KENYATTA UNIVERSITY SCHOOL OF AGRICULTURE AND ENTERPRISE DEVELOPMENT 2ND BIENNIAL INTERNATIONAL CONFERENCE ON ENHANCING SUSTAINABLE AGRICULTURAL PRODUCTION AND MARKETING SYSTEMS Edited by Waceke Wanjohi, Daniel Kyalo Willy, Purity Nguhiu and Cyrus Gichaga HELD FROM 29TH NOVEMBER TO 2ND DECEMBER 2016 AT BUSINESS AND STUDENT SERVICES CENTER, KENYATTA UNIVERSITY KENYATTA UNIVERSITY SCHOOL OF AGRICULTURE AND ENTERPRISE DEVELOPMENT 2ND BIENNIAL INTERNATIONAL CONFERENCE ON ENHANCING SUSTAINABLE AGRICULTURAL PRODUCTION AND MARKETING SYSTEMS Edited by Waceke Wanjohi, Daniel Kyalo Willy, Purity Nguhiu and Cyrus Gichaga HELD BETWEEN 29TH NOVEMBER TO 2ND DECEMBER 2016 AT BUSINESS AND STUDENT SERVICES CENTER, KENYATTA UNIVERSITY I I PROCEEDNGS OF THE 2ND BIENNIAL INTERNATIONAL CONFERENCE ON ENHANCING SUSTAINABALE AGRICULTURAL PRODUCTION AND MARKETING SYSTEMS THROUGH SCIENCE, TECHNOLOGY AND INNOVATIONS Edited by Waceke Wanjohi School of Agriculture and Enterprise development Daniel Kyalo Willy Department of Agribusiness Management and Trade Purity Nguhiu Department of Animal Sciences Cyrus Gichaga School of Agriculture and Enterprise development Design and Layout: Dr. Daniel Kyalo Willy Published and Printed by: Kenyatta University Printing Press, PO BOX 43844, 00100, NAIROBI. ISBN : 1607-4106 © Kenyatta University, School of Agriculture and Enterprise Development. PREFACE The agricultural sector remains a key pillar to economic development in Sub- Saharan Africa and other developing regions. In Kenya for instance, agriculture is one of the key sectors, accounting for 25% of the gross domestic product (GDP). The sector also accounts for 65 per cent of Kenya’s total exports and provides more than 18% and 75% of formal and informal employment respectively. The growth of the national economy is therefore highly correlated with agricultural growth and development. The sector is however faced by a battery of challenges, limiting its potential to play this critical role. The challenges include limited use of modern agricultural technologies extreme weather events, deterioration in soil fertility, post harvest losses and poor access to agricultural extension services. As a result of these challenges, agricultural productivity has been low and agriculture dependent households continue to be impoverished. It is therefore critical that efforts by the government of Kenya in collaboration with other key stakeholders such as research institutes, universities and other collaborators should focus on addressing these challenges. In line with this agenda, the School of Agriculture and Enterprise Development (SAED), Kenyatta University organizes a Biennial Conference every two years. This offers an opportunity to various stakeholders in Agricultural sector to share and exchange knowledge and experiences. The SAED Biennial conference also provides a platform for developing networks between individuals and institutions, through which capacity and expertise can be shared to spur development and address the challenges affecting the Agricultural sector. The second conference was organized in collaboration with World Vision Kenya. It brought various stake holders in agriculture from Kenya and other African countries. The deliberations centered on the theme “Enhancing sustainable agricultural production and marketing systems through science, technology and innovation”. We are therefore pleased to present the book of proceeding of the 2nd SAED Biennial Conference which was held form 29th November to 2nd December, 2016 at Kenyatta University, Nairobi, Kenya. The conference received an overwhelming number of scientific papers cutting across varied themes. These papers are contained in this book of proceedings and therefore will be of interest to wide readership. Acknowledgement This conference was organized jointly between Kenyatta University School of Agriculture and Enterprise development and World Vision-Kenya. The organizing committee wishes to sincerely appreciate the support from the management of Kenyatta University and World Vision-Kenya for their invaluable support during the planning and hosting of the conference. The contribution by members of staff from both Kenyatta University and World Vision-Kenya in various capacities is highly appreciated. We are indebted to all conference national and international participants, who submitted quality scientific papers which culminated into a successful conference. i CONTENTS CONTENTS ....................................................................................................................................... i WELCOMING REMARKS BY THE DEAN, SAED .................................................................. v ABOUT THE CONFERENCE ORGANIZERS ......................................................................... vii Kenyatta University, School of Agriculture and Enterprise Development (SAED) .......... vii World Vision Kenya .................................................................................................................... vii PROFILE OF THE CHIEF GUEST ............................................................................................ viii PROFILES OF KEYNOTE SPEAKERS ....................................................................................... ix PROFILES OF EXHIBITORS ........................................................................................................ xi Biovision Africa Trust (BvAT) .................................................................................................... xi Centre for Agriculture and Biosciences International (CABI) ............................................... xi Monitoring and Evaluation Food Security Programme (MEFOSEP) ................................... xi Agrochemicals Association of Kenya (AAK) / CropLife Kenya ........................................ xii Twaweza Initiative ...................................................................................................................... xii KCB Bank LTD – Ruiru Branch ................................................................................................ xiii Equity Bank ltd- Kenyatta University Branch ........................................................................ xiii OSHO Chemical Industries Ltd ............................................................................................... xiii Kenya Livestock Marketing Council ....................................................................................... xiv The National Sericulture Research Centre, Kenya Agricultural and Livestock Research Organization (KALRO) ..................................................................................................... xiv International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) ................... xv Kenya Plant Health Inspectorate Services ............................................................................... xv CONFERENCE FULL PAPERS ................................................................................................... 16 1. Plenary Session 2: Livestock Production and Health Issues ..................................... 16 1.1 Analysis of consumer preferrence in product attributes: A case study of indigenous chicken’ eggs in Kenya ....................................................................................................... 16 1.2 Livestock Keepers in Isiolo County, Embrace New Adaptive Strategies as Mitigation Measures against the Effects of Drought ......................................................................... 25 1.3 Milk production in Kenya: Challenges and opportunities ............................................. 32 1.4 Infectious Bovine Keratoconjuctivitis (Pinkeye) Infection in a Dairy Cattle Herd in Kenya .................................................................................................................................... 41 1.5 Disorders of the claw and their association with laminitis in smallholder zero-grazed dairy cows ............................................................................................................................ 48 ii 1.6 Participatory technology transfer and impact in poultry production and marketing: The case of farmers in Makueni County .......................................................................... 61 2.Parallel Session 1: Crop Management, Biotechnology and Climate Change .............. 75 2.1 Evaluation of seed quality and phenotypic characteristics of jute mallow (Corchorus olitorius) accessions from north rift region, Kenya.................................75 2.2 Morphological and Biochemical Traits of Sorghum Fostering Abiotic Stress Tolerance Superiority among Cereals: A Review ............................................................................. 92 2.3 Exploration of marine microalgae as potential for biofuel production ....................... 101 3.Parallel Session 2 : Technology diffusion impact and sustainable Agribusiness value chains .................................................................................................................................. 113 3.1 Effect of audience simulation on the knowledge of dieback disease management among cocoa farmers in Ife East Local Government Area of Osun state, Nigeria .. 113 3.2 Analysis of spillover effects of Integrated Pest Management strategy for suppression of Mango fruit flies in Meru County, Kenya ................................................................. 124 3.3 Challenges facing strawberry farming in central Kenya .............................................. 133 3.4 Evaluating health and environmental effects of mango integrated fruit fly management technologies in Meru County, Kenya. .................................................... 142 4. Parallel Session 3 : Crop Management, Biotechnology and Climate Change .......... 154 4.1 Bostrichidae beetles associated with Acacia xanthoploea at Kenyatta University and Mitaboni in Nairobi and Machakos Counties, Kenya ................................................. 154 4.2 Effect of Poly culture cropping on infestation of maize by the vectors of Maize lethal necrosis viruses in Bomet county of Kenya ................................................................... 171 5.Parallel Session 4 : Soil health management, input access and conservation issues 178 5.1 Time preference and resource use behavior by wetland users: The case of ewaso narok wetland, Kenya....................................................................................................... 178 5.2 The effect of invasive plant (Lantana camara) on soil chemistry at conservation area ol-donyo sabuk national park, Kenya ............................................................................ 191 5.3 Determinants of use of integrated soil fertility management strategies and phosphorus management for maize production intensification in Runyenjes sub- County, Kenya. .................................................................................................................. 198 5.4 Combining agroforestry with conservation agriculture for soil fertility enhancement in the dryland context: case of Machakos County, Kenya .......................................... 214 5.5 Effects of selected plant leaf extracts on soil ph and betacarotene levels in amaranthus species grown under varying conditions in Kiambu county...................................... 230 6. Parallel Session 5: Agricultural Innovations, Intensifications and Risk Management.................................................................................................................................. 239 iii 6.1 Farmer knowledge and utilization of rainwater harvesting and conservation technologies in a semi-arid agro-ecozone of kenya ..................................................... 239 6.2 Efficacy of Selected Crude Plant Extracts on Post-Harvest Pathogens in Tomato Fruits in Kirinyaga County, Kenya ............................................................................................ 255 6.3 Evaluation of Belt 480 SC (Flubendiamide 480 g/L) against tomato leaf miner, Tuta absoluta at Mwea in Kirinyaga County ........................................................................... 269 6.4 Developing an effective pest management strategy for French beans in Kenya to protect export market ....................................................................................................... 278 6.5 Evaluation of Tihan OD 175 (Flubendiamide 100 g/L + Spirotetramat 75 g/L) against sucking and chewing arthropod pests of cotton at Mwea in Kirinyaga County ..... 282 7.Parallel Session 6: Crop Management, Biotechnology and Climate Change ............ 294 7.1 Physicochemical characterization of selected rice (Oryza sativa L.) Genotypes based on gel consistency and alkali digestion .......................................................................... 294 7.2 Impact of nitrogen fertilizer application on groundwater nitrate and ammonium concentration in different farm systems in Ainabkoi, Uasin Gishu County. Kenya. 303 7.3 Supplementing households income through sericulture technology. ........................ 316 7.4 Seed certification as a means of curbing emerging diseases: A study of Maize Lethal Necrosis disease in Kenya ................................................................................................ 326 7.5 Mango pollinators and their spatial differences across a forest-farm gradient in Kilimambogo, Machakos County, Kenya ...................................................................... 333 7.6 Low cost insect proof and micro-climate regulating Agronets are effective in managing infestation of American Bollworm (Helicoverpa armigera Hubner.) on field tomatoes in Nairobi and Kiambu counties, Kenya ...................................................... 340 7.7 Appropriate pollinator management scheme for kenyan farmlands .......................... 346 7.8 Avocado (Persea americana) yield as influenced by pollinators in Kandara, Murang’a County, Kenya ................................................................................................................... 351 8.Parallel Session 7: Technology diffusion impact and sustainable Agribusiness value chains .................................................................................................................................. 356 8.1 Effect of phosphorous on yield and growth of soybean (Glycine max L) varieties in central highlands of Kenya .............................................................................................. 356 8.2 A comparative analysis of the determinants of gross margins from African Indigenous Vegetables in semi-arid and humid rural areas of Kenya ...................... 367 8.3 Determinants of Intensity of Participation in Avocado Producer and Marketing Groups in Murang’a County, Kenya .............................................................................. 377 9. Parallel Session 8: Crop Management, Biotechnology and Climate Change ........... 385 iv 9.1 Evaluation of the effect of (A. indica, A. sativum and L. camara ) leaf extracts on root- knot nematode management on tomatoes in Mwea-East, Kenya. ............................. 385 9.2 Assessment of sweet melon pollinators grown under field conditions at KALRO- Kabete, Kenya. ................................................................................................................... 394 9.3 Assessment of genetic diversity of Kenyan Dolichos bean (Lablab purpureus) (L) using simple sequence repeat (SSR) markers .......................................................................... 399 9.4 Response of yield and yield components of maize as affected by tied ridges, fertilizers and cropping system in the semi- arid areas of Machakos County, Kenya 409 CONFERENCE ORGANIZING COMMITTEE .................................................................... 416 OVERALL COORDINATOR ................................................................................................... 416 MAIN COMMITTEE MEMBERS ............................................................................................ 416 SUB-COMMITTEE MEMBERS................................................................................................ 416 LIST OF PARTICIPANTS......................................................................................................... 418 v WELCOMING REMARKS BY THE DEAN, SAED It gives me great pleasure to warmly welcome you to the 2nd SAED Biennial Conference taking place from today the 29th November 2016 to 2nd December 2016 at the Business and Students’ Services Center, Kenyatta University, Nairobi Kenya. In 2012, the School of Agriculture Board of the then 4 year old School made a bold decision to be holding biennial conferences with the primary aim of creating a platform to disseminate research findings; bringing various actors in the industry together for dialogue and enhancing the visibility of the young School. In 2014, we held a successful 4 days inaugural conference and I am happy that the Conference is now fully integrated in the school calendar. The overall theme of the conference is “Enhancing Sustainable Agricultural Production and Marketing Systems”. Prof. Waceke Wanjohi, PhD. FRSB, FKNAS Dean, School of Agriculture and Enterprise Development. It is notable that though agriculture contributes 25% of the GDP, employs over 70% of the workforce and is key in the attainment of the middle income economy, its characterized by low productivity (both in terms of low return to land and labour). Raising agricultural productivity in a sustainable way remains therefore a major national focus as captured in the Kenya Development blue print; the Kenya Vision 2030. The high food losses (54%) at the production, postharvest handling and storage stages and the high food wastages (46%) at the processing, distribution and consumption stages further aggravates the matter. According to a recent FAO report, a staggering 1.3billion tonnes of food per year is wasted leading to economic losses in the tune of USD 750 billion (75trillion) annually and adding 3.3billion tonnes of greenhouse gases hence negatively affecting the environment. Compounding this challenge is the aging farming population in that the average farmer in Kenya today is 60 years old against a life expectancy of 63 years while the youth comprising 33% of the Kenyan population have in general a negative attitude towards agriculture. Given the slow economic development pace in Kenya, it is hard to expect the industry and service sectors to grow substantially to absorb all the youths entering the job market. Therefore, agriculture, the alternative sector, has to play a critical role of providing employment opportunities to the youths. Increasing commercial viability of agricultural enterprises so as to make agricultural returns attractive is also critical. Successful agribusiness investments stimulate agricultural growth through the development of new markets and a vibrant input supply sector. For this reason the School of Agriculture has focused on rebranding the curriculum to make it attractive to the youth, with an emphasis on agriculture as a business and promotes and supports the engagement of the youth in agriculture through various innovative initiative. Though technologies to enhanced sustainable agricultural production systems have been developed, the diffusion of technology in Kenya has been slow. vi Further, due to infrastructural challenges, incidences of high (market) transaction costs are rampant and often leads to market failure. In addition, most small holder farmers also are not fully integrated into high value markets so as to take advantage of globalization of agricultural value chains. This conference therefore brings together the practitioners in agriculture, policy makers, research institutions, civil society and the private sector to deliberate on the challenges revolving around these key challenges. The dialogue will revolve around five subthemes namely: Subtheme 1: Crop management, Biotechnology and Climate change Management of Emerging Crop Pests and Diseases, Emerging issues in Agricultural Biotechnology, building farmer resilience and risk management strategies in the wake of climate change. Subtheme 2: Livestock production and health issues Strengthening Integrated Crop and Livestock production systems, Enhancing Livestock productivity through Improved Livestock health and new production techniques. Subtheme 3: Technology diffusion, Impact and Sustainable Agribusiness value chains: Facilitating access to technology, fast adoption and diffusion of Innovations in Agricultural Value chains, building environmental sensitive agribusiness value chains, Agricultural commercialization and Impact of agricultural innovations on farmer outcomes. Subtheme 4: Soil health management, Input access and Conservation issues: Integrated strategies for enhancing Soil health, Soil and Water Conservation and enhancing performance of irrigated agriculture, yield response to fertilizer use and Innovative strategies for facilitating input access. Subtheme 5: Emerging and Topical issues in Agriculture Agricultural Innovations, ICT Applications in Agriculture & Agribusiness, Youth involvement in Agri-preneurship, Mobile phone applications in Agriculture, Gender issues in Agriculture, Access to resources and women Participation in commercial agriculture. The aims of this conference are fivefold; share research findings which seek to address some of the major challenges affecting the agricultural Sector with a broad audience; bring various actors in the industry together for dialogue and for networking; provide a platform for mentoring upcoming research scientists; create an opportunity for the players in the sector to exhibit their services and products and strengthen engagement and linkage between the actors in the industry and the School of Agriculture. Partnering with us in organizing this conference is the World Vision Kenya; an Organization which is very well known by the audience having been in existence since 1950 and with significant impact. The 2016 SAED Biennial conference has attracted participants and exhibitors from different organizations and comprises of 4 plenary sessions, six parallel sessions and exhibitors’ session. It is therefore my hope that you will make the most of this conference as you interact and learn from this diverse audience. On my own behalf and on behalf to the entire School, I once again welcome you to the 2nd SAED Biennial Conference and wish you fruitful deliberations and engagement. vii ABOUT THE CONFERENCE ORGANIZERS Kenyatta University, School of Agriculture and Enterprise Development (SAED) Kenyatta University, School of Agriculture and Enterprise Development was established in December, 2007 in response to the growing need of competent leaders who can meet the rapidly changing market demand in the Agricultural related sectors and to contribute in the achievement of Kenya’s Blue Print Vision 2030. Agriculture is one of the six components under the economic pillar to deliver 10% economic growth per annum. The School provides an enabling environment for academic excellence, research and community engagement with a focus to producing socially responsible, highly motivated and responsive professionals with a significant outlook and perspective to potentials and challenges in Agriculture. The school has a strong emphasis on agribusiness. The objectives of the school are to: (i) Re-orient the training of agricultural scientists to encompass coordinated pluralism that integrates a wide range of specializations including enterprise development in Agriculture. (ii) Promote agricultural theory and practice as a business for sustainable development. (iii) Promote the primacy of agriculture in sustainable socio- economic development. (iv) To promote partnership, networking and collaboration with the Public and Private sectors at the all levels. (v) To create a conducive environment for teaching, learning and carrying out research in agriculture. (vii) To provide consultancy services in major areas of agriculture. World Vision Kenya Founded in 1950, World Vision is an international partnership of Christians serving children, their families and communities in conditions of poverty in nearly 100 countries, primarily through programmes of community development, disaster management and promotion of justice. World Vision began operations in Kenya in 1974. Its more than 800 committed development staff members work in 52 Area Development Programmes in 35 counties country-wide. Through valued partnerships, World Vision Kenya supports communities to access the knowledge and resources needed to improve the well-being of children and overcome poverty. World Vision has four strategic goals regarding the well-being of children: (i) Children enjoy good health. (ii) Children are educated for life. (iii) Children experience the love of God and their neighbours, always being mindful that World Vision is a guest in cultures that hold many diverse beliefs. (iv) Children are cared for, protected and participating in the decisions that affect their lives. World Vision Kenya works with the most vulnerable children and communities regardless of religion, race, ethnicity or gender. Vision Statement: Our vision for every child, life in all its fullness; Our prayer for every heart, the will to make it so. Mission: World Vision is an international partnership of Christians, whose mission is to follow our Lord and Saviour Jesus Christ, in working with the poor and oppressed to promote human transformation, seek justice and bear witness to the good news of the Kingdom of God. viii Our Core Values: We are Christian | We are committed to the poor | We value people | We are stewards | We are partners | We are responsive PROFILE OF THE CHIEF GUEST Prof. Paul Wainaina is a Professor of Philosophy at Kenyatta University. He holds Bachelor of Education (Honors) from the University of Nairobi. He completed his graduate training from the same University where he obtained his MA majoring in Philosophy of Education. He later completed his PhD in Philosophy of Education from the University of Alberta, Canada where his research focus was on axiological and epistemological foundations of education. He is a keen philosopher with extensive research work on Ethics in Education. Prof. Wainaina has risen through the academic ranks; from Lecturer to Full Professor. He has been elected Dean, Faculty of Education, University of Namibia, in the Republic of Namibia. Prof. Paul K. Wainaina AG. Vice Chancellor, Kenyatta University Prior to his appointment as acting Vice Chancellor, Kenyatta University, Prof. Wainaina has served in various positions within Kenyatta University including being a member of the Kenyatta University Management Board and University Council. He has also served as a member of the Kenyatta University Staff Pension Scheme as well as Acting Deputy Vice Chancellor (Administration), Acting Deputy Vice-Chancellor (Academic Affairs), and Deputy Vice Chancellor (Administration), Kenyatta University. ix PROFILES OF KEYNOTE SPEAKERS Prof. Wainaina is a PhD holder in Agricultural Economics, Master in Business Administration, Bachelor of Education and Associate Professor in the School of Business University of Nairobi. He has extensive working experience with University of Nairobi where he is an Associate Professor, Kenya Vision 2030 Delivery Secretariat where he was Acting Director General as well as Director for Social and Political Pillars, World Bank where he was a Senior Education Economist, CARE International in Kenya where he was Regional Coordinator and University of Nairobi Enterprises and Services Limited, where he was Deputy Managing Director and Business Development Manager. I have been a board member of Capital Market Authority, Lamu Port South Sudan and Ethiopia Transport Corridor. Presently, I am a council and board member of Management University of Africa, council member of Riara University, alternate board member for Africa Group B Partnership in Statistics for Development in the 21st Century under Organization for Economic Co-operation and Development, and a member of Uwezo Kenya National Advisory Committee. Prof. Gituro Wainaina, Associate Professor, School of Business - University of Nairobi & Former Ag. Director General, Vision 2030 Secretariat Mr. Jeremiah Nyaga Operations Director, World Vision Kenya Jeremiah Nyagah, is the Operations Director of World Vision Kenya, "a Christian relief and development organization dedicated to promoting the well-being of children, families and communities around the world. Jeremiah is a project management specialist with holistic management experience gathered over 17 years in Programme management, Design, Monitoring & Evaluation (DM&E), Change management and leadership. He has had a successful career in humanitarian programming, having worked with Pathfinder International, Child Fund (CCF then) and now with World Vision. He joined World Vision Kenya as an Area Development Programme Manager and has served in other roles over the years; which include Regional DM & E Coordinator, National Transformational Development Indicators Measurement Coordinator, National DM & E Coordinator and Associate Program’s Director. Jeremiah assumed the role of the Operations Director in October 1, 2015. Jeremiah is a Bachelor of Education graduate from Egerton University. He obtained a postgraduate degree in Development studies from the University of South Africa (UNISA). In addition he has a Master’s degree in Organizational Leadership from Africa International University. x Mr. Tennyson Williams, a skilled leader and manager with experience spanning more than 15 years in INGO management and leadership. As, an international development specialist with a background in zoology and zoonotic diseases, Tennyson has held a number of senior management positions directing Africa-wide regional programmes with multi-donor funding. Originally from Sierra Leone, he was until joining World Society for the Protection of Animals (WSPA) the ActionAid Kenya Country Director responsible for directing programmes, policy and advocacy, and shaping the development of the organization in Africa as well as internationally. Mr. Tennyson Williams, Regional Director for Africa, World Animal Protection Prior to that, he was ActionAid International Director for West and Central Africa with strategic leadership, governance and programme oversight responsibilities in Burundi, Democratic Republic of Congo, Ghana, Liberia, Nigeria, Rwanda, Senegal, Sierra Leone and The Gambia. Before that he was ActionAid Country Director for Sierra Leone focusing on poverty reduction and humanitarian influence and credibility, and driving forward WSPA’s campaigns and programmes, particularly in the areas of Disaster Management, Humane and Sustainable Agriculture, Ending Inhumane Culling and Wild Life. Tennyson joined WSPA in August 2013, with the overall mandate of establishing WSPA as a leading animal protection organization in Africa, enhancing its reach, Evelyn Namubiru-Mwaura has over fifteen years of experience in international development, strategy and governance through appointments with AGRA, the World Bank, UNREDD, UNDP-GEF, International Forestry Resources and Institutions (IFRI) and Makerere University. She provides policy and strategic leadership in the development and delivery of AESA’s scientific agenda for Africa. The holder of a PhD in public policy from Indiana University in the US has won several awards including the Elinor Ostrom-Johan Skytte Fellowship, the International Foundation for Science award for field research, and the Compton Foundation Peace & Security Fellowship and the Australian Sponsored Training Scholarship. Dr. Everline Namubiru Mwaura, Strategy and Policy Manager xi PROFILES OF EXHIBITORS The conference attracted eighteen (18) organizations who participated in the exhibitions category. These were drawn from a wide range of organiozations enganged in different activities cutting across the agricultural value chain: Agro-input producers and marketers, Regulators, Financial Institutions, Farmer and Community development groups, Non- Governmental Organizations and Research Organizations. Below are profiles of some of the exhibitors; Biovision Africa Trust (BvAT) Biovision Africa Trust (BvAT) is a not-for-profit organization established in Kenya in 2009 by the Biovision Foundation for ecological development in Switzerland and supported by the International Centre of Insect Physiology and Ecology (ICIPE) in Nairobi. The Trust’s goal is to alleviate poverty and improve the livelihoods of smallholder farmers by supporting dissemination of information and knowledge on appropriate technology to improve human, animal, plant, and environmental health. The overall goal of the Trust is to sustainably improve the lives of the people in Africa while conserving the environment as the basis for all life Our vision is a food secure African continent with healthy people living in a healthy environment. The trust runs several projects to help it achieve its objectives:  The organic farmer magazine  The organic farmer radio  Outreach project For more information please visit: www.Infonet-biovision.org Centre for Agriculture and Biosciences International (CABI) CABI is an international not-for-profit organization that improves people’s lives worldwide by providing information and applying scientific expertise to solve problems in agriculture and the environment. Our approach involves putting information, skills and tools into people's hands. CABI's 48 member countries guide and influence our work which is delivered by scientific staff based in our global network of centres. CABI is also a leading global publisher producing key scientific publications, including the world renowned CAB Abstracts (the leading abstracting and indexing database in the applied life sciences), Global Health, as well as Compendia, books, eBooks and full text electronic resources. Behind each of our products is a team of subject specialists committed to delivering the most relevant and authoritative information to researchers worldwide. We also offer CAB Direct, CABI's online development platform, providing a single point of access to all your CABI database subscriptions. CABI has over 100 years’ experience in scientific publishing. Monitoring and Evaluation Food Security Programme (MEFOSEP) MEFOSEP established a food security monitoring information system, by using anotice board which provides a target plan and updated information based on the actual findings that are taking place among the farmers in the community. MEFOSEP have upgrade this xii system to use of a mobile app to know the right spacing required and to calculate for farmers on the quantity of farm inputs (fertilizers, manure and seeds) they need and the expected yield once they insert the acreage of land and the recommended fertilizer after conducting soil test analysis. It also monitors famer’s production activities. These information from the app is relayed into a central server system which aggregates it and update it on the notice board. Objectives: 1 To improve on preparedness to mitigate agricultural challenges by undertaking timely (early) warning through sequential disseminating information on the actual outcome of each season on the notice board and providing intervention in the published articles and in the website blog for discussion. 2 To improve food security by identifying food security gaps through data analysis and participatory dialogue meeting with the farmers, agricultural staffs to compare each seasons with other previous seasons and the target plan. Agrochemicals Association of Kenya (AAK) / CropLife Kenya The Agrochemicals Association of Kenya (AAK) is the national representative of the International Agrochem Industry represented worldwide by CropLife International. AAK is the umbrella organisation in Kenya for manufacturers, formulators, repackers, importers, distributors, farmers and other users of pest control products. The association has been in existence since 1958 under various names with the most recent one being Pesticide Chemicals Association of Kenya (PCAK). The objectives of the association are: Promotion and public education on the safe and responsible use of pesticides, providing an agency for liaison with government and other stakeholders on matters of mutual interest, promote just, fair and honourable trade practices among its members while weeding out counterfeits, encourage research in areas that will improve pesticide use as well as work with regulators on labelling, registration and taxation of pesticides. Twaweza Initiative Twaweza initiative is a young women group founded in the year 2010. The group is found in Korogocho- Kariobangi slum in Nairobi, Kenya. The group was formed to address the challenges facing young women and the girl child in Korogocho slum. The group has 20 members in total. (15 Female and 5 Male) The challenges that the group seek to address are; poverty, prostitution, early pregnancies, early marriages, abortions, school dropout and crime among other problems. These problems are majorly due to low level of income. The group seeks to address these challenges through economic empowerment to members and young women in the society. This is done through various economic activities that the group is involved in. The activities that the group engages in so as to empower its members are: production of sandal, bags, bracelets, mats bead work and sanitary towels. The group is also involved in environmental conservation and community work like reproductive health programs in the local school. xiii We use ecological waste materials which contribute to pollution, in our production. Most of our raw materials include: (i) Old fabrics (Renin) (ii) Organic cotton (iii) Old tires Contacts: TWAWEZA INITIATIVE, P.O BOX 7714-00300 NAIROBI. Email; twawezainitiative@gmail.com Tel: 0714356845/0706382359/0715188634/ 0739561662. www.twawezainitiative.com | Facebook: Twaweza Initiative | Twitter: Twaweza Initiative KCB Bank LTD – Ruiru Branch KCB Group is registered as a non-operating holding company which started operations as a licensed banking institution with effect from January 1, 2016. The holding company oversees KCB Kenya – incorporated with effect from January 1, 2016 – and all KCB’s regional units in Uganda, Tanzania, Rwanda, Burundi, Ethiopia and South Sudan. It also owns KCB Insurance Agency, KCB Capital, KCB Foundation and all associate companies. KCB offers a wide range of products in agriculture including: KCB Dairy loans ; KCB Agribusiness Working Capital Loans; KCB Agribusiness Asset Finance Loans; KCB Horticulture product; KCB Kenya Seed product; KCB Mavuno product KCB Bio Digestor loan product. Equity Bank ltd- Kenyatta University Branch Equity Group Holdings Limited, formerly Equity Bank Limited, is a Kenya-based commercial bank. The Company provides financial services to individuals and small and medium sized enterprises. The Company operates in six geographical markets: Kenya, Uganda, South Sudan, Rwanda, Tanzania and Democratic Republic of Congo. Its three customer facing lines of businesses include consumer, small and medium enterprises (SME's) and corporate. The consumer business line focuses on salaried customers or customers receiving other regular remittances, such as pension. The facilities granted under its SME's line of business are for purposes of meeting working capital needs, property development or acquisition of assets. The corporate line of business comprises large enterprises. The Company offers a range of products, including Equity loan, Vijana loan, Fanikisha loan, Farm input, Mortgage loan, Asset finance loan, Trade finance, Development loan, Business Loan and Biashara Imara. Products in support of agriculture include: Kilimo biashara loan (small scale), Kilimi Biashara (Agribusiness), Kilimo Maendeleo Loan and Kilimo kisasa loan. OSHO Chemical Industries Ltd Osho Chemical Limited is a leading marketer, manufacturer and distributor of agricultural inputs, industrial chemicals and veterinary products. Its product range include: Agricultural Inputs – Agricultural chemicals, Bio-pesticides, Fertilizers, Knapsack sprayers, Greenhouse sheeting and Growth media. Industrial Products: Industrial Chemicals, Industrial & Institutional Detergents, Oxides and Dyes, Screen-printing auxiliaries, Other Industrial inputs and equipment. Veterinary Products: Dewormers, Acaricides, Mineral Supplements and Metabolic Stimulant, Trypanocidals, Antimirobials, Disinfectant and Antiseptic, Poultry Products, Cattle Sprayers, Toxorid, Bloatosil, Vaccine xiv and Identification Equipments. In addition to this Osho undertakes third party Toll manufacturing. Kenya Livestock Marketing Council The Kenya Livestock Marketing Council (KLMC) is an umbrella organization of livestock producers and traders in arid and semi-arid areas of Kenya. It was established in 2000 as a non-profit making services organization to counter the challenges the livestock farmers are facing. KLMC is an non-political, non-religious/commercial livestock pastoralists organization and is an affiliate of KENAFF,AIN, KEPSA, Kenya Chamber of Commerce, joint Arab Chamber of Commerce and NEALCO. KLMC believes that pastoralism is the key economic activity in the Arid and Semi-Arid areas of Kenya and it can be harnessed to promote economic growth and development in these regions and reduce vulnerabilities. KLMC recognizes the fundamental challenge facing the pastoralists is their inability to access better market for their livestock and livestock products. In order to overcome this challenge, KLMC is working closely with other development partners and stakeholders to source for better market for livestock and livestock products and disseminate market information to producers and traders on timely basis within the country , regionally and internationally. The National Sericulture Research Centre, Kenya Agricultural and Livestock Research Organization (KALRO) The Kenya Agricultural and Livestock Research Organization (KALRO) is a premier national research organization established under the Kenya Agricultural and Livestock Research (KALR) Act No. 17 of 2013 as a successor of four former agricultural research institutions; Kenya Agricultural Research Institute (KARI), Tea Research Foundation (TRF), Coffee Research Foundation (CRF) and Kenya Sugar Research Foundation (KESREF). It is mandated to conduct agricultural research of strategic and national importance and produce public goods in the form of technologies, information and new knowledge. KALRO comprises of 16 interactive and interdependent network of semi- autonomous institutes (comprising of over 50 centres and sub centres) situated in different parts of the country each with specific mandates that contribute to the overall vision and mission of KALRO. The National Sericulture Research Centre is one of the KARLO centers with a national mandate for promoting and catalysing growth of sericulture industry in Kenya. The center focuses on Developing business incubation and technology centre where entrepreneurs will receive apprenticeship to gain practical experience in the industry, Establishing strong business models for sericulture industry in Kenya; Building strong research outputs to support the industry, in collaboration with key partners and Establishing a national grainage for the silkworm seeds to cushion and protect domestic sericulture. The center is involved in the production and marketing of the following products: Raw silk, Mulberry xv cutting and saplings, Spunned floss, Silk and silk blended fabrics,Mulberry tea,Silkworm pupa for animal and poultry feed, dry cocoons and shells For more information please visit: www.karlo.org International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) ICRISAT is an international non-profit organization that undertakes scientific research for development. It works in agricultural research for development across the drylands of Africa and Asia making farming profitable for smallholder farmers while reducing malnutrition and environmental degradation. ICRISAT works across the entire value chain - from developing new varieties to agri-business and linking farmers to markets. Because of its focus on the drylands, the organization has an extra specialization on crops that survive in these harsh climates. These are: Sorghum, Finger Millet, Pearl Millet, Chickpea, Pigeonpea and Groundnuts. ICRISAT has offices in Kenya, Malawi, Mali, Mozambique, Niger, Nigeria, Ethiopia, Zimbabwe and India. It is a member of the CGIAR System Organization. For more information please visit: www.icrisat.org Kenya Plant Health Inspectorate Services Kenya Plant Health Inspectorate Service (KEPHIS) is the government parastatal whose responsibility is to assure the quality of agricultural inputs and produce to prevent adverse impact on the economy, the environment and human health. Vision: The Lead Regulator and Facilitator of Globally Competitive Agriculture Mission: To Provide a Science Based Regulatory Service by Assuring the Quality of Agricultural Inputs and Produce to Promote Food Security and Sustainable development Strategic Objectives  To protect plants from pests, weeds and invasive species.  To facilitate review and strengthening of the policy, legal and regulatory framework, so that it is in tandem with both local and international agricultural sector emerging issues.  To contribute towards improved levels of agricultural productivity.  To support compliance to market requirements.  To build adequate technical and infrastructural capacity to facilitate efficient and effective delivery of the KEPHIS mandate.  To mobilize adequate financial resources and ensure optimal allocation and utilization to enable full implementation of planned programmers and activities.  To enhance synergies through information and resource sharing with stakeholders and partners. For more information visit: www.kephis.org 16 CONFERENCE FULL PAPERS 1. Plenary Session 2: Livestock Production and Health Issues 1.1 Analysis of consumer preferrence in product attributes: A case study of indigenous chicken’ eggs in Kenya aCharles Ndenga*, aEric K. Bett, bLucy W. Kabuage aDepartment of Agribusiness Management and Trade, Kenyatta University, P.O Box 43844-0100, Nairobi, Kenya. bDepartment of Agricultural Resource Management, Kenyatta University, P.O Box 43844 0100, Nairobi, Kenya, email: lwkabuage@gmail.com *correspondence :charlesndenga@yahoo.com, Tel; 0725093929, Abstract Indigenous chicken eggs are highly preferred in Kenya by majority of both rural and urban consumers. In the recent past, research has focused on improving productivity levels resulting to Improved indigenous chicken eggs with less focus on consumer preferences. Despite their efforts, Indigenous chicken eggs dominate the market and the country is egg deficit. This study sought to identify and analyze preference attributes of indigenous chicken’ eggs that influences choice and consumption. The specific objectives were to; identify and rank indigenous chickens’ eggs attributes that influences choice and consumption and analyze the influence of these attributes on price of eggs. Multistage sampling technique was employed to select 180 respondents in Nairobi and Makueni counties. Kendall coefficient of concordance test was used to identify and rank the attributes of indigenous chicken’ eggs while semi-logarithmic functional form of hedonic pricing model was employed to analyze the influence of eggs attributes on price. Results indicated that consumers buying indigenous chicken eggs prefer and are willing to pay premium prices for brown shelled, non oval eggs. Small and medium sized eggs were also more preferred and received price premiums compared to the larger sized eggs. Small and medium sized eggs were perceived to be organically produced with less chemical components. The study recommended for incorporation of this attributes by breeders. Farmers can enhance the preferred egg attributes by adopting better production systems while retailers can use the attributes in formulating appropriate marketing strategies in indigenous chicken eggs business. Keywords: Attributes, choice, Hedonic pricing. INTRODUCTION Indigenous chickens(IC) contributes 71% of the total egg and poultry meat produced in Kenya and therefore impacts significantly on the rural trade, welfare and food security of smallholder farmers (Philip, 2007). Over 21 million people in rural areas depends on IC for livelihood support. Consumers exhibit high preferences for indigenous chicken 17 eggs and are willing to pay 41.53% more compared to other eggs (Bett et al., 2011). This preference is due to increasing knowledge on health complications associated with food consumption (Changhee, 2011). Today’s consumers want an ever-widening variety of food products with various characteristics of nutrition convenience, food safety, environment and other traits. In general, changes in food consumption patterns are primarily as a result from increase in income and changes in food prices. As income increases, so is the demand for quality and safety particularly for food products. The new awareness is attributed to dietary changes associated with more disposable income and urban growth. For instance, in 2005, Kenya produced 60,000 tonnes of eggs in shells and imported 240 tonnes to fill the shortage in supply (FAO, 2008). The deficit is an indication of the local market failure to stimulate production. Production of indigenous chicken is characterized by low levels of inputs and outputs (Okitoi, 2007)with low productivity levels, which limits their potential for commercialization. This has resulted in development of improved indigenous chicken that have high productivity levels compared to IC. Emphasize on IIC eggs through high productivity levels has left out consumer preferences an important component in acceptability and marketing of products. Nonetheless, improved indigenous chickens’ eggs have failed to fill the widening gap in demand for eggs and IC eggs dominate market share. There is therefore a knowledge gap on what consumers prefer in the IC eggs that is probably not in the recently released IIC eggs. Lack of consumer preference analysis could be a factor that limits utilization, subsequently low production of newly released IIC eggs. The problem therefore is insufficient information on the factors that determine choice of eggs by consumers in the market. The overall objective of this study is to fill this gap in knowledge. The paper will achieve this through its twofold objectives; to identify and rank attributes of indigenous chicken eggs that influence choice and consumption and to analyze the influence of these attributes on price of eggs. The results will inform and guide producers of improved indigenous chickens’ eggs on the management practices and production systems to adopt in order to enhance attributes that fulfils market requirements hence increasing chicken’ egg prices and demand. Consequently, traders will be able to adopt strategies in transportation, handling, storage and transformation in order to improve retail level chicken egg prices through emphasis on retail level attributes that are important to end users. Several studies have applied hedonic pricing to disentangle preference attributes from bundled goods and their economic valuation. Egg types have different attributes which determine their attractiveness to consumers. These attributes are heterogeneous, making each type distinct. Hannah (2015), employed hedonic price model to analyze the effects of common beans attributes on price in Kiambu Kenya. Results indicated that consumers were willing to pay premium prices for taste, price and cooking time. Dalton (2003), used hedonic price model to evaluate consumption attributes perceived important by rice consumers in West Africa. Results showed that grain elongation and swelling were important in relation to the amount of rice prepared and the amount that can effectively feed a household. Timothy et al., (2006) sought to analyze cattle prices in central corridor of West Africa by employing hedonic price model. Specifically, the 18 study was to determine if market participants have systemic preferences for cattle attributes and their willingness to pay premium price for the attributes. The results of the study indicated that animal age, sex, breed, body condition, purpose of purchase, season of sale and market location were the most significant factors influencing short run cattle prices in the study region. Ramatu et al. (2014) sought to determine the quality characteristics of dressed local and imported chicken that influence consumer preference and how these affect the prices of chicken. The study employed hedonic pricing model which indicated that consumers were willing to pay premiums for imported, non fatty and tender attributes of chicken. Changhee (2010) analyzed retail prices of eggs in Korea using hedonic price model in order to identify those attributes that affect prices and the respective value of each attribute. Results indicated that the status of eggs fertility, organic feeding and free range feeding are the main attributes that positively affect the retail price of eggs. MATERIALS AND METHODS Theoretical framework Egg types have different attributes which determine their attractiveness to consumers while making purchase decision (Becker, 1965 and Lancaster, 1966). According to these theories, consumers choose indigenous chicken eggs that maximize their utility based on consumption characteristics. The underlying assumption postulates that products consist of utility-bearing attributes and that the values of those attributes collectively contribute to the price of the product (Rosen, 1974). This approach is called the hedonic pricing method in which the price of indigenous chicken egg is viewed as a composite of implicit values/prices of each individual attribute. The price of a good is a function of the amount of attributes that it contains and of the values placed on them (Carman, 1997). This can be represented as:- 𝑃 = 𝛽0 + ∑ (𝛽𝑗𝑍𝑗 + ε) … … … … … … … … … … … … … … (i) 𝑚 𝑗=1 Where the vector Z stands for a particular variable of indigenous chicken egg, 𝛽0is the intercept; 𝛽𝑗 is the regression coefficient and ε is the random error term satisfying the classical regression assumption. The study area and sampling This study was conducted in Makueni and Nairobi counties of Kenya. Makueni county lies between latitude 1035’ South and longitude 370 10’ East (Makueni county integrated plan (CIDP), 2013). Nairobi county on the other hand lies between latitude 10 17’ south and longitude 36 0 49’ East and has nine sub counties. The target population for this study consisted of all consumers of indigenous chicken eggs in Nairobi and Makueni counties. Multistage sampling technique was used and 180 consumers selected in the four sub counties as follows; Kaiti 33, Makueni 57, Starehe 43 and Westlands 47 consumers. 19 Data Analysis Hedonic pricing model was used to analyze the influence of attributes on price of eggs. The price is the dependent variable upon which egg attributes are regressed. 𝑙𝑛𝑝𝑖 = 𝛼0 + ∑ 𝛽𝑠ℎ𝑒𝑙𝑙 𝑐𝑜𝑙𝑜𝑟 + 𝑘 𝑛=1 ∑ 𝛽𝑠𝑖𝑧𝑒 + 𝑘 𝑛=1 ∑ 𝛽𝑐𝑙𝑒𝑎𝑛 + 𝑘 𝑛=1 ∑ 𝛽𝑠ℎ𝑎𝑝𝑒 + 𝑘 𝑛=1 ∑ 𝛽𝑝𝑎𝑐𝑘𝑎𝑔𝑒 + 𝑘𝑛=1 ∑ 𝛽𝑓𝑟𝑒𝑠ℎ + 𝑘 𝑛=1 ∑ 𝛽𝑟𝑒𝑔𝑖𝑜𝑛 + 𝑘 𝑛=1 ∑ 𝑆𝐷 + 𝜀 𝑘 𝑛=1 … … … … … … … … … … (ii) Where P is the price of IC egg in Kenya Shillings, SD represents socio-demographic factors influencing market price βs are parameter estimates and ɛ is stochastic error term. Overall variables described above are dummy variables. In a semi logarithmic functional form (adopted for this study), the effect of a dummy variable on the dependant variable is not equal to the first derivative of the regression function with respect of the dummy variable in question, unlike the effect of a continuous variable (Kennedy, 1981). This implies that the first derivative ( 𝛽) is a potentially imprecise approximation of the effect of the dummy variable on the dependent variable. Among the approaches available to correct this is the method suggested by (Kennedy 1981). Following this method, the effect of a change of xk is from zero to one on P, can be calculated as follows (Kennedy, 1981). 𝑔𝑘 = 𝑒𝑥𝑝 (𝛽𝑘 − 1 2 𝑉 (𝛽𝐾)) − 1 … … … … … … … … … … … … … … . (𝑖𝑖𝑖) Where v(𝛽𝑘 ) is the estimated variance of the estimated coefficient 𝛽𝑘 RESULTS AND DISCUSSION Socioeconomic characteristics of the sampled households Out of the sampled respondents in both counties,56% comprised of male while the rest were female as shown in table 2 implying that in Urban centres there is a large percentage of male as compared to women who in most cases are found in rural areas engaged in farming activities. This can be attributed to rural – urban migration in search of employment by majority of men(Table II). The mean age for the respondents was 36years (table II). The youngest respondent was aged 17 years and was a male while the oldest was aged 85 years and was a female. The literacy rate of 96.5% seemed to be higher in the study area compared to the national average literacy rate of 87.01% (World Bank, 2009). Majority of the respondents; 48% had secondary level of education, 18% had primary level of education, 14% had Diplomas and 12% had Degrees. The average family size for the current study was 4 members with a minimum of 1 and a maximum of 11 family members. Among the sampled respondents,72% were married while 27% were single. 35% of the respondents received a monthly income of upto Ksh. 10,000 (table II). Only 21% of the respondents earned a monthly income of above Ksh. 30,000. Attributes of I.C Eggs I.C eggs that were identified during the preliminary survey were presented to respondents for verification and ranking. The attributes were; shell color, size, shape, cleanliness, price, freshness and package. Results of Kendall’s ranking indicated that size, price, shell color and freshness of the I.C eggs were the most important attributes influencing consumer choice and consumption. Cleanliness and Package of the IC eggs 20 were the least ranked attributes influencing choice and consumption (Table III). The F- test results indicated that 36% of the respondents were in agreement on ranking of the IC eggs attributes at 99% confidence level. The null hypothesis was thus rejected and alternative which stated that there is significance agreement among respondents in ranking the IC eggs attributes adopted. Influence of Indigenous Chicken Eggs’ Attributes on price For the indigenous chicken eggs, a total observation of 180 respondents was made. The estimation of semi-logarithmic form of equation was completed using the ordinary least squares method (OLS). The model included seventeen independent variables which were all dummies with the exception of age and household size of the respondent. Heteroskedasticity and multicollinearity is a problem associated with cross-sectional data, hence they were tested using Breusch-pagan test and variance inflation factor (VIF) respectively. A mean VIF was 1.79 and no variable individually had a VIF of above 5 (rule of thumb) for presence of multicollinearity. The hypothesis for homoskedasticity in error term was thus accepted (P>0.10).The significant effect of each independent variable on the price of eggs was tested with a t-statistic. The hypothesis of a coefficient not different from zero is rejected for one variable at α=0.01, three variables at α=0.05 and two variables at α = 0.10. The remaining variables are found statistically insignificant. Insignificant coefficients of the variables suggested that consumers either do not have adequate information to incorporate the characteristics into their buying decisions or that they place no value on such attributes when they buy indigenous chicken eggs. The results of F-test (F=3.41significant at one percent) indicates that the independent variables as a set significantly affect the dependent variable(Price of I.C egg) .The R-square (0.26) indicates that 26% of the variability in I.C egg price is explained by the empirical model. Table IV shows estimates of coefficients, corresponding standard errors and marginal implicit prices. Marginal implicit prices are calculated by multiplying the average price with the relative change which is the corrected unbiased partial derivative of price with respect to each product attribute (Kennedy, 1981). The attributes are interpreted with respect to the default dummies (Gujaratti, 1995). White shell color is negatively related to price. Ceteris paribus a change in shell color from brown to white significantly (P<0.05) results in 8.1% decrease in price of indigenous chicken eggs. This indicates that compared to brown eggs, white eggs will attract price discounts of Ksh.1.4/egg. This is contrary to our expectation of price premium in favour of white shell. The study found that 78% of respondents preferred white eggs compared to 22% who had purchased brown shelled eggs. This implies that although consumers show preference for white shell, they are unable to value it while making purchase decision probably because of wide spread brown eggs. The estimated implicit price show a strong potential for developing niche markets for I.C eggs. This attribute was ranked third in importance by consumers while making their purchases. Cleanliness and freshness, though with an expected positive relationship to price, were not significant (P>0.10). There was no significance difference between prices of clean and non clean eggs and fresh and non fresh eggs. Further still during Kendall ranking of attributes’ importance, the two attributes ranked least on the 21 scale. This indicates the wide availability of fresh and clean eggs in the market to an extend that consumers do not make distinction between the two categories when valuing eggs. Shape of an I.C egg significantly (P<0.10) influenced the price of eggs. Ceteris paribus the price of an oval egg tends to be 7% (Ksh. 1.25/egg) less compared to that of a non oval egg. This premium price for non oval eggs could be attributed to other specific roles/purposes upon which such eggs are purchased (beautification, tourism or breeding) and the perception that they have more value than the ordinary oval eggs. Indeed, the study only found out that only 6% of consumers had purchased non oval eggs implying the scarcity nature of such eggs. This implies that breeders, producers and retailers can capitalize on this new emerging market by increasing such eggs in the market. As expected, eggs prices are found to be significantly influenced by the size. Small sized eggs were found to significantly (P<0.10) positively influence market price. Ceteris paribus, the price of a small sized egg tends to be 15% (Ksh. 2.76/egg) more than that of an extra large egg (default). Consequently, medium eggs significantly (P<0.05) influenced price and the price of a medium sized egg tends to be 9% (Ksh. 1.77/egg) more than that of an extra large egg ceteris-paribus. This study also found out that consumers bought 7%, 52%, 38% and 3%of small, medium, large and extra large eggs respectively. This implies the perception that small eggs are produced organically and hence more nutritious with minimal health related problems than large and extra large eggs which showed no significant difference in their prices. Producers and retailers should endeavour to fulfil this niche market by targeting it with small and medium sized eggs so as to fully maximize on returns. The study also found that size was the most ranked egg attribute in making the purchase decision. The price of eggs was also found to respond to other socio-economic attributes such as household size and region. The study found that large household paid a premium of 2.0% compared to small sized households indicating market imperfection. This was significant at (P<0.01). On the other hand, price of eggs varied significantly across sub counties, for instance the price of eggs in Westlands were 15% (Ksh.2.7/egg) higher (P<0.5) than the price of eggs in Kaiti. Consequently, consumers in Makueni and Starehe paid premium compared to those in Kaiti although this was not statistically significant. The price difference is due to differences in geographical locations and socio-economic status of the consumers in these regions. 22 TABLES Table I: Explanatory Variables and the Priori expectations for the study Dummy Variable Categories Priori Expectation Shell color White Brown + - Freshness Fresh Otherwise + - Shape Oval Otherwise + - Size (extra large) small medium large - + + + Table II: Socio Economic Characteristics of the sampled households Characteristic % MIN MAX MEAN SD Gender Male Female 57 43 Age __ 17 85 36.09 12.89 Household size __ 1 11 4.1 2.35 Marital status -married -single -widow -divorced 72.5 26.5 0.5 0.5 Monthly income ksh.30,000 35 28 16 21 23 Table III: Kendall Ranking of the Indigenous Chicken’ Eggs Attributes. IC egg attribute Rank Sum of Ranks Price 2 512.5 Size 1 502 Shell color 3 666.5 Cleanliness 6 829 Freshness 4 678.5 Shape 5 794 Package 7 917.5 N 175 W 0.17 F calculated 36.51 F- Critical 2.10 Significance 1% TableIV: Estimation of Hedonic Price Model for Indigenous Chicken’ Eggs Variable Coefficient Std error Relative impact Implicit price Shellcolor( white) -0.07920** 0.03230 0.0818 1.4 Freshness(fresh) 0.0446 0.0334 0.0450 0.8 Shape(oval) -0.0687* 0.0389 0.0703 1.25 Cleanliness(clean) 0.01860 0.03648 0.01809 0.32 Size (extra large) small medium large Default 0.1463* 0.0954** 0.06409 - 0.0750 0.0459 0.0430 - 0.1542 0.0989 0.06529 - 2.76 1.77 1.16 Sub county (Kaiti) Westland Starehe Makueni Gender (Male) Education Income Age Marital status House hold size Default 0.1423** 0.02364 0.0182 0.0392 0.0050 0.0121 0.0000 -0.01043 0.01936*** - 0.0584 0.05361 0.0482 0.03018 0.0133 0.01020 0.0012 0.0327 0.0058 - 0.1509 0.0224 0.0171 0.0395 0.0049 0.0121 0 0.0099 0.0195 - 2.70 0.40 0.307 0.70 0.88 0.21 0 0.17 0.34 Constant 2.6271 0.1188 F 3.41*** R2 0.2637 Adjusted R2 0.1864 N 180 24 CONCLUSION This study aimed to establish consumer preference attributes in indigenous chickens’ eggs. Hedonic pricing model was employed to analyze preference attributes and their influence on price in Makueni and Nairobi counties. Findings of the research have shown that shell colour, size and shape are important attributes influencing price and consumption of IC eggs. Furthermore, socio-demographic characteristics of the respondents such as household size and region (sub-county) influenced prices of indigenous chicken eggs confirming existence of market imperfection in the I.C market. The study made the following recommendation; first, Farmers in the study region and beyond should consider altering their feeding and management practices of indigenous chicken so as to influence incorporation of preference attributes in the IC eggs for sale through quality feeds and sound production systems. Second, researchers should incorporate attributes of choice based on consumer preferences in their breeding work to enhance acceptability of eggs in the market. REFERENCES Becker, G.S (1965). A theory of the Allocation of Time. The economic Journal, 75(299): 493-517 Bett, H., & Bokelmann, K. P. (2011). Hedonic price analysis to guide in breeding and production of indigenous chicken in Kenya. Livestock research for Rural Development 23 (6) . Carman, F.H. (1997) Agricultural Price Analysis. In Padberg D.I., Ritson C. and Albisu L.M. (eds). Agro-food marketing, CAB INTERNATIONAL, 51 - 84. Changhee, K. and Chanjin, C., (2011). Hedonic Analysis of retail egg prices using store scanner data: An application to the Korean egg market. Journal of Food Distribution Research, 42(3). Dalton, T.J.,(2003). A hedonic Model of Rice Traits. Economic values from farmers in West Africa 25th International Conference of Agricultural Economists (IAAE) (PP. 1552 - 1560). Durban, S.A: Document Transformation Technologies. Damodar Gujarati (1995). Basic Econometrics, 3rd Edition. New York: McGraw-Hill. FAO, (2008). faostat. Retrieved october 26, 2016 from FAOSTAT: faostat.fao.org. Hannah, W. G., (2015). Analysis of Consumer Preference in Product attributes: A case of common Beans in Kiambu County, Kenya. Msc. Thesis, Kenyatta University, Nairobi, Kenya. Kennedy P.E (1981) Estimation with Correctly interpreted Dummy variables in Semi- logarithmic Equations. The American Economic Review, 71(4) :p. 801. , K. (1991). Modern consumer theory. Edward Egler. Makueni County (2013). First County Integrated Development Plan 2013 – 2017. Makueni, Kenya. 25 Nairobi City County (2014). County Integrated Development Plan. Nairobi, Kenya. Okitoi, L.O, Ondway, H.O., Obali, M.P. & Murekefu, F.(2007). Gender issues in poultry production in rural households of Western kenya. Livestock Research for Rural Development, 19(2), Article No. 17. Ramatu, M. Al-Hassan, Noah Larvoe and Abigail A. Adaku. (2014). Hedonic price analysis of dressed chicken in Ghana. international journal of Business and social science, 5(12) Rebecca, J. (2012). Consumer willingness to pay for Dairy goat milk in Siaya, Kenya. Msc.Thesis. Rosen, S. (1974). Hedonic prices of implicit markets.Product differentiation in Pure competition. The Journal of Political economy 82 , 34 - 55. Timothy, W. (2006). Ahedonic analysis of cattle prices in the central corridor of West Africa: Implications for production and marketing decisions. conference paper. Uptoon, M. (2000). The livestock revolution: Implication for smallholder Agriculture. Acase study for milk and poultry production in Kenya. Apoultry discussion paper. Food and Agricultural Economics 2(1): , 23 - 32. USAID. (2010). partnerships for safe poultry in Kenya(PSPK) Programme: regional poultry value chain analysis. Nairobi: USAID. World Bank (2015). Kenya Literacy rate. World Bank Development indicators, 2015 1.2 Livestock Keepers in Isiolo County, Embrace New Adaptive Strategies as Mitigation Measures against the Effects of Drought *Anastasia Kagunyu1, Steve Nangendo2 and Joseph Wanjohi3 1Kenya Agricultural and Livestock Research Organization- Sericulture P.O.BOX 7816 Thika 2 Institute of Anthropology, Gender and African Studies, University of Nairobi P.O.BOX 8711 Nairobi 3Catholic University of Eastern Africa, P.O.BOX 8O5-00200 Nairobi *Correspondence author; kagunyu2010@gmail.com mobile; +254723770295 ABSTRACT Droughts are generally associated with the failure of the seasonal rains. The two major rainfall seasons in Kenya are the long-rains (March to May) and the short rains (October to December). Drought has become a perennial problem in Kenya with chronic vulnerability being concentrated in Arid and Semi-Arid lands (ASALs). Traditionally, pastoralists have used indigenous coping strategies to address these disasters, but recent studies have shown that the recurrence of droughts has weakened pastoralists’ indigenous coping strategies. The overall objective of this study was to investigate the new adaptive strategies being used by livestock keepers of Isiolo against negative 26 effects of droughts. Household heads were interviewed with the help of a semi– structured questionnaire. Two focus group discussions were held and the participants of each group were men and women who were above 45 years of age. The trend of droughts and the effects of droughts on Borana livelihoods and new adaptive strategies applied. Quantitative data derived from the household interviews were edited, coded and analyzed using the Statistical Package for Social Sciences (SPSS) software version 20 spread sheets. Descriptive statistics were run to give frequencies and percentages. The study findings revealed that livestock keepers have adopted new coping strategies such as small crop production, camel rearing, small scale business and informal employment. According to the study findings the pastoralists have limited skills and capital to implement sustainable projects. This study therefore came up with two recommendations; the need for livestock keepers to be trained on appropriate skills on crop production and entrepreneurship. There is also need to link the livestock keepers to financial institutions who would give loans to livestock keepers to so that they can come up with sustainable business. Key words: Droughts, coping strategies, pastoralists, livestock, Isiolo County INTRODUCTION Drought has become a perennial problem in Kenya with chronic vulnerability being concentrated in Arid and Semi-Arid lands (ASALs). According to KNBS (2010), ASALs account for 36% of the total population, around 14 million people. Droughts are associated with the failure of the seasonal rains. The two major rainfall seasons in Kenya are the long-rains (March to May) and the short rains (October to December). Natural disasters such as drought, floods, earth quakes and storms are not new to the arid and semi-arid areas. Traditionally, pastoralists have used indigenous coping strategies to address these disasters, but recent studies have shown that the recurrence of droughts has weakened pastoralists’ indigenous coping strategies. Livestock production is the major economic activity in the ASALs of Northern Kenya. The communities living in these zones have been livestock keepers since time in memorial. However, the impact of drought, increasing insecurity, and famine has led to a growing emergence of sedentary livestock keepers and experimentation with alternative livelihoods. Over recent years, pastoral communities have had to employ other supportive activities to supplement pastoralism, which has proved to be ineffective in meeting all their economic and social needs (Watson and Van Binsbergen 2008). Golicha et al. (2012) work in Northern Kenya provides information on some of the alternatives livelihoods being adopted by the pastoralists in Marsabit County which includes crop farming, apiculture, casual work, small scale businesses formal employment, weaving, crafting and sand harvesting among others. There was inadequate information about the alternative adaptive strategies applied by the Borana of Isiolo in their response to the effects of droughts. As a result, this study was initiated whose overall objective was to investigate the new adaptive strategies being used by livestock keepers of Isiolo to cushion themselves against the negative impact of droughts. The study was guided by two objectives; 27 Specific Objectives To examine the effects of droughts on the Borana pastoralists livelihoods To establish the type of adaptive strategies which are being used by the Borana community of Isiolo County MATERIALS AND METHODS Study site and unit of analysis This study took place in Isiolo Central which is a sub -County of Isiolo County situated in northern Kenya. The study focused only on the Borana community, although there other communities because it is the largest community in the County. This being the case the study was concentrated in the areas occupied by the Borana Community. The unit of analysis was the individual man and woman household. Sample size and sampling procedure This study applied both random and non-random sampling strategies. In selecting the study site simple random sampling technique was applied whereby the names of the three Isiolo sub-counties were written on pieces of paper which were folded several times. The pieces of paper were then put in a container which was shaken before the researcher picked one of them. This happened to be Isiolo Central. Three villages were selected purposively with the help of the local provincial administration who assisted to researcher to identify the three villages which were occupied by the Borana community. To get a representative sample size this study used a formula used by Mugenda and Mugenda (1999: 43-44). The accessible population in this study was 4000 households. to be used to calculate samples size.Using the above formula sample size is: To cater for those households that would decline to participate or drop out during the process of investigation, the study proposed a sample size of 400. Systematic sampling was applied to select households for the interview. The sampling interval was determined by the equation given below. Where: n= required sample size N=Population size n=400 N=4000 (i.e., 1 in 10) 28 Data collection methods Household interviews Household interviews were conducted with household heads, male or female, through the administration of a questionnaire to the respondents. The interviews were conducted in Borana local language by enumerators who were from the community. The questions answered by household heads how their livelihoods have been affected by droughts and new adaptive strategies of pastoral communities. Focus group discussions Two focus group discussions were conducted and each group had 10 participants who were selected purposively. Each group had men and women who were above 45 years of age. The issues discussed included the effects of droughts on the Borana community livelihoods and their new adaptive strategies. Data processing and analysis Data obtained from focus group discussions were sorted out and interpreted in relation to the research general objective, to enable the researcher to provide overall interpretation of the findings showing how thematic areas and issues relate to one another. On the other hand, quantitative data derived from the household interviews were edited, coded and analysed using the Statistical Package for Social Sciences (SPSS) software version 20 spread sheets. Descriptive statistics were run to give frequencies and percentages. RESEARCH FINDINGS Effects droughts on the livelihoods of the Borana community The study findings indicate that droughts have affected the livelihood of the Borana pastoralists to a great extent. The respondents said that their wellbeing has been affected in three ways as shown in Figure 1. 28 29 30 31 32 33 34 35 36 37 P e rc e n ta g e Effects of droughts on livelihoods of the Borana Deaths of Livestock Displacement of families Poverty 29 Figure 1: Effects of droughts on livelihoods of the Borana Thirty-six per cent of the respondents stated that severe and prolonged droughts had contributed to the death of livestock, which is the major economic activity of the community. Droughts had caused fluctuations in the livestock population through increased mortality and reduced birth rates due to decreased forage and water availability. Thirty-three per cent of the respondents stated that droughts had caused displacements as people moved with their livestock in search of water and pasture for many months leaving women, the elderly and children behind. In some cases this causes family breakdowns since some of the women engage in immorality so as to get finances to feed the remaining family members. Lastly, 31% of the respondents stated that poverty had become rampant in the Borana community. This argument was supported by focus group discussants who said that massive losses of livestock drove off large numbers of pastoralists from the pastoral system, which was their major source of livelihood. Effects of droughts on the environment The findings of this study reveals that droughts had affected the natural resources used by the Borana to feed their livestock (Figure 2). This is because half (50%) of the respondents said that trees and grass had dried up, 30% stated that rivers and water pans had dried up, while 20% said that desertification had increased. Figure 2: Observed effects of climate variability on environment New Adaptive Strategies The study findings suggest that climate variability has made the Borana community to diversify their livelihoods. Focus group participants stated that traditionally pastoralism was a full-time occupation and everyone from an early age was engaged. However, frequent and prolonged droughts had made the community to look for alternative ways of getting income. According to the results of this study, 33.3% of the respondents stated that they engaged in causal labour, 30.3% have turned to small-scale farming, while 24% engaged in small-scale business. Lastly, 12.5% of the respondents 50% 30% 20% Drying up of trees and grass Drying up of water pans and rivers Increased desertificatio n 30 indicated that people had joined salaried employment. Figure 2 summarises the alternative livelihoods of the respondents. Figure 3: Alternative adaptive strategies of the respondents This study indicates that even though livestock keeping was the most preferred source of livelihood among these pastoralists, there was increasing income diversification in the community particularly as an immediate survival mechanism during and after drought periods. Focus group discussants stated that many pastoralists are exiting from the traditional ‘highly mobile’ form of nomadism and entering into agro- pastoralism or sedentarisation and other livelihood options. According to Oxfam GB (2010:17) some households send members of the family as labourers to towns or other regions, or even abroad (usually to Middle Eastern countries) to work and send remittances for household use. More and more households send children to school in the hope of improving their access to employment in the future, while other households move and settle close to food distribution centres to secure better access to food. Shifting of livestock preference from Cattle to camel rearing Respondents were asked about the most preferred livestock species in the region. A majority (71.5%) of them said that they preferred camels. The reason they gave for this was that camels are most suited for the region as they can stay for many days without water and food. This is in agreement with Field (2005:93) who states that camels are the best suited in the ASALs due to their biological and physiological adaptations, which help them to cope with the hot and arid environments. In addition, they have varied diets which include shrubs and their height allows them to browse tall trees. Focus group participants were in agreement that drought had affected the range vegetation and this had impacted on the herd composition of the Borana. According to them: In the last 50 years we used to have good grasslands and forages. This enabled us to have different livestock species. Today our range lands that were once covered with good grass-land have been replaced with unpalatable hardy and woody species. A change in vegetation composition from grass land to woody and unpalatable plant species, has forced pastoralists to 0 5 10 15 20 25 30 35 P e rc e n ta g e Alternative adaptive strategies Casual labour Farming Business Salaried employme nt 31 alter their livestock composition from grazing to browsing species, that is, from cattle and sheep to goats and camels. (FGD participants) Field (2005:100) estimated that the volume of milk produced by camels is six times that produced by indigenous cattle found in the ASALs. In addition, the milk is believed to have medicinal value, helping in the management of high blood pressure and diabetes in human beings (Ngeiywa et al., 2012:20). Focus groups discussants revealed that even the clans among the Borana, namely, Karaiu Mbere and Aswa, who used to consider camel rearing and utilizing camel products as a taboo have started rearing them and using camel products. However, 22.3% of the respondents said that they would prefer goats and sheep because they give birth twice per year and sometimes give birth to twins. Focus group discussants indicated that small ruminants were easy and cheap to restock after a disaster. This argument concurs with that in Spore Magazine (2008:14) which states that livestock keepers have started rearing small ruminants as a mitigation measure against climate change. Those who were in favour of small ruminants indicated that they applied the zero-grazing system of production, whereby the owners were not required to go to insecure grazing areas. In addition, goats were considered to cause difficulty to livestock raiders as they disperse when approached. On the other hand, 4.5% of the respondents said that they preferred rearing cattle and the reason they gave for this was that when they sell cattle they get higher incomes as compared to other livestock. In the past, the Borana community was attached to cattle; they used to claim that they are the “owners of cattle”. Traditionally, any person in the Borana community who never owned cattle was considered to be very poor and such a person was loaned cows by relatives so as to build a herd. The research findings suggest that the trend has changed and even development agents like Arid Lands gave camels to pastoralists as a restocking strategy. Finally, 1.5% of the respondents indicated that they had engaged in poultry production as these are never stolen by raiders. CONCLUSION AND RECOMMENDATIONS The finding of this study have revealed that drought has a great impact on the Borana livelihoods. As a result of this the community has engaged itself in new adaptive strategies which included causal labor, formal employment, small scale farming and business. However, it came out clearly that the alternative adaptive strategies the Borana community is engaged in are short-lived and not sustainable since the community has limited skills and capital to enable it have better gains. This study therefore came up with two recommendations; the need for livestock keepers to be trained on appropriate skills on crop production and entrepreneurship. There is also need to link the livestock keepers to financial institutions which would give loans to livestock keepers so that they can come up with sustainable initiatives. 32 REFERENCE Field, C. 2005.Where there is no Development Agency: A Manual for Pastoralists and their Promoters. Aylesford, Kent, UK: Natural Resources International. Golicha D, M Ngutu and H Charfi 2012. Livelihood diversification options for pastoralists 4(11) 12 Kenya National Bureau of Statistics 2015. Economic Survey 2015. Nairobi, Kenya: Kenya National Bureau of Statistics. Ngeiywa, J., J. Njanja, C. Toritich, A. Khalif, G. Wamwere and A. Guliye 2012. Camel Manual for Service Providers. Nairobi: Kenya Camel Association. Oxfam GB 2010. Pastoralism Demographics, Settlement and Service Provision in the Horn and East Africa: Transformation and Opportunities. London: Overseas Development Institute. Watson, DJ and van Binsbergen J. 2008. Livelihood diversification opportunities for pastoralists in Turkana, Kenya. ILRI Research Report 5. ILRI (International Livestock Research Institute), Nairobi, Kenya. 43 pp. 1.3 Milk production in Kenya: Challenges and opportunities Christine Ombuna, Kenya Industrial Research and Development Institute, P.o Box 30650 -00100 Nairobi, Kenya. Corespondence: christineombuna@gmail.com, +254 708 455 734 Abstract Dairying in Kenya is characterized by small scale and labour intensive operations such as hand milking. The aims of this research were to determine trends in milking by hand, by milking machines and technology needed for dairying. The research used an interview schedule and convenience sampling method in Naivasha, Nakuru, Eldoret, Kitale, Nyandarua and Nairobi regions. A total of 74 farmers milking by hand, 8 farmers using milking machines, 50 experts in Kenyan dairy sector, 5 dairy processors and 9 dairy equipment dealers were surveyed. The results indicated that the breeds reared were either pure or crosses of Friesian (56%), Aryshire (30%), Jersey (7) and others 7%. The difference in breeds reared by farmers milking by hand and machine was not significantly different (P<1). About 9.7% had adopted milking machines and skilled hand milking personnel were 13%. The duration of milking by hand was mostly (62%) above recommended time (seven minutes). Majority of farmers (95%) and 96% of experts recommended adoption of milking machines for high milk quality, improved production and health of animal throughout lactation periods. The reasons farmers were not using milking machines were the cost, few herd and electricity. Farmers who milked by hand and supplied milk to dairy processors were 42%, of this, 37% had their 33 milk rejected for failure to meet quality standards. Each of sampled dairy processors received 60%-70% of milk processing capacity per day. Adoption of milking machines has been prevented by high initial and operation costs, lack of designs applicable for farms with less than 20 cows, sensitization and diversified marketing. The ones using milking machines are challenged with inadequate water and electricity. Key Words: Hand Milking, Machine Milking, Challenges, Technology needs INTRODUCTION Background of the problem Kenya produces over 3 billion litres of milk per year, 70% of which is from dairy cattle. Total milking herd is approximately 27.7 millions of animals (FAO, 2011). Milk production is mainly from cattle, camels and goats. Smallholder farmers dominate dairy production (80%), with limited land and small dairy herds (Majiwa et al., 2012). Milking is one of the major and final operations that determine profitability of a dairy farm. However, farmers are faced with several challenges that include low productivity, poor hygiene and hand milking routines. Hand milking exposes dairy animals to injury, dangers of disease transmission and incomplete emptying of the udder which complicate the health of the cow as well as subsequent milk yield (Dzidic, 2004; Millogo, 2010). Hand milking is slow, very tiresome and unhygienic. It is difficult for milkers to milk each cow within seven minutes. These challenges can be mitigated by investing in machine milking. The cost of farm machinery is high and/or not suitable for small scale farms which have minimized mechanization, hindering maximum yield. The models of milking machines imported into Kenyan market; majority of small scale farmers cannot afford to buy and use. They require high initial capital and huge investment in electricity supply. Secondly, the design specifications for milking machines may be based on requirements of dairy industry of Countries of origin which would vary with Kenyan dairy industry requirements. Factors such as type/breed of dairy cattle reared, milk production per cow and average number of dairy cattle per farmer influence design specifications. The research was conducted in some Kenya’s dairy hub (Naivasha, Nakuru, Eldoret, Kitale, Nyandarua and Nairobi) that established local dairy trends. The aims were: to assess current usage and demand of milking machines, determine type and average number of cattle reared per farmer, challenges faced by farmers milking by hand and by a machines and obtain measures required to be undertaken to improve Kenyan dairy sector. 34 MATERIALS AND METHODS Research Methods A stratification was done into six regions; Naivasha, Nakuru, Eldoret, Kitale, Nyandarua and Nairobi. Figure 1 shows a section of Kenya map where the survey was done. A further classification of respondents was done into farms milking by hand, farms using milking machines, experts, dairy processors and dairy equipment/machines dealers. Fifteen samples of farms doing hand milking were chosen from each region except Nairobi. Due to low adoption of machine milking, uniform sample size per region was unattainable. Four farms machine milking were surveyed in Nakuru and two in Naivasha. In Nyandarua, Eldoret and Kitale one sample was surveyed in each region. Experts were drawn from Dairy Training Institutes, dairy processing industries/Equipment dealers and Research and Development institutes. The minimum qualification was a Bachelors degree and in a position of a lecturer, supervisor or researcher in dairy sector. They were interviewed at their respective place of work. Six dairy equipment/machine dealers were from Nairobi and three from Nakuru. The dealers would not be found in other regions. Two dairy processors were sampled in Nakuru and Nyandarua, one was from Naivasha. No processor was assessed in Eldoret, Nairobi and Kitale. A convenience sampling method was used to reach respondents. The survey was done from January to March 2015. It involved observation on milk production systems, interviewing and filling questionnaires that had been pretested. The data was analyzed using SPSS version 22 and Excel 2007 Microsoft programme. The analysis was mostly descriptive. RESULTS AND DISCUSSIONS A total of one hundred and forty six samples were obtained, seventy four were farms milking by hand, fifty experts, eight farms using milking machines, nine dairy equipment/machines dealers and the remaining five were dairy processing plants. Trends in milk production and management The study showed that majority (54%) of farmers milking by hand had a dairy herd of 1 to 10 cows (mean of 6). About 27% had a herd of 25 – 58 cows. Table I shows number of cows owned by farmers milking by hand. The breeds were either pure or crosses of Friesian (56%), followed by Aryshire (30%), Jersey (7%) others were 7%. Friesian cows dominated small holder farms in Western Kenya (Omondi et al., 2014) which agrees with this research. Majority of milking people were Semi killed 58%, unskilled 29% and skilled 13% (had training). About 60.5% of farms hired milking people. Unfortunately, most milking personnel highlighted that they were not happy with milking routine work. This was noted as the rate of hiring was temporary and frequent workers’ exit. About 39.5% farms used family members as milkers. If different people milk, clear milking routine instructions and training are necessary for quality milk production ( Ruegg, et al., 2005). 35 The farms using milking machines were 9.7%; the ones who used herringborn (Parlour system) milking machines had a herd of 400-800 (37.5%) and 100-400 (50%). About 12.5% had herd of around 25 -40 cows and had adopted bucket milking machines. Table II shows total herd owned and milked by farmers using milking machines. In about 62.5% of farms machine milking, milking herds were below 50% of the total herd. The mean of milked herd to total herd was 51%, a higher rate than one indicated by Kenya Dairy Board (32.3%) survey of 2013. The means of Kenya dairy board was based on farmers milking by hand and those using milking machines (all dairy farmers) which might have pulled the mean down. The mean of this study (51%) is based on farmers using milking machines. This indicates that, embracing technology in farms increases productivity. The breeds reared by farmers using milking machine were; Friesian which accounted to 56%, Aryshire 32%, Jersey at 7% and other breeds were 5%. The difference in breeds reared by hand milking and machine milking farmers were not significantly different (P<1). The time spent in milking each cow is very important in ensuring that maximum milk yield is attained throughout the lactation periods. The time of milking each cow by hand is a function of milk yield per cow and skill of a milking person. Seven minutes is the recommended maximum. However 62 % of farmers milking by hand exceeded this time with a number of them going up to twenty minutes. It was encouraging to note that some farmers had information that they should milk within seven minutes. The duration of milking by a machine is a function of milk yield of a cow and pulsation rate of a machine. All farmers who used milking machines were within the recommended time (seven minutes), most of them (63%) attained five minute and below. Day to day milk variation of machine milked cows is 6-8% while a hand milked and restrictedly suckled cow is high, 18 -21% (Millogo, 2010). There is 5.5% increase in milk yield in standardized milking procedure compared to a variable milking procedure (Ruegg et al., 2005). The procedure may, among other factors be influenced by level of motivation or stress of a milking person ,which varies from day to day and caused by many factors surrounding a person unlike a milking machine. Hand milking farmers who supplied milk to processing plants were 42%, while 47.3% sold direct to consumers such as hotels, institutions, shops and neighbors. A 10.7% did not sell their milk. Those who supplied milk to a processing plant, 37% indicated that the milk would often be rejected because of failure to meet quality standards. Quality assurance managers highlighted that the rate of rejection was high to milk from hand milking farmers and to that supplied by brokers. These was attributed to milking procedures and dirty milking parlors that are muddy during rainy season which increases contamination between milking person and milk. Brokers’ urge of making more money, skim or neutralize milk which causes rejection. Farmers shy from supplying milk to processing plants due to low prices offered by the plants. Another reason was fear of rejection and therefore, sell direct to consumers who are not able to analyze its quality before use. Milk is contaminated immediately after milking with bacteria and the level of contamination becomes harmful when milk reaches a consumer (Millogo, 2010). Many 36 small scale farmers in Romania were hand milking, had no milk storage tanks to store milk at right temperatures and used dirty milking parlours (Popescu, 2015), depicting unhygiene conditions where milk is produced. Their milk was often rejected by milk processors for failure to meet European Union standards; which agrees with findings of this study. Consumers must be careful in consuming milk that is direct from farmers or has not been processed. Challenges and opportunities in milk production A 95% of farmers milking by hand indicated that they would like to use milking machines, while 2.5% said that they do not want a machine and the remaining 2.5% were undecided. Farmers highlighted various challenges (Table III) that have prevented them from using milking machines. The farms that were hand milking and connected to electricity from national grid were 55%, 32.4% were connected to three phase while 22.6% had single phase. The ones not connected to electricity accounted to 45%. This agrees with factors in table III; that major reason why farmers are not using milking machines is not access to electricity, but the cost involved and herd size. Table I indicates that about 27% of farms have 25- 58 dairy cows and some were connected to electricity but, were hand milking yet there are milking machines in the market for such a capacity. This has to do with the cost involved and lack of awareness of existence and importance of using milking machines. The use of milking machines in Kenya dates back to 1970s. KALRO Naivasha (Dairy Research Institute) was the oldest in the samples with over 40 years of operation. Herringbone (8 aside) milking machine was installed in the institute in 1970s and was still in operation. Other farms had improved from bucket type of milking machines to Herringbone due to need of quality and zero contact of human beings with milk. About 50% of farms had used milking machines for more than 10 years, an indication of reliability. Majority, 63% highlighted that they faced challenges in using milking machines that included blackouts or low voltages, parts breakages, low pressures and resistance of the cow when introduced to a machine for the first time. The farms using milking machines that were connected to electricity from national grid and had a standby generator were 62.5%. While 37.5% depended only on electricity from national grid and reverted to hand milking in case there was no electricity. The abrupt revert from machine milking to hand milking creates variation (Johnson, 2000). It would be good if all farmers would have power back up. A dairy farm can generate its own electricity that can be derived from biogas and/or solar. Unfortunately, most farms had not ventured into these technologies for their dairy production. Cleaning a parlor (herring borne) milking machine requires an approximate of 600 to 1200 litres of water after every milking. In some of milking parlours, it was evident that the cleaning process was not followed as per requirements of manufacturer. Some measuring cylinders had stained due to poor cleaning. Some farmers used different detergents from ones recommended and were inaccessible to enough water. A bucket type of milking machine requires an approximate of 20 – 30 litres of water for cleaning. Hand milking requires about 5-8 litres of water. It might require farms to have more 37 access to water and electricity if they adopt milking machines. These are some of operation costs that hinder farms from using milking machines. Views from Experts in Kenyan Dairy Sector Concerning Milk Production The experts interviewed were Animal scientists, Veterinary doctors, Engineers, Dairy technologist and/or farm managers. They gave their views as indicated below. A 96.3% of respondents recommended farmers to use milking machines by giving the following reasons; efficient milking with complete emptying of udder within seven minutes, high milk quality, and low chances of mastitis. Does not require skilled hand milking personnel, accuracy in recording and monitoring milk yields per cow, consistency, reduce theft of milk and reduce milk rejection at selling points. The other, 3.7% failed to recommend the use of milking machines citing that it is expensive to buy, install and operate, some animals are stressed up and teats block or die. The expert’s opinion on efficient milking with complete emptying of udder within the required time (seven minutes) agrees with the findings of the author’s survey with farmers using milking machines, most of them attained five minutes and below. High quality milk from farmers using milking machine is true as their milk was not rejected by milk processors. Milk from hand milking farmers was often rejected. There is high somatic cell count which influence milk quality in smallholder dairy farms milking by hand (Vo Lam, 2011). The high somatic cell count is as a result of poor udder health. Careless milking hygiene causes subclinical mastitis at cow and herd levels. There is high chances of controlling mastitis by machine milking than hand milking (Graeme, 2012) which agrees with experts’ opinion. Milking machine is unlikely to cause mastitis and harm on a farm if it is installed, operated, maintained, tested and functions according to standards (Laven, 2016). Herring borne milking machine is designed with a measuring cylinder which records milk yield per cow. The system pumps milk to a milk cooling storage tank that can be locked minimizing milk theft. The experts were asked about the measures that have been put in place to encourage farmers to use milking machines. About 22.3% indicated that there have been exhibitions and demonstrations by machine dealers and extension officers during farm fields and shows. In addition, commercial banks advance loans for purchase of milking machines. Majority of experts (77.7%) felt that no measures had been put in place to encourage farmers to use milking machines. About 4.4% of farmers milking by hand indicated that they do not know the importance of using milking machines, 2.2% were not aware that milking machines exists . This calls for more sensitization on existence and importance of using milking machines. When asked about the measures to be undertake to improve dairy production, the experts recommended that a public organization to work with farmers to design and develop milking machines affordable and applicable to small scale farmers. Increase in supply and reliability of water and electricity should be increased. Adequate water intake increases milk yield (Vo Lam, 2011), increasing water to farms should be a major 38 undertaking. The government should lower prices or partner with donors in promoting the use of milking machines. In addition, the experts suggested a proper civic education in advantages of mechanization and reduction in the cost of farm machinery and spare parts. County governments were urged to invest in milk cooling plants in dairy hubs. Milk marketing was noted by experts as one of the challenges facing dairy sector. They felt that an improvement is required in marketing channels, infrastructures, milk pricing and guidelines to reduce exploitation by middlemen. Development of improved marketing systems is a gateway to increased production (Kedija, 2007). Animal production attributes such as an introduction of high yield and quality genetics, training on pasture and fodder production technologies were recommended. Better ways of feeding, breeding, health and policies that encourage animal production and processing were emphasized as well. Promotion of research to solve local problems was additionally suggested by experts. Dairy Processing Industries and Equipment/Machine dealers The milk processing industries surveyed received about 60% - 70% of their milking processing capacity per day. The volumes rejected were 6,000 – 7,000 litres per day. Milk was received from brokers, individual farmers or processor’s own cooling plants. On delivery, milk was checked for fat and protein percentages, acidity, density, total pathogenic germs and somatic cell counts. The standard fat is 3.5%, number of pathogenic germs is a maximum of 100,000 per ml of milk and total somatic count is a maximum of 400,000 per ml of milk. The parameters reflect where milk was produced and the degree of disinfection used (Popescu, 2015). The parameters determined the rejection or acceptance of milk. Industries are supposed to run throughout to conserve energy. Due to in ability to receive 100% of their capacity per day, short shut down happens which are costly to the industry. The extra energy cost incurred has a high probability of transfer to a consumer. Some farmers were not supplying milk to dairy industries because of low prices offered by industries. Proper milk pricing motivates farmers to improve production. This requires industries to review prices or inputs they give to farmers in order to encourage farmers to supply milk to them. Kenya imports milking machines majorly from Sweden, Turkey and China. The classes of these machines are; bucket, trolley (single or double) and Parlors (Herringborn). Stock turnovers of milking machines were low. Out of nine dairy equipment dealers surveyed; only three had stocked milking machines. The dealers were concentrated in Nairobi, the capital city, forcing farmers to travel long distances for purchase of machines/Equipment and spare parts. 39 TABLES Table I: Dairy cows owned by farmers milking by hand Dairy Cows Frequency Percentages 50 -60 10 13.5 27% This category can use bucket milking machines 40-49 4 5.4 25 -39 6 8.2 15-24 10 13.5 73% This category may find it uneconomical to use any current designs of milking machines 10-14 10 13.5 5-9 16 21.6 1-4 18 24.3 Total 74 100 Source: Author’s calculations Table II: Total Dairy Cows and Milked Cows for farmers using milking machines Total Dairy cows Milked cows Percentage Milked 162 48 30 325 140 43 320 144 45 150 58 39 40 25 63 450 220 49 473 320 68 800 600 75 Source: Author’s calculations Table III: Factors that have prevented farmers from using milking machines No Factors Preventing the use of milking machines Percentage of farmers 1 It is costly to buy, install and operate 53.3 2 Have few cows 20.2 3 Have no access to electricity 11.1 4 Have no knowledge of the importance of using milking machines 4.4 5 Low milk yields 4.4 6 Lack of appropriate farm structures 4.4 7 Not aware that milking machines exist 2.2 Source: Author’s calculation 40 CONCLUSIONS Milk production is dominated by smallholder farmers; about 73% own less than 25 dairy cows and around 46% own less than 10 cows. The adoption of milking machines is in farms with a milking herd of 25-600 and total herd of 40-800 cows. Majority (95%) of farmers milking by hand and 96% of experts would like or suggested adoption of milking machines for high milk quality, improved production and health of animal throughout lactation periods. Lack of adoption of milking machines by majority of farmers has been prevented by high initial and operation costs, lack of design applicable for farmers milking less than 20 cows. Lack of sensitization and diversified marketing has also hindered adoption of milking machines. The ones using milking machines are faced with a lack of enough water and electricity. Farmers milking by hand are usually unable to get regular/skilled milking people, their milk fail to meet European Union or local standards and lack guidelines on milk selling prices. To improve dairy production, experts’ feels that the government needs to promote mechanization, improve animal genetics, health, feeds, research, marketing and processing of dairy products. ACKNOWLEDGEMENTS Thanks to almighty God for power and opportunity to do this work. Profound appreciations go to Kenya Industrial Research and Development Institute (KIRDI), where the research was authorized and facilitated. A lot of thanks to Wanjala . G; research scientist at KIRDI, for successfully networking with stakeholders and participating in data collection. Gratefulness to Mwangi .J and Kithuka . J for participation in project proposal development and in data collection. Appreciations also go to all respondents for allowing access to their premises and responding to the interviews. REFERENCES Agatha Popescu. 2015. Research on the trends in milking livestock and milk production in Romania. Scientific paper series management, Economic Engineering in Agriculture and Rural Development. Vol 15, Issue 1. Alen Dzidic. 2004. Studies on milk ejection and milk removal during machine milking in different species. Technical University of Munchen. Andrew P. Johnson. 2000. A proper milking routine, the key to quality milk. National Mastitis Council Annual meeting proceedings, Page 123. FAO. 2011. Dairy development in Kenya by H.G Muriuki Rome. Graeme A. Mein. 2012. The role of the milking machine in mastitis control. Elsevier, volume 28, pp307 - 320. Majiwa.E.B, Kavo .M.M and Murage H. 2012. Smallholder Dairying in Kenya. The Assessment of the Technical Efficiency using the stochastic production Frontier model. Jagst Vol.14(2). 41 Kenya Dairy Board. 2014. Milk production and marketing in Kenya, survey report. Kedija Hussein. 2007. Characterization of milk production system and opportunity for market orientation. Haramaya University. Omondi W and Njehia, B. 2014. Herd characteristics on smallholder dairy farms in Western Kenya. J Anim .Scie. 4 (8) 996 -1003. Ruegg. P, Morten.D and Doug.R. 2005. The seven habits of higly successful milking routines. http://milkquality.wisc.edu/. Richard Laven. 2016. Masititis control and management. Mastitis part 9 – The milking machine. National Animal Disease Information Service (NADIS). Vinsoun Millogo. 2010. Milk production of hand milked Dairy cattle in Burkina Faso. Swedish University of Agricultural Sciences; Uppsala. ISBN 978 – 91 – 576 – 7481– 4. Vo Lam. 2011. Milk production on smallholder Dairy cattle farms in Southern Vietnam. Department of Animal Nutrition and management, Uppsala. ISBN 978 -91-576- 7582- 1.4 Infectious Bovine Keratoconjuctivitis (Pinkeye) Infection in a Dairy Cattle Herd in Kenya Purity N Nguhiu and Obed K Tum Department of Animal Sciences School of Agriculture and Enterprise Development Kenyatta University, P.O. Box 43844- 00100, Nairobi, Kenya ABSTRACT Infectious bovine keratoconjuctivitis (IBK) or Pinkeye is a highly contagious, non fatal eye disease of cattle worldwide but has marked economic impact due to production losses, rapid spread of the disease in a herd and treatment costs. A cross sectional study was done to determine the causes of pink eye infections in Ilulla dairy farm in Eldoret of Uasin Gishu County, Kenya from November 2015 to February 2016. Any cattle that showed signs of eye disease (lacrimation, photophobia and corneal opacity) were recruited for this study. A total of thirty heads of cattle that were positive on clinical examination of the affected eyes, were restrained in a crash and conjunctiva swabs were collected for bacteriological evaluation. Bacterial cultures were grown on blood agar media and colonial morphology recorded. Bacteria were identified after staining bacterial smears with Gram’s stain. Antibiotic sensitivity tests were done for positive colonies. Moraxella bovis, a gram negative short rod which occurred in pairs or singly was confirmed in 33% of the ocular infections, other bacteria isolated included Staphylococcal aureus at 67% and Escherichia coli at 7% of the infections. Majority of the Moraxella infections were observed in calves 4 out of 8 (50%) and heifers 5 out of 14 (35.7%) while only 1 out of eight (12.5%) adults were infected. The infections were 42 observed in Friseans, Ayrshires and Guernseys. The bacteria were sensitive to wide range of antibiotics tested but were resistant to sulphamethoxazole. Some of the risk factors observed include was high density of face flies, feeding of dry dusty hay and the long dry spell during the time of the study. The infected animals were treated using Ampliclox® and measures to control flies around the milking parlour were instituted. Key words: Keratoconjuctivitis, pink eye, dairy cattle, treatment INTRODUCTION Infectious bovine keratoconjuctivitis (IBK) or Pinkeye is a highly contagious, non fatal eye disease of cattle worldwide but has marked economic impact on the cattle industry. Ocular infections in cattle are usually acute and tend to spread rapidly. Pinkeye in cattle easily costs producers an average of $100 a head due to reduced weight gain, added treatment costs and discounts on sale day. University research has shown that pinkeye can reduce weaning weights in calves as much as 40 to 60 pounds (AG0066, 1998). Morexiella bovis (M. bovis) is the primary infectious agent initiating pinkeye. Other microorganisms initiating pinkeye include Chlamydia, Mycoplasma, and Acholeplasma, or viruses such as the Infectious bovine rhinotracheitis virus (IBRV), which can either add to the severity of the disease process or may serve as predisposing factors permitting a secondary infection with M. bovis (Ref). Other predisposing factors are excessive ultraviolet light (sunlight), the face fly (Musca autumnalis), the housefly (Musca domestica), the stable fly (Stomoxys calcitrans), plant material, and dust (Smith, 2002). Ultraviolet (UV) light is especially a problem for cattle lacking pigmentation around the eye. Lack of pigmentation allows increased UV radiation to sensitize the eye, resulting in inflammation and subsequent infection (Radostits et al. 2000). Most of the ocular infections are transmitted by flies, commonly the face fly (Musca autumnalis) but also by the housefly (Musca domestica), stable fly (Stomoxys calcitrans) and Arcyophora longivalvis, a moth that feeds on the ocular secretions of cattle (Carter et al., 1995). The disease is seen in both young and the adult animals. However, it is more common in beef breeds and is aggravated by grazing in tall grass, by a dry dusty environment and by the presence of the insect vectors (Carter et al., 1995). Young animals are affected frequently with one or both eyes affected (Radostits et al, 2000). Several studies have shown that the economic impact of ocular infections can be significant due to decreased weight gain in young animals, reduced milk yield, cost of treatments and labor decreased value of calves as a result of disfigurement of the eyes and losses in slaughter cattle (Smith, 2002). Disease outbreaks are common during the dry season due to presence of dust and other mechanical irritants. Other predisposing factors for the disease are; prolonged exposure to sunlight (ultra violet rays), concurrent infections, vitamin A deficiency and breed susceptibility has also been reported with Bos indicus being least susceptible (Coetzer and Tustin, 2004). Animals having pigmented skin around the eye are more resistant to infection (Coetzer and Tustin, 2004). Young animals are affected frequently with one or both eyes affected (Radostits et al, 2000). 43 Previously before this study, the herdsmen at Illula farm had reported an increase in the number of ocular infections, treated them with cloxacillin eye ointments and tetracycline powder with some success but resurge infections occurred at the onset of the drier period. MATERIALS AND METHODS Area of study This study was done at Illula dairy farm, which is a privately owned farm, in the outskirts of Eldoret town, Uasin Gishu County, Kenya. The farm lies at the coordinated of 00311N, 35016 E. It is in the highlands of Kenya at an altitude of 2073 M, average temperatures of low 16.9 0C and high of 23 0C. The area receives rainfall throughout the year with annual rainfall of 1100mm, peak months being April/ May and July/ August. The drier months with low rainfall are October to February. The area has 12 day and 12 night hours. The area is prime agricultural land with good soils to support pastures. The farm is 400 acres and supports 150 adult cattle, 50 calves, mainly Friesians, Ayrshire and a few Guernsey and 100 Dorper sheep. The main economic activity is dairy production. Study animals All the cattle that showed signs of eye disease (lacrimation, photophobia and corneal opacity) were recruited into the study and they totaled 25. They comprised of nineteen Friesians (4 adults, 8 heifers, 7 calves), three Ayrshires (2 adults, 1 heifer) and three Gurnsey (all heifers). The age classes were Adults > 24 months, Heifers >9 < 24 months and calves 1 to 8 months. Five Friesians (2 adults, 2 heifers and 1 calf) which had no clinical symptoms were also included as negative controls as they had no obvious symptoms of ocular infections. The animals had specific ear tag numbers and names. A total of thirty animals comprising of 8 adults, 14 heifers and 8 calves were selected. Clinical Examination and sample collection The ear tag number, name, breed and age of each of the study animals were recorded. The study animals restrained in a crush for close clinical examination and collection of conjunctiva swabs bacteriological evaluation. Samples from the conjunctiva of both eyes of each study animal were taken with sterile cotton swabs. Each swab was rolled in the eye for about 30 seconds for each of the eyes. Care was exercised to avoid injury to the animal’s eye during sampling, and also contamination of the sample by the animal’s skin. It was done carefully to avoid touching the eyeball and any other part of the animal with the swab. The swabs were placed into a sterile labeled tubes corresponding to the ear tag numbers and transported with an hour to the laboratory for bacteriological analysis. A total of thirty samples were collected and submitted to the laboratory. 44 Bacteria culture and sensitivity tests Each sample was streaked onto plates with blood agar and incubated aerobically at 37℃ for 24 hours. The colonial morphology was examined and described. Bacteria smears were made from individual colonies, stained with Gram stain for microscopic examination and identification of the bacterial organisms. Isolation and identification of Morexiella bovis were carried out according to previously described protocols [Holt, 1994]. Antibiotic sensitivity tests were done using Himedia KGL 2/4 octodisc impregnated with Ampicillin, Tetracycline, Streptomycin, Cotrimazole, Sulphamethazole, Kanamycin, Gentamycin, and Chloramphenicol and incubated at 370C for 24 hours. RESULTS The study animals comprised of 8 calves, 14 heifers and 8 adults while the breeds represented were 23 Friesians, 4 Aryshires and 3 Guernsey. The cattle with signs of ocular infection were in poor body condition, and were mostly yearling heifers. Most cases were unilateral, but bilateral eye infection was observed in one Ayrshire heifer. The signs noted were excessive lacrimation with matting of the facial hair, photophobia, with animals presenting with closing the affected eyes to avoid exposure to sunlight, reddening and swelling of the eyelids including the third eyelid (Figure 1). Out of the 30 samples collected 10 (33%) were positive for Moraxella bovis, Staphylococcus aureus accounted for 18 samples (60%) while 2 samples (7%) were positive for Escherichia coli. Among the cases that were positive for Moraxella bovis, 6 were heifers, 3 calves and 1 adult, and on breed distribution the infections were observed in 8 Friesians, 1 Ayrshire and 1 Guernsey. The prevalence of Moraxella bovis infections was recorded at 6% of the herd with majority of the infections occurring in heifers. No eye infections were observed in the dorper sheep in this farm. The colonial characteristic of Moraxella bovis showed scanty growth on blood agar of small round whitish colonies (Figure 2). Figure 1: Ayrshire cow swollen eyelids, lacrimation and matting of facial hairs 45 Figure 2: Colonial characteristics of Moraxella bovis on blood agar from eye infections of dairy cattle in Eldoret, Kenya On microscopic evaluation, the organisms occurred in pairs and were gram negative bacteria. The isolated organisms were sensitive to most of the antibiotics used, with largest zone of inhibition with ampicillin and Sulphamethazole having no effect (Figure 3). Figure 3: Antimicrobial sensitivity test for the Moraxella bovis eye infections of dairy cattle in Eldoret, Kenya 46 High percentages of the cattle with signs of ocular infection were in poor body condition, and were mostly yearling heifers. Most cases were unilateral, but few animals with bilateral eye infections were present and were walking by making careful steps to avoid hitting objects this was witnessed mainly in heifer Ayrshire. Other signs noted were lacrimation leading to matting of the facial hair, many flies were seen around the eyes and the entire face feeding on the tears, photophobia presenting with animals were seen closing the affected eyes to avoid exposure to sunlight and corneal ulceration in some animals (Figure 1). The common clinical findings in cases of ocular infections were; photophobia, blepharospasms and epiphora; later the ocular discharge may become mucopurulent, opaque areas in the center of the cornea appears in about two days, and by day six the entire cornea will have a gray-white to yellow color with deep central ulceration of the cornea. Some of the animals were blind either unilateral or bilaterally. DISCUSSION Most of the animals affected were heifers (46%) followed by calves (26%) a few adults were affected. Other findings included low weight gain in young animals and reduced milk yield in adults which has leads to high economic loss as well as treatment cost. These findings are similar to studies in Australia in beef cattle (Champness, 2008; Snowden et al., 2005). The high prevalence of face flies (Musca autumnalis) suggested them as the main predisposing factors of this disease which is in agreement with the work of authors (Cheng, 1967). Other predisposing factors that are also found in this study included tall grasses which could cause irritation of the eyes and feeding of dry hay. Ocular infections in Illula Dairy farm Eldoret lead to losses associated with decreased weight gain in young animals, reduced milk yield, costs of treatment and labor, with culling of most of the cattle. The treated animals recovered within three to five weeks, with only a few affected eyes having a persistent white scar on the cornea as reported by other authors (Coetzer and Tustin, 2004). In this study, though similar findings were reported, none of the cases developed the severe forms of the disease. This may be attributed to the prompt treatment of most cases due the presence of veterinarians and experienced herdsmen who reported most cases early. Most cases of eye infections have been reported in younger cattle. In this study, most cases of eye infections were in yearling heifers which are in agreement with the previous reports on this condition (Champness, 2008). This may be due to the low immunity in younger cattle since they are naïve. The Staphylococcus aureus and Escherichia coli that are identified in this study were considered as commensal hence not cause of the eye infections observed. In conclusion, Moraxella bovis were the only infectious agents isolated from eye infections in these cattle therefore it was the cause of this disease and occurrence of face 47 flies was the main predisposing factor. The authors recommended prompt treatment of eyeinfections and control of flies particularly in the calf pens and milking parlour. ACKNOWLEDGEMENTS We wish to acknowledge and to thank the Veterinarians and laboratory staff at the Regional Veterinary Lab, Eldoret for use the lab facilities and assistance with this project. Gratitude also goes to the Management of lIlula farm, Eldoret for permission to carry out the research on their animals and the assistance of the animal health assistant on the farm. REFERENCES Blood D.C, Radostits O.M, Gay C.C and Hinchcliff K.W, (2000). Veterinary Medicine, A Textbook of Diseases of Cattle, Sheep, Pigs, Goats and Horses-9th edition. Harcourt Publishers Ltd. page 412-415. Champness D. (2008). Pinkeye in Beef cattle. ISSN 1329-8062 Published and Authorised by: Department of Environment and Primary Industries, 1 Spring Street Melbourne, Victoria. The original author of this agriculture note was Neil Farquhar, and the previous version was published in September 1998 Cheng TH. Frequency of pinkeye incidence in cattle in relation to face flies abundance. J Econ Entomol. 1967; 60:598–599. [PubMed] Coetzer J.A.W AND Tustin R.C. (2004). Infectious diseases of livestock 2nd edition Volume 3, Oxford University Press, Cape Town.1487-1490. Holt JG. Bergey's Manual of Systematic Bacteriology. 9th ed. Baltimore: Williams & Wilkins; 1994. pp. 296–302. Snowden, G.D., Van Vleck, L.D., Cundiff, L.V., and Bennett G.L. Genetic and environmental factors associated with incidence of infectious bovine keratoconjuctivitis in preweaned beef calves. Journal of Animal Science 2005 Mar,83(3): 507-518 48 1.5 Disorders of the claw and their association with laminitis in smallholder zero- grazed dairy cows *J. Nguhiu-Mwangi1, P.M.F. Mbithi2, J.K. Wabacha3 and P.G. Mbuthia4 1University of Nairobi, Faculty of Veterinary Medicine, Department of clinical studies, P.O. Box 29053-00625, Nairobi; 2 Office of the Vice Chancellor, University of Nairobi, P.O. Box 30197-00100, Nairobi; 3African Union-IBAR-Nairobi, P.O. Box 30786-00100, Nairobi; 4University of Nairobi, Faculty of Veterinary Medicine, Department of Veterinary Pathology, Microbiology and Parasitology, P.O. Box 29053-00625, Nairobi. *Corresponding author: email: jamesnguhiumwangi@gmail.com Abstract Lameness in dairy cattle adversely impact on productivity. A cross-sectional study was done in which 300 zero-grazed dairy cows from 29 smallholder dairy units were examined for claw lesions in the peri-urban areas of Nairobi, Kenya. The objective of the study was to determine the characteristics of claw disorders and their association with laminitis. Lameness for each cow was determined by assessing the locomotion score through a prescribed locomotion scoring system. Each cow was restrained in a crush and the claws thoroughly washed and examined for any claw external lesions or disorders including signs of laminitis. About 2 mm of the horn of the sole for each claw was trimmed to confirm laminitis by presence of sole haemorrhages. Data was analyzed using Genstart statistic software. Prevalence of laminitis was 70.3% out of which 49.3% was subclinical laminitis and 21.0% was chronic laminitis. Claw deformities found to be strongly associated with chronic laminitis were claw overgrowth (χ2 = 96.69, r = 0.6, p < 0.0001), horizontal grooves (χ2 = 61.27,r = 0.5, p < 0.0001), concave claws (χ2 = 59.39, r = 0.4, p < 0.0001), flat claws (χ2 = 57.87, r = 0.5, p < 0.0001), presence of sole haemorrhages (χ2 = 50.16, r = 0.4, p < 0.0001) double soles (χ2 = 42.57, r = 0.4, p < 0.0001) and white line separation (χ2=37.78, r=0.4, p<0.0001), while sole bruising was moderately associated with chronic laminitis (χ2 =11.02, r = 0.2, p = 0.0009). Sole haemorrhages were the only lesions strongly (χ2=89.45, r=0.6, p<0.0001) associated with subclinical laminitis. It was therefore concluded that chronic laminitis was strongly associated with deformity types of claw disorders and diffuse sole haemorrhages, while subclinical laminitis was strongly associated with only sole haemorrhages. Key words: Lameness, sole haemorrhages, claw deformities, subclinical laminitis, chronic laminitis INTRODUCTION The physiological function of the bovine claw and the pathogenesis of non-infectious claw disorders depend on the state of pododermal microvasulature (Hisrschberg and Plendl, 2005). Biologically, laminitis is a systemic disease syndrome that primarily manifests in the claws, hence rightly being referred to as “claw horn disruption” (CHD), which is a term that more appropriately describes its main effects on the claws (Hoblet 49 and Weiss, 2001). Impairment of the corium microvasculature resulting from altered systemic pH, as well as the action of circulating endotoxins and vasoactive substances triggers processes that lead to laminitis (Nocek, 1997; Cook et al., 2004). Laminitis initially develops as insidious subclinical form with delayed manifestation of the clinical symptoms. By the time it gets to the phase of clinical manifestation, the disruptive changes within the claw may already be irreversible (Greenough, 1987; Nocek, 1997; Belge and Bakir, 2005; Nguhiu-Mwangi, 2007; Nguhiu-Mwangi et al., 2008). Apart from causing lameness directly in cattle, laminitis through claw-horn disruption (Hoblet and Weiss, 2001) is also incriminated as a significant contributor to the occurrence of other non-infectious claw disorders such as white line separation, sole ulcers, heel erosion, sole bruising, double soles and sole haemorrhages (Smilie et al., 1991; Nocek, 1997; Belge and Bakir, 2005). Laminitis may not drastically reduce productivity in dairy cows. However, due to its effects on the quality of the horn of the claws that predispose to the occurrence of other claw disorders and the irreversible damage within the claws, it may eventually have long-term effects on the performance of the dairy cow that may subsequently result in the culling of the animal (Nocek, 1997; Belge and Bakir, 2005). Chronic laminitis in cattle causes gross misshaping of the claw that can easily be mistaken as mere hoof deformity (Rhebun and Pearson, 1982; Greenough, 1987). The claw is becomes more elongated, flattened, widened and the dorsal wall becomes concave (Greenough, 1987; Nocek, 1997), while horizontal ridges on the dorsal surface of the claw become deeper and more prominent, giving it a rippled appearance (Weaver, 1993; Nocek, 1997). Some of the lesions associated with laminitis can be evident externally, but others are more discernible only after trimming the horn over the sole and the white line (Nocek, 1997; Belge and Bakir, 2005). It has been stated that laminitis could be the cause or result of some of the non- infectious claw disorders affecting dairy cows (Smilie et al., 1991; Nocek, 1997; Belge and Bakir, 2005), but its association with these claw disorders has not been presented in any literature. The purpose of this paper is therefore to show how subclinical and chronic phases of laminitis are associated with other non-infectious claw disorders in dairy cows. MATERIALS AND METHODS The study Area The study was carried out in the urban and peri-urban areas of Nairobi County, Kenya. Nairobi is the capital city of Kenya with an area estimated at 696 square kilometers and a populated of over 3.1 million people. It lies between 01º18´S and 36º45´E. It is 1798 meters above the sea level and has an annual rainfall estimated at maximum of 765 mm and minimum of 36 mm with two distinct seasons of March to June and October to December. Nairobi and its surrounds have a high concentration of smallholder zero- grazed dairy units due to its ready market for milk and related milk products. 50 Selection of Study Smallholder dairy Units A purposive selection of 32 smallholder zero-grazing dairy units was done. Each had 5- 20 adult dairy cows. The median number of cows to the units was 10. Purposive selection method was used because of the difficulties of farmers allowing their animals to be used for the study, hence it was based on the willingness of farmers to participate and to allow their cows to be used. The farms were selected with the assistance of local veterinarians and veterinary paraprofessional workers with whom the farmers were acquainted. Selection of the cows A total of 300 dairy cows were selected for the study from the 32 smallholder zero- grazing dairy units. A cow qualified to be recruited into the study population if she had calved at least once. Among the cows that were selected, 40% were in the 1st or 2nd parities and 60% had 3 or more parities. Lame and non-lame cows were included in study. They included the four main dairy breeds, which were Friesian, Ayrshire, Guernsey and Jersey or their cross breeds. The cows were selected using simple systematic sampling method in which the cows that met the selection criteria in each smallholder zero-grazing unit were serially tagged with numbers 1, 2, 3,-- n, where n was the serial number of the last cow with selection criteria in the unit. The assigning of the serial numbers to the cows was randomly done by a farm worker who was not aware of the research objectives in order to avoid being biased. The investigator then did the systematic sampling of the serially numbered cows in each unit by selecting every alternate cow within the serial numbers. For example cow numbers 1, 3, 5, 7-- n or 2, 4, 6, 8-- n until the last cow within the smallholder zero-grazing unit. Therefore the cows within a unit that met selection criteria were either all picked from only even numbers or only odd numbers. For any smallholder zero-grazing unit that had less or up to a total of 5 eligible cows, all of them were included in the study, but systematic sampling was done only for units with more than 5 eligible cows. Examination of the Selected Cows Each of the selected cows was assessed for lameness using a locomotion scoring system adapted from Sprecher et al. (1997). This was followed by each cow being restrained in a crush for examination of claw disorders. In most of the smallholder zero-grazing units, the restraint facilities were poor and therefore only the hind limbs were examined because of the difficulties of accessing the forelimbs while the cows were within these facilities. Moreover, it is clearly known that a higher percentage of claw disorders in cattle affect mainly the hind limbs (Tranter et al., 1993), hence examining the hind limb would be a reliable indicator of the claw disorder status. With the cow well restrained in a crush, each hind limb claw was evaluated on the dorsal surface for signs of laminitis and other disorders. Each limb was raised off the 51 ground by fastening it with a rope tied to an overhead cross-bar, thoroughly washed and the sole examined for any gross lesions. A 1-2 mm thickness of the horn of the sole was trimmed-off using a quittor knife in order to expose any underlying lesions. Trimming was non-invasive because it did not reach the level of the corium, hence no pain was caused to the cows. Foot anaesthesia using 2% lignocaine hydrochloride was applied when examining any cow with painful claw lesions. Each claw lesion was pre-coded with a specific code number and recorded in designated data entry forms using a simple numerical system. Numerical “1” was used for presence of subclinical, chronic laminitis or a claw disorder and numerical “0” for absence of claw disorders (i.e. when the claw was normal). The “1” or “0” entries made it easier and convenient to important the data into computer analysis software. Diagnosis of laminitis was mainly based on presence of haemorrhages occurring on the weight-bearing surfaces of the claws as red or yellow discolourations. The haemorrhages would either be on one or more zones of the weight-bearing surface and would either occur alone or concurrently with other non-infectious claw lesions such as sole bruising, sole ulcer, heel erosion, white line separation and double soles (Nocek, 1997; Belge and Bakir, 2005). Normal-looking claws but with sole haemorrhages were diagnosed as having subclinical laminitis, but those with sole haemorrhages concurrently with other claw disorders/deformities were diagnosed as having chronic laminitis. Those with no sole haemorrhages were diagnosed as having only the manifesting disorder. Location of each lesion on the weight-bearing surface of the claws was classified according to the 6 conventional zones as described by Greenough and Vermunt (1991) as shown in Fig. 1. Fig. 1: Zones of the weight-bearing surface of the claws of cattle (Greenough and Vermunt, 1991). Key to Figure 1: Zone 1-white zone at the toe, Zone 2 - abaxial white zone, Zone 3 - abaxial wall-bulb junction, Zone 4 – sole-bulb junction, Zone 5 – apex of the sole, Zone 6 – heel bulb. 52 The distribution of haemorrhages on the weight-bearing surfaces of the claws was also categorized according to 5 conventional haemorrhage scores also described by Greenough and Vermunt (1991), which is presented in Table 1. Table 1: the 5-score scale of sole haemorrhages that occure on the weight-bearing surface of the claws of cattle (Greenough and Vermunt, 1991) Score Description 0 No haemorrhages 1 Slight discolouration 2 Moderate discolouration 3 Severe haemorrhages 4 Exposed corium DATA MANAGEMENT AND ANALYSIS The data were stored in Microsoft Office Excel 2003 (Microsoft Corporation, 2003) spread sheets. They were cleaned, verified and validated to be correct as per the entries from the record sheets. The data were imported into SAS© 2002-2003 (SAS Institute Inc. Cary, NC, USA). Descriptive statistics were computed for each claw disorder per cow and then for the entire study population. The prevalence rate of each claw disorder was calculated independent of other claw disorders. It was calculated as the number of cows (CL) affected by the specific claw disorder divided by the total number of cows (300) examined multiplied by 100 to give the percentage (%). Prevalence (%) = CL x 100 300 After descriptive statistics, the test of association between laminitis and the non- infectious claw disorders and the manifesting symptoms was done in three stages. The first stage was Chi-square (χ2) statistics to test for unconditional associations as well as strength of those associations between laminitis and the other claw disorders at the 5% significant level (p < 0.05) and also the correlation coefficient (r) for direction of the associations. The second stage was multiple logistic regression models that factored all the claw disorders against subclinical and chronic phases of laminitis. These also revealed both the strength and direction of the associations using β-estimate (β-e) and odds ratio (O.R.). The third stage was stepwise logistic regression models, which revealed the claw disorders that were invariably associated with the two phases of laminitis being considered. The effects of confounding among the claw disorders in contributing to the associations were dealt with in the analysis through the stepwise selection of the factored claw disorders. 53 RESULTS The results of this study revealed that the dairy cows in the smallholder zero-grazing units in the urban and peri-urban areas of Nairobi had a 70.3% prevalence of laminitis of which 49.3% was subclinical and 21% was chronic. There was also relatively high percentages of occurrence of claw disorders other than laminitis, such as haemorrhages (52%), sole bruising (45%), overgrown claws (30.3%), heel bruising (27%), white line separation (18%), horizontal ridges (17.8%), double (underrun) soles (17%) and flattened claw (12%). Claw infections had low percentage (4.7%). The percentages of occurrence of these claw disorders are summarized in Table 2. Table 2: Prevalence of laminitis and the non-infectious claw disorders that were found during prospective cross-sectional study in dairy cows in 32 smallholder zero-grazing units in urban and peri-urban areas of Nairobi, Kenya. Claw disorders Number of cows (n = 300) Prevalence (%) Subclinical laminitis 148 49.3 Chronic laminitis 63 21.0 Sole haemorrhages 156 52.0 Sole bruising 135 45.0 Overgrown claws 91 30.3 Heel erosion/bruising 82 27.3 White line separation 54 18.0 Horizontal ridges on dorsal wall surface 53 17.8 Double (underrun) soles 51 17.0 Flattened claws 36 12.0 Concaved claws 15 5.0 Corkscrew claws 14 4.7 Sole ulcer 8 2.7 NB: Some of the cows had more than one claw disorder per claw, hence the reason for total number adding to more than 300 and prevalence of all claw disorders adding to more than 100%. Correlation coefficient (r) and chi-square (χ2) statistics indicated that subclinical laminitis has significant positive and strong association with sole haemorrhages (r = 0.5, χ2 = 89.45m p < 0.0001). The haemorrhages associated with subclinical laminitis were 54 invariably observed in zone 4 of the weight-bearing surface of the claw and this was significant (r = 0.3, χ2 = 22.83, p < 0.001) and occasionally but significantly in zone 6 (r = 0.1, χ2 = 5.87, p < 0.0154) (Table 3). Table 3: Correlation between subclinical laminitis and non-infectious claw disorders in dairy cows examined during the prospective cross-sectional study in 32 smallholder zero-grazing units in urban and peri-urban areas of Nairobi, Kenya. Claw disorders Number of cows With each claw disorder (n =300) χ2 r-value p-value Direction of association Sole haemorrhages 156 89.44 0.6 < 0.0001 Positive Sole haemorrhages in zone 4 104 22.84 0.3 < 0.0001 Positive Sole haemorrhages in zone 6 42 5.87 0.1 0.0154 Positive Overgrown claws 91 15.94 -0.2 < 0.0001 Negative Concaved claws 15 15.37 -0.2 < 0.0001 Negative Flattened claws 36 14.62 -0.2 < 0.0001 Negative Corkscrew claws 14 14.30 -0.2 0.0002 Negative Horizontal grooves 53 6.09 -0.1 0.0136 Negative NB: Some of the cows had more than one claw disorder per claw, hence the reason for total number adding to more than 300. The degree of haemorrhages associated with subclinical laminitis was slight to moderate (score 1 and 2) as described in Table 1. However, through multiple logistic regression and stepwise logistic regression models, it was found that the only claw disorders having significant strong positive association with subclinical laminitis were the occurrence of sole haemorrhages (β-e = 3.672, O.R. = 39.33, p = 0.0072), particularly in zone 4 (β-e = 1.399, O.R. = 4.05, p < 0.0001) as shown in Tables 4 The claw disorders that had weak positive significant association with subclinical laminitis were sole bruising (β-e = 0.821, O.R. = 2.27, p = 0.0093) and heel bruising (β-e = 0.820, O.R. = 2.27, p < 0.0139) (Table 4). The rest of the claw disorders observed were negatively associated with subclinical laminitis as shown by correlation coefficient and β-estimate values (Tables 3 and 4). 55 Table 4: Results of multiple logistic regression models on associations between subclinical laminitis and non-infectious claw disorders in dairy cows examined in a prospective cross-sectional study in 32 smallholder units in urban and peri-urban areas of Nairobi, Kenya. Claw disorders β-estimate (β-e) S.E (β) Odds ratio (O.R.) 95% C.I. χ2 p-value Sole haemorrhages 3.672 1.367 39.33 2.70-573.14 7.22 0.0072 Sole ulcer -2.213 1.105 0.11 0.01-0.95 4.02 0.0451 Horizontal grooves -1.559 0.469 0.21 0.08-0.53 11.07 0.0009 Sole bruising 0.821 0.315 2.27 1.22-4.22 6.77 0.0093 Heel bruising 0.820 0.334 2.27 1.18-4.37 6.05 0.0139 NB: S.E (β) = standard error of β, CI = confidence interval Overgrown claws, horizontal grooves on the dorsal surfaces of the claws, concave shape of the dorsal wall of the claw, flattened claws, corkscrew claws, double (underrun) soles, white line separation, sole bruising and presence of haemorrhages on the weight- bearing surfaces of the claws were all found to have strong positive significant association with chronic laminitis (Table 5). Unlike in subclinical laminitis, the sole haemorrhages associated with chronic laminitis were found to be severe in scores 2 and 3, which was significant (χ2 = 50.16, p < 0.0001). These haemorrhages were widely distributed in most zones on the weight-bearing surface of the claws and have varied strength of associations with chronic laminitis as follows: zone 2 (χ2 = 18.24, p < 0.0001), zone 3 (χ2 = 23.45, p < 0.0001), zone 4 (χ2 = 11.05, p = 0.0009), zone 5 (χ2 = 5.23, p = 0.0222) and zone 6 (χ2 = 4.48, p = 0.0304) as shown in Table 6. Sole ulcer was weakly but significantly associated (r = 0.118, χ2 = 4.16, p = 0.0412) with chronic laminitis. Although multiple regression models showed that all these claw disorders were significantly associated with chronic laminitis (Table 7), stepwise logistic regression models indicated the only claw disorders that were invariably significant with chronic laminitis were overgrown claws (β-e = 3.006, O.R. = 10.93, p < 0.0001) and deep horizontal grooves on the dorsal wall of the claws (β-e = 2.196, O.R. = 12.30, p < 0.0001). 56 Table 5: Correlation between chronic laminitis and non-infectious claw disorders in dairy cows in a prospective cross-sectional study in 32 smallholder zero-grazing units in urban and peri-urban areas of Nairobi, Kenya. Claw disorders Number of cows with each type of claw disorder χ2 r-value p-value Direction of association Overgrown claws 91 96.69 0.6 < 0.0001 Positive Horizontal grooves 53 61.27 0.5 < 0.0001 Positive Concaved claws 15 59.39 0.4 < 0.0001 Positive Flattened claws 36 57.87 0.4 < 0.0001 Positive Corkscrew claws 14 45.71 0.4 < 0.0001 Positive Double (underrun) soles 51 42.57 0.4 < 0.0001 Positive White line separation 54 37.78 0.4 < 0.0001 Positive Sole bruising 135 11.02 0.2 0.0009 Positive Sole ulcers 8 4.16 0.1 0.0412 Positive Sole haemorrhages 156 50.16 0.4 < 0.0001 Positive Haemorrhages zone 2 38 18.23 0.2 < 0.0001 Positive Haemorrhages zone 3 43 23.45 0.3 < 0.0001 Positive Haemorrhages zone 4 104 11.05 0.2 0.0009 Positive Haemorrhages zone 5 26 5.23 0.1 0.0222 positive Haemorrhages zone 6 42 4.48 0.1 0.0343 positive 57 Table 6: Results of multiple logistic regression models on associations between chronic laminitis and non-infectious claw disorders in dairy cows examined in a prospective cross-sectional study in 32 smallholder zero-grazing urban and peri-urban areas of Nairobi, Kenya Claw disorders β-estimate (β-e) S.E. (β) Odds ratio (O.R.) 95% CI χ2 p-value Sole haemorrhages 4.397 4.056 0.3 0.03-2.45 39.92 < 0.0001 Haemorrhages zone 2 6.040 4.257 13.7 1.11-167.30 19.60 < 0.0001 Haemorrhages zone 3 6.132 4.178 24.8 1.89-327.21 23.63 < 0.0001 Haemorrhages zone 4 61.630 4.194 4.0 0.42-38.45 11.74 < 0.0001 Haemorrhages zone 5 4.990 4.087 7.06 0.60-83.35 5.99 0.0144 Haemorrhages zone 6 7.227 4.253 17.70 1.49-210.79 4.54 0.0332 Overgrown claws 2.402 0.817 11.04 2.23-54.74 93.55 < 0.0001 Flattened claws 3.543 1.037 34.56 4.53-263.74 93.55 < 0.0001 Horizontal grooves 2.553 0.705 12.84 3.23-51.07 60.62 < 0.0001 Sole ulcer 1.573 1.523 4.82 0.24-95.45 4.19 0.0406 Sole bruising 1.840 0.821 6.30 1.26-31.47 10.24 0.0014 White line separation 2.352 0.828 10.50 2.07-53.26 35.08 < 0.0001 Double (underrun) soles 3.035 0.856 20.80 3.88-111.33 39.70 < 0.0001 DISCUSSION The high prevalence of laminitis and non-infectious claw disorders has been attributed to presence of cow-level and farm-level risk factors prevailing within the smallholder dairy systems in the zero-grazing units (Nguhiu-Mwangi et al., 2008). The low plane concentrate feeding in these smallholder dairy systems only predisposes to the occurrences of subclinical and chronic laminitis rather than acute laminitis, which is normally associated with high level of grain feeding. The results of this study indicate that all cows with subclinical and chronic laminitis invariably had sole haemorrhages as previously documented (Nocek, 1997; Belge and Bakir, 2005). However, the severity of haemorrhages and the zones on the weight- bearing surface of the claw in which the haemorrhages were found varies between subclinical and chronic laminitis. Strong positive association between sole haemorrhages and laminitis could be attributed to changes in pododermal microvasculature that occurs in laminitis, which results in transvascular seepage of serum that manifests as haemorrhagic discolourations in the horn of the sole (Nocek, 58 1997; Hirschberg and Plendl, 2005). It has further been reported that a number of claw- horn lesions initially develop as haemorrhages of the sole or white line (Vermunt and Greenough, 1991). The invariable association of haemorrhages in zones 4 and 6 with subclinical laminitis could be explained by the fact that the horn in these zones is thinner than the rest of the zones, hence the haemorrhages that occur in the corium would show on the surfaces of these zones and then later in chronic laminitis appear on the surfaces of the zones with thick horn (Nguhiu-Mwangi et al., 2007). Furthermore, the initial pathogenesis of subclinical laminitis involves the process of extravascular escape of blood components into the tubules of the horn of the sole and the heel bulb (Greenough and Vermunt, 1991), with haemorrhages gradually becoming visible on the surface of the sole over a period of time, hence sole haemorrhages are a good indicator of subclinical laminitis (Shearer and van Amstel, 2000; Nguhiu-Mwangi et al., 2007). This explains the fact that by the time laminitis advances to chronic phase, the haemorrhages are widely spread to most zones of the weight-bearing surface of the claw. Moreover, the deformities that occur in chronic laminitis result in shifting of weight-bearing to the sole-heel junction area which is weaker. This exacerbates microvascular damage in the corium, leading to more severe and widespread sole haemorrhages in chronic laminitis than in subclinical laminitis (Greenough and Vermunt, 1991; Nocek, 1997). Association of sole bruising and heel erosion with subclinical laminitis may result as a cascade of processes. In these processes, subclinical laminitis causes production of weak horn that is easily bruised/eroded at the sole and the heel. Bruised and eroded horn of the sole and heel become excessively thin, allow easy transmission of pressure to the corium from the treading ground against the weight of the cow particularly in zones 4 and 6 (Shearer and van Amstel, 2000). This causes microvascular damage within the corium, which predisposes to occurrence of subclinical laminitis, subsequently leading to production of weak horn. This cascade cycle continues. Claw deformity disorders shown to have positive strong association with chronic laminitis can probably be attributed to the disruption of the growth of keratinized horn during laminitis, which subsequently alters the shape of claws, thus making them longer, flattened and broadened (Nocek, 1997). The dorsal angel becomes markedly reduced, while the dorsal wall becomes shaped to concave (Weaver, 1993). It is this misshaping of the claws that makes the chronic laminitis be mistaken as claw deformity (Rhebun and Pearson, 1982; Greenough, 1987) with the affected cow developing abnormal gait and lameness. Presence of any of these claw shapes concurrently with characteristic rippled appearance of the dorsal surface due to prominent horizontal grooves (Weaver, 1993; Nocek, 1997), could positively indicate presence of chronic laminitis. Moreover, laminitis is incriminated as a predisposing cause for development of claw deformities (Greenough, 1987; Weaver, 1993). Conversely, the resulting misshaping of the claws enhances more laminitis by altering the altering the treading 59 angle with more weight-bearing on the softer parts of the sole towards the heel, which in turn leads to transmission of traumatic pressure into the underlying corium. The association of some claw conditions such as double (underrun) soles, white line separation, sole bruising, white line separation, sole ulcer and heel erosion with chronic laminitis is probably related to the fact that chronic laminitis is reported to be a predisposing cause to these conditions (Nocek, 1997; Smilie et al., 1999; Belge and Bakir, 2005). Severe microvascular haemorrhages in the corium leads to accumulation of blood/serum between the corium and the horn of the sole. This eventually causes separation of the horn from the underlying corium. The new horn produced by the corium is separate from the old horn and this results in double soles with outer old horn and inner thin new horn (Nocek, 1997). The formation of double soles is a process that occurs in chronic laminitis and not in subclinical laminitis, which explains its strong association with chronic laminitis. Similarly, the white line zone is naturally the weakest point of impact on the weight- bearing surface of the claws and easily succumbs to avulsion under pressure of the weight of the animal against the treading ground (Baggott and Russell, 1981). During laminitis, the white line zone becomes even more vulnerable to avulsions owing to softening and impairment of the horn, which is more prevalent in chronic laminitis (Nocek, 1997; Smilie et al., 1999; Belge and Bakir, 2005). The cows found with sole ulcers in this study had very advanced laminitis. It has been reported that in chronic laminitis, the pedal bone changes its configuration and drops distally, compressing the corium towards the horn of the sole. The horn then gives way at the axial part of zone 4 and 5, hence resulting in sole ulcer that exposes the corium to protrude externally, which explains how sole ulcer is associated with chronic laminitis (Baggott and Russell, 1981; Hull, 1993; Nocek, 1997). It is therefore concluded from the findings of this study that non-infectious claw disorders in cattle are strongly associated with laminitis. The subclinical phase of laminitis being strongly and invariably associated with sole haemorrhages particularly in zones 4 and 6 of the weight-bearing surface of the claw, while chronic laminitis being strongly associated with deformities that cause disorders of the claws. However, laminitis can be the “cause” or the “effect” of non-infectious claw disorders. The main risk factors of laminitis and the non-infectious claw disorders in the dairy cows in the smallhoder zero-grazing units are comprehensively described in a previous publication (Nguhiu-Mwangi et al., 2008). ACKNOWLEDGEMENTS The authors are grateful to the Deans Committee of the University of Nairobi for financial support to carry out this study. We also appreciate the farmers who opened their smallholder zero-grazing units for data collection. 60 REFERENCES Baggott, D. G and Russell A. M (1981): Lameness in cattle. British Veterinary Journal 137: 113- 132. Belge, A and Bakir, B: (2005): Subclinical laminitis in dairy cattle: 205 selected cases. Turkish Journal of Veterinary Science 29: 9-15. Cook, N. B; Nordlund, K.V and Oetzel, G.R (2004): Environmental influences on claw horn lesions associated with laminitis and subacute ruminal acidosis in dairy cows. Journal of Dairy Science 87: E36-E46. Greenough, P.R (1987): An illustrated compendium of bovine lameness. Part I. Modern Veterinary Practice 68:6-9. Greenough, P.R and Vermunt, J. J (1991): Evaluation of subclinical laminitis in a dairy herd and observations on associated nutritionaland management factors. Veterinary Record 128: 11-17. Hirschberg, R. M and Plendl, J (2005): Pododermal angiogenesis and angioadptation in the bovine claw. Microscopy Research and Technique 66: 145-155 Hoblet, K. H and Weiss, W (2001): Metabolic hoof horn disease. Claw horn disruption. Veterinary Clinics of North America. Food Animal Practice 17:111-127. Hull, B. L (1993): Sole abscesses, vertical wall cracks, horizontal wall cracks, Rusterholz ulcer. In: Current Veterinary Therapy 3. Food Animal Practice. Howard J. L (ed.). W. B. Saunders, Philadelphia. pp 864-866. Nguhiu-Mwangi, J (2007): Characteristics of laminitis and associated claw lesions in dairy cows in Nairobi and its Environs. PhD Thesis. University of Nairobi. Nguhiu-Mwangi, J; Mbithi, P.M.F; Wabacha, J. K and Mbuthia, P.G (2007): Prevalence of sole haemorrhages and its correlation with subclinical and chronic laminitis in dairy cows. Bulletin of Animal Health and Production in Africa 55: 232-242. Nguhiu-Mwangi, J; Mbithi, P.M.F; Wabacha, J. K and Mbuthia, P.G (2008): Factors associated with the occurrence of claw disorders in dairy cows under smallholder production systems in urban and peri-urban areas of Nairobi, Kenya. Veterinariski Arhiv 78: 343-355. Nocek, J. E (1997): Bovine acidosis: implications on laminitis. Journal of Dairy Science 80:1005-1028. Rhebun, W. C and Pearson, E. G (1982): Clinical management of bovine foot problems. Journal of the American Veterinary Medical Association 1982: 181: 572-579. Shearer, J.K and van Amstel, S (2000): Lameness in dairy cattle. In: Proceedings of 2000 Kentucky Dairy Conference. Lexington. pp1-11. http://www.uky.edu/Ag/AnimalSciences/dairy/dairyconference/dc07 Smilie, R.H; Hoblet, K. H; Eastridge, W. P; Weiss, W. P; Schnitkey, G. L and Moeschberger, M. L (1999): Subclinical laminitis in dairy cows: use of severity of hoof lesions to rank and evaluate herds. Veterinary Record 144: 17-21. Sprecher, D. J; Hostetler, J. B and Kaneene, J. B (1997): A lameness scoring system that uses posture and gait to predict dairy cattle reproductive performance. Theriogenology 47: 1179-1187 61 Tranter, W.P; Morris, R.S; Dohoo, I.R and Williamson, N.B (1993): A case-control study of lameness in dairy cows. Preventive Veterinary Medicine 15: 191-203. Vermunt, J.J and Greenough, P. R (1996): Sole haemorrhages in dairy heifers managed under different underfoot and environmental conditions. British Veterinary Journal 152: 57-73. Weaver, A. D (1993): Aseptic laminitis of cattle, interdigital and digital laminitis, interdigital phlegmon (interdigital necrobacillosis). In: Current Veterinary Therapy 3. Food Animal Practice. Howard, J. L. (ed.). W.B. Saunders. Philadelphia. pp 867-870. 1.6 Participatory technology transfer and impact in poultry production and marketing: The case of farmers in Makueni County Kabuage, L.W.1*, Kamau, C.1, Kuria, J.2, Okitoi, L.3, Oyoko, G.4, Njuguna, K.5, Waithaka, K.1, 1Department of Animal Sciences, Kenyatta University, P.O. Box 43844-00100, Nairobi, Kenya 2Department of Veterinary Pathology and Microbiology, University of Nairobi, Kenya 3Kenya Agricultural and Livestock Research Organization (KALRO), Kakamega, Kenya 4Blake and Morton Agribusiness Consultants, Nairobi, Kenya 5State Department of Livestock, Nairobi, Kenya * Corresponding Author: lwkabuage@gmail.com Abstract Indigenous chickens (IC) constitute 80% of the poultry population in Kenya and kept by over 80% of rural households. They contribute to household food security, income and social-cultural roles. Demand for IC and their products has increased in urban areas without corresponding increase in productivity at farm level. A research project was undertaken in Makueni County where constraints to IC production and marketing were identified, followed by implementation of appropriate interventions. The intervention strategy entailed participatory technology development and transfer with farmers. A survey was then conducted in the study area post-project, aimed to evaluate the uptake and impact of the interventions. Primary data was collected from a sample of 40 farmers using a structured questionnaire and direct observations. Data was analysed using descriptive statistics. Results of the survey showed positive change between the ante- and post-project periods. The flock sizes per household increased, with farmers keeping more of improved indigenous chicken (IIC) ecotypes than the local IC. The productive performance of IIC was higher compared to that of IC. The mean age at onset of lay was 28 months for IC hens and 23 months for IIC while the corresponding number of eggs laid per hen per year was 43 compared to 74. The marketing price of eggs was on average Ksh. 9 and 13 for IC and IIC, respectively. Owing to a faster growth rate, the IIC birds were marketed earlier than IC (21 versus 27 weeks of age) and fetched a higher price of Ksh. 531 compared to Ksh. 420 for IC. Hatchability and survivability rates for all birds increased to approximately 90 and 86%, respectively. Proper housing of birds and usage of brooders for chick rearing increased. Almost all farmers (97.5%) practiced 62 disease control measures while a similar percentage formulated their own feeds and supplemented their flocks twice per day post-project. The significant impact from the study highlighted the importance and effectiveness of participatory technology development and transfer methods when working with rural communities. Such methods provide for faster and more sustainable gains in technology adoption, improved farm productivity and incomes, hence improved livelihoods. Key words: Indigenous chicken, farmers, interventions, participation, impact INTRODUCTION The poultry industry is estimated to contribute 33% of the total global meat consumption and this proportion is anticipated to increase at 2 -3% per year (Mengesha, 2013; FAO, 2012). Poultry production remains a vital economic activity among smallholder farmers in developing countries (Vincent et al., 2010). According to Kryger et al., (2010), an estimated 80% of rural households in developing countries recognize poultry production as a major economic activity. Indigenous chicken (Gallus domesticus) constitute over 70% of the total chicken population (FAO, 2011). Moreover, 47% and 55% of eggs and poultry meat respectively, in Kenya are derived from indigenous chicken (King’ori et al., 2012). Therefore, indigenous chicken is a promising subsector in Kenya and especially to small-scale farmers whose production constitutes 80% of the national poultry production (RoK, 2012). Indigenous chicken production creates an avenue towards addressing food insecurity and this translates to improvement of livelihoods among many rural households (Magothe et al., 2012). This is also the case for Makueni County located in an arid and semi-arid region where IC is reported to positively contribute to the livelihoods of many households (Kabuage, 2010). Indigenous chicken distribution remains wide in both rural and peri-urban areas and plays a crucial role in food production and generation of income (Okello et al., 2010). The IC products (meat and eggs) serve as a cheap source of animal protein, while reliability to source petty cash from IC is highly recognized (Adomako et al., 2009). Rearing of IC is preferred because the birds are hardy and well adapted to harsh environments, survive well on low inputs and are efficient in feed conversion (King’ori et al., 2010). Moreover, their products (meat and eggs) are preferred for leanness and good taste. Previous studies reported that consumers’ preference for IC products has triggered increased demand (Bett et al., 2012; Ndenga et al., 2017). The demand has also been linked to population growth, urbanization, health concerns and increased per capita income which is projected to increase consumption by the year 2020 (USAID, 2010). According to WSPA (2012), there exists an unmet demand for indigenous chicken products (meat and eggs). Consumers are therefore willing to pay a premium price for IC products (Bett et al., 2012; Ndenga et al., 2017). Despite the importance of IC, the growth of this subsector is constrained by various factors which include; poor nutrition, poor housing, low genetic potential, feed shortages and diseases, leading to low productivity (RoK, 2012; King’ori et al., 2010). Outbreak of diseases such as; New Castle Disease, fowl typhoid and Gumboro in Kenya 63 accounts for approximately 50% of the total IC losses (Olwande et al., 2010). To address the stated constraints, various technological interventions at farm level would be required. These include: proper housing, feed supplementation, disease and pest control, and improved IC genotypes. Implementation of such technologies would be expected to improve IC productivity among smallholder farmers in rural households (RoK, 2012; Njuguna et al., 2017). Enhanced productivity would consequently boost financial returns and help to address the grinding poverty experienced by most rural households. As a pertinent strategy to increase IC productivity, various rural development stakeholders such as the Kenya Agricultural Productivity and Agribusiness Project (KAPAP) have made efforts to intervene through supporting extension and research projects in selected counties. The current study presents the results of intervention impact generated by a KAPAP supported research project that was implemented in Makueni County to promote indigenous chicken production and marketing. The project entitled ‘Promotion of indigenous chicken value chain in Makueni County, Kenya’ was undertaken in two phases. The first phase entailed participatory identification of constraints to IC production and marketing through a mapping survey conducted along the indigenous chicken value chain (ICVC). Phase two involved formulation and implementation of identified appropriate interventions at farm level. The intervention strategy for the project entailed participatory technology development and transfer with farmers in Makueni for one year. A post-project survey was thereafter conducted in the study area, aimed to evaluate the uptake and impact of the implemented intervention technologies in the project areas in Makueni County. This paper was therefore intended to show the results of the impact evaluation and to depict the situation in the ante- and post-project periods in the County. MATERIALS AND METHODS Study area Makueni County is located in the Southern part of Eastern Kenya. It lies between latitude 1°35′ South and Longitude 37°10′ East and 38°30′ West East (RoK, 2010). The county comprises of an area of 8008.8 Km2. Temperatures range between 12-28 ºC, with bimodal rainfall ranging from 150-650 mm per annum, typical of the arid and semi-arid lands (ASALs) in Kenya (RoK, 2010). Low rainfall and high temperatures in this county hinder crop production thus livestock production remains a priority. The county is located in areas considered to have favourable agro-ecological conditions for production of IC and is listed amongst leading areas in IC production in Kenya (KARI, 2011; MoLD, 2011). The County has high poverty levels but indigenous chicken are found in almost all households. The research project covered two Sub-counties but the main site for the current study was located in Wote Sub-county of Makueni. 64 Data collection and sampling procedure The intervention through dissemination and implementation of improved poultry production and marketing technologies by the KAPAP Project was done in the 2014- 2015 period. This followed a comprehensive mapping and identification of constraints and opportunities in poultry production and marketing, using participatory methodology. The intervention strategy entailed participatory technology development and transfer with farmers. A broad range of key management technologies was disseminated and implemented in over 200 target households. The technologies included: feed formulation, feeding and supplementation, proper housing, bio-security and disease control, use of artificial brooders and rearing of improved indigenous chicken (IIC) ecotypes. One month old IIC chicks were distributed by the project to the target households. The impact survey that forms the subject of this paper was conducted after the end of the project in 2015. A purposive sampling procedure was used to select a sample of 40 households. A structured questionnaire was administered to collect primary data. This was further strengthened by direct observation of farm facilities and records. Secondary data was obtained from the livestock office located in Wote, Makueni County. Data analysis was done through descriptive statistics. RESULTS AND DISCUSSION The following section presents the results of impact of the disseminated technologies on IC performance among the target farmer groups in Makueni County. Distribution of target farmers Figure 1 shows how farmers were distributed in the various study sites in Makueni County which included: Kamunywolo, Mbimbini and Kikumini locations. The evaluation involved 12 farmers from Kamunywolo, 22 from Mbimbini and 6 from Kikumini location. The 22 farmers from Mbimbini constituted the Uviluni common interest group (CIG) whereas the other 18 farmers belonged to the Mwang’a/Mang’auni CIG. Figure 1: Distribution of target farmers in three locations 0 10 20 30 40 50 60 F re q u en cy / p er ce n ta g e Farmers location Farmers Distribution Frequency Percentage 65 Household flock size and composition The results in Table 1 illustrate the diffusion of the improved indigenous chicken technologies in the selected locations. Before the project, the target farmers had a mean flock size of 23 birds, of which 19 were local indigenous chickens. The flock size post- project had increased to an average of 34 birds comprising 26 improved indigenous chicken (IIC) and a reduced number of local IC (LIC) at only 8 birds. The percentage of IIC had risen from the initial 17.39% of the flock to a high of 73.53%, indicating an improvement of 56.14%. The evaluation results therefore showed a major increase in overall household flock sizes and more remarkably, a different flock composition comprising more IIC and fewer local IC. The new composition was expected to significantly enhance household flock productivity due to the higher performance of IIC compared to that of local IC. It was also an indicator that majority of the farmers had focused their attention towards rearing the improved indigenous chicken by reducing the number of the local indigenous chicken in their flocks. To boost the small household flock sizes and productivity, the project had earlier intervened by giving the target farmers an average of 15 one month old IIC chicks. The evaluation results reflected the cumulative impact of interventions arising from the distributed chicks and their far better performance in terms of growth and reproduction, and also the improved productivity associated with implementation of the other disseminated poultry management technologies. Table 1: Flock size and composition before, during and after project Local and Improved Indigenous Chicken Mean Std. Err. Flock composition before, during and after project Local indigenous chicken before project 19.25 2.72 Improved indigenous chicken before project 3.65 0.91 Total number of improved chicks received 14.5 0.94 Post-project flock size and composition 34.1 3.48 Improved indigenous chicken 25.83 3.46 Local indigenous chicken 8.28 1.60 Std. Err. – Standard Error: Source; Own computation, 2016 Results demonstrating the performance of both local indigenous chicken and improved indigenous chicken ecotypes are presented in Table 2. The performance parameters used included: age at onset of lay, average number of eggs per clutch, number of clutches/hen/year, total number of eggs/hen/year, size of an egg, age at selling of birds, and the relevant prices . 66 Table 2: Comparison on performance of improved indigenous chicken (IIC) with local indigenous chicken (LIC) on-farm Parameter Improved Indigenous Chicken(IIC) Local Indigenous Chicken(LIC) Age at first egg lay (months) Number of clutches/hen/year Av. number of eggs/clutch Total number of eggs/hen/year Age at selling of birds (months) Avg, selling price of birds (Ksh) Price of an egg (Ksh) Size of an egg (Visual assessment, scalea 1-3) Mean 22.78 3.48 21.95 74.18 21.20 530.60 12.80 2.73 Std. Err. 0.52 0.18 1.15 2.99 1.24 31.39 0.84 0.11 Mean 27.78 3.08 13.33 43.35 27.03 421.40 9.10 1.60 Std. Err. 1.15 0.15 0.67 2.57 1.79 26.38 0.73 0.12 Source; Own computation, 2016. aScale 1-3, where: 1-2 = Small (45-50 grams); 2.1-3 = Medium (51-55 grams); 3.1-4 = Large (56-60 grams) The results in Table 2 demonstrated the superior performance of improved indigenous chicken compared to that of local indigenous chicken, thus indicative of the benefits accruing from rearing more of the IIC. The average age of hens at onset of lay for the IIC was 23 weeks whereas that of LIC was 28 weeks. The number of clutches/hen/year was similar for both breeds at an average of 3 clutches though the deviation showed that IIC had more than 3 clutches. The IIC hens rarely got broody, thus the prospects of using them to incubate eggs were rare as expressed by majority of the farmers. This led to prolonged laying cycles for IIC compared to the LIC. The average number of eggs/clutch for the IIC was 22 eggs while that of LIC was 13eggs. This clearly pointed at the positive impact of IIC through laying more eggs compared to the local indigenous chicken. The total number of eggs/hen/year for IIC was 74 eggs whereas that of the local indigenous chicken was only 43 eggs. The results further indicated that majority of the farmers sold their indigenous chicken to available markets in Makueni County. Most of the farmers sold their IIC at an average age of 21 weeks at a selling price of Ksh. 531. The local indigenous chicken were sold at an age of 27 weeks of age for an average of Ksh. 420. This clearly indicated that IIC reached the market age and size faster compared to that of the local indigenous chicken. The difference in prices revealed higher preference for the IIC compared to the LIC, demonstrating superior quality for these birds. 67 The average selling price of an IIC egg was Ksh. 13 whereas that of the LIC was Ksh. 9. This indicated the IIC eggs were considered superior by consumers, hence the higher price. Majority of the farmers also confirmed that most of the buyers preferred buying IIC eggs for hatching purposes. This further propelled the demand for these eggs leading to farmers increasing the prices. It was also notable that the average size of IIC eggs was larger at about 51-55g per egg while the LIC eggs were of small size, weighing about 45-50g. The larger eggs from IIC were therefore preferred for both consumption and hatching, with the latter aimed at yielding higher performing chicken ecotypes. Table 3 shows the hatchability and survivability of the indigenous chicken before and after the project. The farmers practiced natural hatching by setting the available eggs to broody hens, with the objective of hatching by 21 days. This simple and low cost method of getting chicks was considered as convenient hence the best by farmers, for expanding their flock sizes. Table 3: Hatchability and survivability of the indigenous chicken Parameter Before the project After the project Mean Std. Err. Mean Std. Err. Eggs set for hatching 10.78 0.58 13.25 0.72 Av. No. of chicks hatched/hen 7.95 0.53 11.98 0.62 Av. No. of chicks weaned/hen 4.40 0.52 10.35 0.69 Source; Own computation, 2016 Before the project, majority of the farmers set an average of 11 eggs for hatching by a broody IC hen. After the project, results showed that the average number of eggs set had increased to 13 eggs, mainly attributed to increased availability of eggs from IIC. The average number of chicks hatched per hen before the project was 8 whereas this number rose to 12 chicks by the end of the project (Table 3). The computed hatchability rate before the project was therefore 73.78% but this increased to 90.38% post-project. The hatchability rate was computed as: Average number of chicks hatched Number of eggs set for hatching ˣ 100 These results demonstrated a positive impact from the training of project farmers on selection of eggs for hatching and improvement of the hatching environment. Farmers had been trained to always consider the critical factors that influence hatchability, with emphasis on proper egg handling, optimal storage conditions, age and size of eggs, before setting for hatching purposes. The survivability rate was computed by considering the number of chicks that reached weaning stage (8weeks) out of the total number that hatched. The results well illustrated the situation before and after the project. Before the project, the survivability rate was 55.35%, but this increased to 86.43% post-project. The increment was attributed 68 to the training of farmers on vaccination, bio-security and other disease control measures, brooding technologies, protection of chicks against predation, flock separation by categories and proper feeding, among other management practices. This was a clear indicator that the chicks’ mortality rate had reduced by the end of the project. Mortality rates were 44.65% and 13.57% respectively, for the periods before and after the end of the project. Survivability rate was computed as: Average number ofchicks at weaning stage Average number of chicks hatched ˣ 100 Table 4 shows the condition of the chicken houses before and after the project. Table 4: Housing standards and management practices Management practice Before project After the project Proportion Std. Err Proportion Std. Err. Condition of chicken house Poor – 0.78 0.07 Poor – 0.00 0.00 Good – 0.22 0.07 Good – 1.00 0.00 Training on housing management No – 0.78 Yes – 0.22 0.07 0.07 No – 0.02 Yes – 0.98 0.03 0.03 Source; Own computation, 2016 The results revealed that only 22.5% of the project farmers kept their IC houses in good condition while 77.5% of the farmers had poor IC housing and poor conditions. However, by the end of the project, all the farmers had improved their IC houses, keeping them clean and in overall good condition. The improvements were attributed to the training of farmers by the project team on how to construct and manage their poultry housing facilities. The training was broad based and had also covered the relationships between the condition of the poultry house with biosecurity and disease control issues, and the mortality rate of birds. The poor condition of poultry houses before the project was attributed to lack of knowledge and training on housing management, and failure to understand the consequences of poor and dirty housing conditions. Upon the training, the post-project results demonstrated that all the farmers within the group had maintained good conditions of their IC housing facility and the largest proportion were cleaning their chicken houses daily. The results shown in Table 5 and Figure 2 below illustrated the frequency of cleaning the IC housing facility. Before the project, majority (57.5%) of the farmers never cleaned their housing facilities. A few (2.5%) farmers cleaned the facility once per week, 12.5% cleaned twice per week and only 7.5% cleaned on daily basis. Post-project results showed that majority (35%) of the farmers cleaned their chicken houses on daily basis, a major improvement compared to the pre-project period. A significant proportion (27.5%) of farmers cleaned twice a week, 22.5% three times a week while 10% cleaned 69 four times a week. Only a mere 5% of the target farmers cleaned the chicken house once a week. The notable commitment to clean their poultry houses on frequent basis post- project was attributed to the farmer training undertaken by the project researchers with emphasis on importance of bio-security in chicken management and housing. Maintenance of hygiene and sanitation in chicken housing was underscored, including the use of disinfectants. This was an essential husbandry practice in disease control for the birds and for production of bio-safe poultry products. Table 5: Frequency of cleaning the IC housing facility Frequency of cleaning IC housing per week Proportion Std. Err. Number of times cleaned housing facility before project 0 0.58 0.08 1 0.23 0.07 2 0.13 0.05 7 0.08 0.04 Number of times cleaned housing facility after project 1 0.05 0.04 2 0.28 0.07 3 0.23 0.07 4 0.10 0.05 7 0.35 0.08 Source; Own computation, 2016 Figure 2: Trends on frequency of cleaning the housing facility 0 10 20 30 40 50 60 70 never clean once twice thrice four times daily basis P er ce n ta g e Number of times per week Graph showing frequency of cleaning housing facility Before project After project 70 Table 6 and Figure 3 illustrate the use of brooders among the project farmers. The results demonstrated that before the project, majority (85%) of the farmers did not utilize brooders in rearing their chicks. However, 32.5% of the farmers post-project had one brooder each, 10% had four brooders and 2.5% had five brooders for rearing their chicks, depending on the number of chicks. Table 6: Diffusion of brooding technologies among the farmer groups Description Proportion Std. Err. Brooders before the project 0 0.85 0.06 1 0.1 0.05 4 0.03 0.03 5 0.03 0.03 Brooders after the project 0 0.53 0.08 1 0.33 0.08 2 0.03 0.03 4 0.10 0.05 5 0.03 0.03 Type of brooders None 0.55 0.08 Carton box brooders 0.23 0.07 Wire-mesh brooders 0.23 0.07 Source; Own computation, 2016 Figure 3: Use of brooders before and after the project 0 20 40 60 80 100 0 1 2 4 5 P er ce n ta g e Number of brooders Brooders use before and after the project Before project End of project 71 As an appropriate strategy, the project stakeholders disseminated 3 types of brooders to the respective farmer groups and trained on their use. The evaluation results from the table and chart above revealed that by the end of the project, 47.5% of farmers had adopted the brooding technologies. This was as a result of the training and demonstration of their importance. It was further noted that carton boxes and wire mesh were the most preferred types of materials for fabricating the brooders. Figure 4 shows the usage of feeders and drinkers by the farmers within the group. Before the project, a big majority (77.5%) of the farmers did not have feeders and drinkers. Out of the 22.5% who had, 10% had 2 feeders in their poultry housing facility. Similarly, out of the 22.5% who had drinkers, 12.5% had only one drinker. With the project intervention, farmers in these groups received one feeder and one drinker, each. They were also trained on how to maintain them through regular cleaning and how to use them depending on the flock ration. At the end of the project, the evaluation results revealed that 97.5% of the farmers had feeders and drinkers in their poultry houses, with majority having at least one feeder and one drinker. It is notable that out of those farmers with these facilities, a sizeable number (25%) had 4 feeders while 22.5% had 4 drinkers. This highlighted the positive impact achieved, in that farmers had perceived and responded positively towards increasing the number of these equipments, so as to enhance chicken productivity. Figure 4: Diffusion of appropriate feeders and drinkers among the group farmers 0 10 20 30 40 50 60 70 80 90 0 1 2 3 4 5 9 P er ce n ta g e Current number of feeders Diffusion of feeders Before project End of project 0 10 20 30 40 50 60 70 80 90 0 1 2 3 4 5 6 P er ce n ta g e Current number of drinkers Diffusion of drinkers Before project End of project 72 Table 7 shows that majority of the farmers did not vaccinate their indigenous chicken before the project period. Only 15% of the group farmers followed the vaccination schedule as recommended by service providers. However, through field demonstrations and workshops organized by the project stakeholders, farmers were trained on how to carry out the required disease control practices using prophylactic methods, particularly vaccinations. They were also trained on the various symptoms of common poultry diseases. By the end of the project, majority of the farmers reported they knew how to identify symptoms of diseases that were likely to cause mortality in chicken and knew the appropriate measures to take. As a result of the trainings, 97.5% of the group farmers indicated they had been vaccinating their flocks post-project. Table 7: Vaccination and disease control practices Variable Mean Std. Dev. Min Max Vaccination before the project 0.15 0.36 0 1 Vaccination after the project 0.98 0.22 0 1 Vaccinator before the project 0.23 0.58 0 2 Vaccinator after the project 1.30 0.52 0 2 Mortality before the project 5.13 2.40 0 10 Mortality after the project 1.85 1.94 0 7 Source; Own computation, 2016 Results from table 8 showed that before the alluded to project, only a very small percentage (2.5%) of the farmers had accessed training on feed formulation. They nevertheless acknowledged they had never put the skills into practice. They were incapacitated by lack of practice and requisite facilities for the exercise. However, all the group farmers were trained by the current project on how to carry out feed formulation, mixing, and supplementation, as a key intervention for improved nutrition and feeding of their chicken. Table 8: Feed formulation and supplementation technology Variable/Activity Mean Std. Dev. Min Max Training on feed formulation before project 0.03 0.16 0 1 Training on feed formulation during project 1 0 1 1 Ability to carry out feed mixing before project 0.03 0.16 0 1 Ability to carry out feed mixing after project 1 0 1 1 Practicing feed mixing before the project 0 0 0 0 Practicing feed mixing after the project 1.53 0.88 1 3 Awareness on feed mixture technology before project 0.025 0.1581 0 1 Awareness on feed mixture technology after project 1 0 1 1 73 Following the procedure and ingredients ratios 1 0 1 1 Source; Own computation, 2016 Several practical demonstrations by the current project stakeholders were conducted to ensure that farmers followed the correct procedure while formulating and mixing feeds. The farmers communally used fabricated feed mixers provided by the project to mix the supplementation diets and then shared out the feed among themselves. This strategy addressed the previously existing problem of farmers who had no knowledge and pertinent equipment to mix feeds. Conducting the exercise as a group further benefited the farmers through cost sharing among themselves when purchasing any required ingredients. Results in Table 9 revealed that before the project, farmers used to supplement their IC only once a day. They used single ingredients, mainly some grains or kitchen waste thrown at the birds. However, the project intervened and farmers were trained on formulation and mixing of balanced diets as indicated above, to supplement their birds. They mainly reared their IC through scavenging which could not adequately provide the required nutrients for high productivity without supplementation. At the end of the project, farmers in these groups were supplementing their flock twice per day with use of mixed, balanced rations. A large part of the rations comprised of grains mainly sorghum (for energy) and sunflower seed (for protein) from their farms. The farmers were trained on agronomic practices of growing these locally adapted crops by the project and had obtained good harvests for use as food and feed. This was therefore a good indicator that at the end of the project, farmers were in an enabling position to meet the nutritional requirements for their flocks, thereby increasing productivity. Table 9: Supplementation of the IC flock Variable Mean Std. Err. Min Max No. of times supplementing flock before project 0.53 0.55 0 2 No. of times supplementing flock after project 1.8 0.69 0 3 Form of supplements before the project 0.85 0.95 0 2 Form of supplements after the project 0.98 0.16 0 1 Source; Own computation, 2016 CONCLUSION AND POLICY IMPLICATIONS The current study evaluated participatory technology transfer and impact in poultry production and marketing. The significant impact from the study highlighted the importance and effectiveness of participatory technology development and transfer methods when working with rural communities. Such methods provide for faster and more sustainable gains in technology adoption, improved farm productivity and incomes, hence improved livelihoods. It is therefore recommend that policies should be formulated to strengthen participatory dissemination of improved poultry production technologies. Enhancing infrastructure and deploying more extension agents would be a significant step to facilitate effective training and extension services, and viable uptake 74 of agricultural and livestock technologies. Additionally, farmers should put more focus on collective action by forming more Common Interest Groups (CIGs) to facilitate effective engagement with stakeholders for production and marketing in the poultry industry. ACKNOWLEDGEMENT The authors acknowledge support from the World Bank through the Kenya Agricultural Productivity and Agribusiness Project (KAPAP) in collaboration with the Indigenous Chicken Value Chain (ICVC) project implemented in Makueni and Kakamega Counties, under Kenyatta University, Kenya. REFERENCES Adomako, K., Hagan, J.K. & Olympio, O.S. (2010). The productivity of local chicken in the Ashanti Region of Ghana. Ghanaian Journal of Animal Sciences, 5(2). Bett H. K., Bett R. C., Peters K. J., Kahi A. K. and Bokelmann W. (2012). Linking Utilisation and Conservation of Indigenous Chicken Genetic Resources to Value Chains. Journal of Animal Production Advances, 2(1), 33 - 51. Food and Agriculture Organization. (2011). State of food and agriculture report. Retrieved from http://ilriclippings.wordpress.com/category/poultry Food and Agriulture Organization. (2012). Small livestock, big impact. Retrieved from http://ilriclippings.wordpress.com/category/livestock- challenges/indigenousbreeds/ Kabuage, L.W. (2010). On farm feed formulation and feeding trials for indigenous poultry. Ministry of Livestock Development. Nairobi, Kenya: Agricultural Sector Programme Support (ASPS) Report. Kingori, A., Wachira, A. & Tuitoek, J. (2010). Indigenous Chicken Production in Kenya. International Journal of Poultry Science, 9(4), 309-316. Retrieved from http://www.pjbs.org/ijps/fin1669.pdf Kryger, K.N., Thomsen, K.A., Whyte, M.A.& Dissing, M. (n.d.). Smallholder Poultry Production – Livelihoods, Food Security and Socio-cultural Significance, Smallholder Poultry Production. FAO. Rome Italy. , 76. Magothe, T.M., Okeno, T.O., Muhuyi, W.B. & Kahi, A.K. (2012). Indigenous chicken production in Kenya: Indigenous chicken current status. World Poultry Science, 68. Mengesha, M. (2013). Biophysical and the socio-economics of chicken production. African Journal of Agricultural Research, 8(18), 1828 - 1836. Ndenga, C., Lucy W. Kabuage.& Eric K. Bett. (2017). Analysis of Consumer Preference in Product Attributes: A Case of Indigenous Chicken Eggs in Kenya. Journal of Economics and Sustainable Development, 8(2). Njuguna, C.N., Lucy W. Kabuage and Eric K. Bett. (2017). Determinants of adoption and intensity of use of brooding technology in Kenya: The case of indigenous 75 chicken farmers in Makueni and Kakamega Counties, Kenya. African Journal of Agricultural Economics and Rural Development, 5(2), 532 - 537. Okello, J.J., Gitonga, Z., Mutune, J., Okelo, R.M., Afande, M. & Rich, K.M. (2010). Value Chain Analysis of the Kenyan poultry Industry. Olwande, P.O., Ogara, W.O., Okuthe, S.O., Muchemi, G., Okoth, E., Odindo, M.O. and Adhiambo, R.F. (2010). Assessing the productivity of indigenous chickens in an extensive management system in Southern Nyanza, Kenya. Tropical Animal Health and Production, 42, 283-288. Republic of Kenya. (2008). National Livestock Policy . Session Paper No. 2, Ministry of Livestock Development. Republic of Kenya. (2010). Agriculture Development Strategy (ASDS) 2010-2020. Nairobi- Kenya: Government Printers. United States Agency for International Development. (2010). Value Chain Analysis of poultry. Partnership for safe poultry in Kenya (PSPK) Program. Vincent, N., Lagat, B.K., Korir, M.K., Ngeno,E.K,. & Kipsat, M.J. (2010). Resource Use Efficiency in Poultry Production in Bureti District, Kenya. Paper Presented at the Joint 3rd African Association of Agric. Economists(AAAE) and 48th Agricultural Economists Association of South Africa(AEASA) Conference. Cape Town, South Africa. World Society for the Protection of Animals (WSPA),. (2012). Livestock production and climate change. Retrieved from http://unfccc.int/resource/docs/2012/smsn/ngo/194.pdf 2.Parallel Session 1: Crop Management, Biotechnology and Climate Change 2.1 Evaluation of seed quality and phenotypic characteristics of jute mallow (Corchorus olitorius) accessions from north rift region, Kenya David Kiprono Lelei Rutto1, Elizabeth N. Omami2, Julius O. Ochuodho3 1Student, Department of Seed, Crop and Horticultural Sciences, University of Eldoret P.O. Box 1125, Eldoret, Kenya. Email: ruttodkl@gmail.com Tel. 0723 960 413 2Dean and Associate Professor, School of Agriculture and Biotechnology, University of Eldoret, P.O. Box 1125 Eldoret, Kenya. Email: elizabethomami@yahoo.com 3Professor, Department of Seed, Crop and Horticultural Sciences, University of Eldoret P.O. Box 1125, Eldoret, Kenya. Email: juliusochuodho@yahoo.com ABSTRACT Jute mallow (Corchorus olitorius) vegetable production in North rift region is low due to many reasons, one of them being poor quality seeds. Yields of crop has remained low 2- 4 tons/ha/annum compared to expected yield of 5-8 tons/ha/annum. Farmers use much as 7kg/ha seed instead of 5kg/ha which is recommended for a density of 250 000 plants/ha that is 40% extra seed. The objective of study was to examine the seed quality 76 of Jute mallow accessions used by farmers at planting and evaluate their morphological diversity. A survey was conducted in North Rift region areas of Iten, Eldoret, Kapsabet, Kitale, Kapenguria, and Kenya Seed Company when farmers were asked about the husbandry of the crop. During the survey, some seeds were collected for quality tests on purity, germination and vigor using ISTA (2004) procedures. Seed sub-samples were then planted in green house and phenotypic characteristics recorded and data analyzed using GENSTAT package. The survey results showed 80% of the farmers growing jute mallow for seed were middle and old age people and 92% were females. Only 14% farmers were trained on vegetable seed production and agronomy. Education level analysis showed 54% of farmers had not attended school. Seed germination for the region was 57% that was below the quality standard of 60%. The phenotypic characterization showed diversity among the accessions with four clusters out namely: Green Early Maturing Short, Brown Early Maturing Short, Green Late Maturing Tall and Brown Late Maturing Tall. It was concluded that seed grown by farmers in North rift region is of poor seed quality and exhibit diversity. It’s recommended that purification be made to improve on genetic constitution of the accessions and proper post harvest handling of the seeds be practiced to improve seed quality. Key words: Indigenous, Vegetable. INTRODUCTION Jute mallow (Corchoru olitorius) vegetables are an integral part of obtaining nutritional security as they are a source of important nutrients essential for the good health (WHO, 2003). Jute mallow is noted for its important contribution to diet by supplying nutrients and rendering food more palatable (Grubben, 2004). In Kenya, there is ample production potential of Jute mallow not only to meet the growing demand for the domestic use requirement, but also produce surplus for export (K’opondo et al., 2005). C. olitorius is reported to be demulcent, deobstruent, diuretic, lactagogue, purgative, and tonic (Ayodele, 2005). It can meet major protein-calorific nutritional needs especially in children, sick, elderly and both expectant and lactating mothers in the rural areas (Smith and Eyzaguirre, 2007). Cooked leaves form a mucilaginous substance that has a character that is highly appreciated especially in areas where people depend on rather coarse food such as millets (Maundu, 1997). Jew’s mallow leaves are normally mixed with cowpeas to reduce their coarseness or to neutralize the bitter taste in Crotalaria brevidens (Abukutsa-Onyango, 2004). Currently, there is no good quality seed of the vegetable crop from the breeders to the farmers (Ndinya, 2005). Production of high quality seeds has not been practiced in many of the African leafy vegetables (Abukutsa-Onyango, 2002). Farmers have been getting re-cycled planting seed of Corchorus olitorius every season from the local markets and farms whose potential is not assured (Okongo, 2005). Yields of crop has remained low 2-4 tons/ha/annum compared to expected yield of 5-8 tons per ha per annum (KARI, 2009), and the major constraints has been use of poor quality seeds which forces farmers to use as much as 7kg/ha instead of 5kg/ha of seed required for a density of 250 000 plants/ha, which 77 amounts to 40% extra seed (MOA, 2009). The use of good quality seed has helped to raise yields (Amarjit, 1995). Also seeds with assured quality can be expected to respond to best agronomic practices and other inputs and increase yield in the expected manner (Bhattacharjee et al., 2000) For example in India good quality seed and use of fertilizer increased wheat production from 12 million tons to 31.3 million tons over a short period of ten years (Feistrizer, 1975). Characterizations of different types of plant resources within and across accessions have been increasing in the last years (Bhattarai et al., 2010). These have followed theoretical basis for inferring genetic diversity for the purpose of both breeding and conservation strategies in their identification (Akoroda, 1995). Despite increased importance of characterization in plant resources; there is scarce information about analysis of this type of data (Beebe, et al., 2000). To fill such a gap of information, research needs to be done to identify diversity of accessions and the parameters for specific plant (Drossou, et al., 2004). Surveys conducted in Kenya show that 50% of the people live below poverty line and live on less than one dollar a day (AICAD, 2003). To try and solve this problem, improved production and sale of vegetables like Jute mallow can play an important role in improving food security, health and general rural livelihoods (Fondio and Grubben, 2004). Importance of Jute mallow Corchorus olitorius is not usually eaten on its own but along other main meals as side dish (Chweya, 1985). The dark green leaves have varying proportions of Calcium, Iron, carotene, vitamin C and proteins required for good health (Schippers, 2002). Research has also shown they are nutritionally superior to exotic in many nutrients as they contain higher amounts of phosphorus, zinc, iodine, fluorine and vitamin B (Yebpella, et al, 2010). To avoid loosing such useful micro-elements it would be best to boil the ‘soup’ for as short as possible or steaming the produce would be much better (Chandha and Oluoch, 2003). The crop is a folk remedy for aches and pains, dysentery, enteritis, fever, pectoral pains, and tumors (Keller, 2004). Ayurvedics use the leaves for ascites, piles, tumors, cystitis, dysuria, fever, and gonorrhea (Kokwaro, 1990). The cold infusion is said to restore the appetite and strength of user. These vegetables fetch good prices in the market (Shiundu and Oniang’o, 2007). The expansion of production of these vegetables continues to be hampered by lack of a reliable source of planting materials and technical information (Ngoze and Okoko, 2005). The objective of study was to examine the seed quality of Jute mallow accessions used by farmers at planting and evaluate their morphological diversity. MATERIALS AND METHODS Survey and Jute mallow plant material collection for research. A survey was conducted in North Rift region areas of Iten, Eldoret, Kapsabet, Kitale, Kapenguria, and Kenya Seed Company when farmers were asked about the husbandry of the crop and C. olitorius seed samples simultaneously collected. A sample size of 500 farmers in the region was visited by use of stratified random sampling method. The 78 information collected using a structured questionnaire included age, sex, education background and seed and agronomy training of each farmer. The seed samples collected were then well-labeled and taken to the University of Eldoret seed laboratory for quality tests as per ISTA, (2004) procedures. Some seed sub-sample was thereafter grown in the green house for diversity evaluation. Seed testing as per ISTA, 2004 Analytical purity test Jute Mallow collected vegetable seeds were mixed well by passing through a seed divider several times and subsequently reconstituting them. The sub-samples for each source were obtained by successively halving the sub-sample until sample of required weight was got. Working samples of 15 grams were taken. Weight was taken for each sample and recorded. The working samples were placed on a purity work board and meticulously examined and separated to determine; seed purity. Weights of each group were separated and their percentages worked out from their initial weight of each sample (15 g) and results reported. Germination test Four replicates of 100 random seeds were obtained from the pure seed sample using a vacuum seed counter (with 100 holes) and placed on moistened filter paper inside a Petri dish, ensuring the seeds were sufficiently spaced. Seeds were placed in growth chamber set at 30oC temperature. Evaluation of germination was done with 1st count on 3rd day and last count on 5th day and recorded as follows: Normal and abnormal seedlings and dead seeds. Electrical conductivity (Seed vigor) test Distilled water of 250 ml was measured and put in 24 well labeled flasks and covered with aluminum foil to control any contamination. Other 2 flasks of distilled water of 250 ml was each set aside as control and covered with aluminum foil. The flasks were put on bench for 24 hrs at room temperature (20-23oC). Jute mallow seeds (50) were added to each flask and left to soak for 24 hours. The soaked seeds was then swirled gently for 10-15 seconds and conductivity ( S cm-1g-1) of soak water was measured upon meter showing constant reading. Between measurements, dip cell was rinsed twice in distilled water and dried using clean dry paper towels. The conductivity per gram of seed weight for each replicate was calculated after accounting for the background conductivity of the original water and the average of the four replicates provided the seed lot result. Phenotypic characterization of Corchorus olitorius at the green house The experiment was done in the green house. Boxes measuring 50 cm x 30 cm x 30 cm were filled with well mixed naturally fertile well – drained, friable, sandy clay soil of Ferro - orthic acrisol with an average pH of 5.0 and seed samples from markets centers, farms and Kenya Seed Company planted. At the green house, boxes were laid out using randomized complete block design each seed source with three replications in three blocks. Phenotypic characteristics from each sample were done and recorded as follows: 79 Plant height: Five random plants from the net plot (box) were measured for stem height from the base to the tip at harvesting time. Number of pods per plant: Five plants were sampled, from the net plot at harvest and the average number of pods determined. Leaf petiole length: Five plant leaf petioles were sampled, from the net plot at harvest and the lengths determined. Leaf Lamina: Five plants were sampled, from the net plot at harvest and leaf lamina shape recorded, its length and width determined. Leaf Serration: Five plants were sampled, from the net plot at harvest and the direction of serration recorded. Leaf tip shape: Five plants were sampled, from the net plot at harvest and the leaf tip shape recorded. Leaf color: Five plants were sampled, from net plot at harvest and the leaf color recorded. Flower characteristics: Five plants were sampled, from the net plot at harvest and the flower shape, color, number of sepals and its length and width recorded or determined. Fruit characteristics: Five plants were sampled, from the net plot at harvest and the fruits’ shape recorded and its length and width determined. Seeds per pod: Five plants were sampled, from the net plot at harvest and number of seeds per pod determined. RESULTS Survey and Jute mallow seed production Most of the farmers growing Jute mallow vegetable for seed were middle and old age people accounting for over 80% of total farmers interviewed. Men were missing in Jute mallow seed business with only 8% present in Eldoret and Kitale at the middle age (41- 59 years) making female dominate with 92% in its production in the region. Training of farmers Results on training of farmers showed that on average only 14% of all the total farmers interviewed in North Rift region had been trained on agronomy, crop protection and seed production aspects of Jute mallow. The remaining 86% farmers were doing Jute mallow seed production without any form of training at all. The trained farmers per area are: Kitale-24% (highest in the region), Eldoret-16%, Kapsabet-12%, Iten-11% and Kapenguria-8% (lowest). Education level of farmers Education level analysis showed that 54% of the farmers had not attended school. This means that over half of the farmers doing seed production are illiterate that is they cannot read and write and so cannot prepare or use farm records. The low level of education could be one of the hindrances to use of farm records in areas of North Rift 80 region as a tool for seed quality improvement but rely on their memories for information storage. Seed supply outlets of Jute mallow The farmers interviewed showed 98% depended on the informal seed supply sector for their Jute mallow seed, while only 2% depended on the formal seed supply sector. It was also observed that 69% of farmers use own saved seed for planting, while 24% got Jute mallow seed from neighbors, 5% from Informal seed sector (local market) and only 2% from Formal seed sector market. Fertilizer use for quality crop Fertilizer use showed 66% of the farmers interviewed do not use any fertilizer during the production of Jute mallow. Those farmers who apply fertilizer at planting do not top dress as required for improving quality of seed and crop that meet market grade /requirements. Method of planting Jute mallow During the planting of Jute mallow, 73% farmers broadcasting seed, thereby using higher seed rates as compared to 27% drilling in lines. It was observed that only 19% farmers use the optimum seed rate of 5 kg/ha, 10% farmers under use seed rate - (3% use 3 kg/ha and 7% use 7 kg/ha) as opposed to 71% of farmers that over use the seed rate - (29% use 6 kg/ha and 42% use 7 kg/ha). Indicators of Jute mallow pod maturity and harvesting method Majority of the farmers (84%) indicated that pod has matured when it turns from green color to tan indicating seed has reached harvest maturity, or the stage of seed development when it has dried to a moisture level at which it can be harvested without appreciable risk of mechanical damage. The method of harvesting Jute mallow seed by farmers involves cutting the whole plant as done by 75% of the farmers as opposed to 25% farmers who chose to hand pluck pods that have matured. Handling of Jute mallow seed from harvest to storage On the method of Jute mallow seed extraction, 98% farmers extracted their seed from pods by beating with sticks. In Jute mallow seed drying, it was observed that all the farmers (100%) dry their Jute mallow seed in direct sunlight as opposed to drying in shade for 4 days and only 1 day in sunlight. In seed storage, most of farmers (83%) store their Jute mallow seeds in pots/tins and plastic/synthetic. On storage duration, half of the farmers (53%) had saved their seeds for over 5years. Those that had stored for 4years was 23%, 3years was 20%, 2years was 15%) and those stored for 1year was 12%. 81 Seed quality testing of Jute mallow Analytical purity results The processed seed from Kenya Seed Company had the highest purity percentage (100%) compared with seed from both the markets (99%) and farms (98%). Purity analysis results showed that all the Jute mallow vegetable seed from the region were of superior quality as per ISTA (2004) procedures stating seed quality of above 98% purity is considered of superior quality. Germination results Seed germination for the North rift region was 57%, the highest germination was from processed source; Kenya Seed Company (86%) and the lowest germination was from farms source with 53%, though germination from markets were equally low (54%). Compared to ISTA, (2004) minimum Jute mallow seed germination requirement of 60%, Jute mallow seeds from North rift region do not meet this minimum standard. Radicle seedlings (normal and abnormal seedlings), result showed Kenya Seed Company had highest seedling emergence (88%) and Kapsabet farms (37%) had lowest compared with markets (59%) and farms (58%), (table 1). Also though seedling emergence of Kitale farms (73%) and Eldoret farms (72%) were high, germination percentage were reduced to Kitale (67%) and Eldoret (67%) by high number of abnormal seedlings by 6% (Kitale farms) and 6% (Eldoret farms). This trend was also observed in seed from other areas of North Rift region. On average, over half of seeds from North Rift region germinated on the 1st day. Electrical Conductivity of Jute mallow (Seed Vigor) The conductivity results showed that all seeds from the North rift were of high vigor. Kenya Seed Company had the leachate conductivity of 2.69 µScm-1g-1 (lowest) and Kapsabet farms had the Electrical conductivity of 7.09 µScm-1g-1 (highest) though so, all seeds were still of good quality as they were within range of blow 10 µScm-1g-1 level, within high vigor as per ISTA, (2004). Phenotypic characterization of Jute mallow Phenotypic characterization showed diversity among the accessions with four varietal clusters out namely: Green Early Maturing Short (GEMS), Brown Early Maturing Short (BEMS), Green Late Maturing Tall (GLMT), and Brown Late Maturing Tall (BLMT). Plant height measurements (cm) at harvesting C. olitorius plant height showed significant differences (p=0.001) for sources and locations. There was no interaction between variety and locations. The early maturing plants were short (28 cm) or dwarf or less than 50 cm (table 2). GEMS and BEMS were same in height (28 cm) and differentiated by stem color. GLMT and (BLMT) (82 cm) were tall and differentiated by color. 82 Pod count C. olitorius pod count results showed no significant differences (p=0.001) for source and locations. GEMS and BEMS pod count per plant showed same number of pods (6), likewise to GLMT and BLMT with 17 pods per plant, though BEMS highest was 14 pods from Kenya Seed Company (table 3) and stem color could only differentiate them. Leaf measurements i) Petiole length C. olitorius petiole length analysis showed no significant difference (p= 0.001) for sources and locations, nor between varieties. The GEMS and BEMS from markets and farms had same length of 1.3 cm and Kenya Seed Company had petiole length of 1.2 cm. GLMT and BLMT both from market and farm had same petiole length of 1.7 cm (table 4). ii) Leaf lamina C. olitorius leaf lamina shape was lanceolate with no significant difference (p=0.001) in leaf width. The late maturing variety had leaf width of 2.8 cm while early maturing variety had 2.6 cm (table 5). There was significant difference in leaf lengths (p=0.001) across varieties. Leaf length of late maturing (8.6 cm) was longer than early maturing of (6.0 cm). iii) Leaf serration. All GEMS, BEMS, GLMT and BLMT had serration present facing leaf apex. iv) Leaf tip. All GEMS, BEMS, GLMT, and BLMT leaf tips had acute shape. v) Leaf color. All GEMS, BEMS, GLMT, and BLMT leaves were green. Flower characteristics. All the GEMS, BEMS, GLMT, and BLMT flowers shapes were solitary and small and were in opposite side to the leaves, Color being yellow, Sepals were 5 and narrow. C. olitorius flower measurements showed no significant difference (p=0.001) for varieties. The GEMS, GLMT, BEMS, and BLMT had same flower width and length (1, 1.2 cm) (table 6). Fruit characteristics i) Size: C. olitorius fruit width and length measurements showed no significant difference (p=0.001) for varieties. The late maturing (1.3 cm) are significantly same in width as 83 early maturing (1.2 cm). There was variation for late maturing (3.9 cm) and early maturing (3.7 cm) in reference to fruit length (table 7). ii) Shape: All GEMS, BEMS, GLMT, and BLMT had fruits that were cylindrical, 10 ridged, dehiscing by 5 with traverse septa between the seeds. Seed characteristics All GEMS, BEMS, GLMT and BLMT had pyramidal shape length of 1 mm and dark grey in color. DISCUSSIONS Survey and seed collection The seed survey showed that most of the farmers growing Jute mallow vegetable were middle age (41-59 years) and old (60 years and over), accounting for over 80% of the total farmers interviewed meaning that Jute mallow growing was not done by young generation as observed also by Abukutsa- Onyango, (2002). In North Rift region, it was observed that men do not engage in growing Jute mallow crop in the farms which they consider best women work. Different scenario was observed at market where some 8% male practiced Jute mallow seed business in Eldoret and Kitale markets and were only at the middle age (41-59 years). Few farmers (14%) in North Rift region had been trained on vegetable seed production and agronomy, leaving 86% farmers doing vegetable seed production without having attended any training. This showed that they were using their own indigenous knowledge in Jute mallow seed production. There were 69% farmers having undergone at least one kind of training on the crop. No farmer had attended four or more trainings in the whole of North Rift region. This could be one of the reasons why Jute mallow yield has remained low at 2-4 tons/ha/annum as compared to the expected yield of up to 5-8 tons/ha/annum concurring with findings put forward by MOA, (2009). Education level showed that few farmers (46%) had attended school meaning that over half of the farmers (54%) doing seed production were illiterate i.e. could not read and write, so cannot prepare or use farm records to know their level of quality seed production, but relied on their memories to store information. Low level of education could be one of the hindrances to use of farm records tool in North Rift region. This in turn could have contributed to low quality of Jute mallow seed production in the region by farmers. Such decline in seed quality and crop productivity therefore reduces profits, which concurs with what Ellis and Roberts, (1981) put forward that low quality seeds leads to non acceptance of the vegetable by consumers because of low quality product, supporting also what George, (1985,1999) found out. Analytical purity of Jute mallow seeds Analytical purity showed that Kenya Seed Company had the highest purity percentage (100%) compared with seed from the regions, markets and farms. Iten market and 84 Eldoret farms had the lowest purity (98%). According to ISTA, (2004), the quality of seed analysis is considered superior if purity percentage is above 98% meaning that all the Jute vegetable seed from region were of superior quality. Though so, Louwaars, (1994) observed that purity per se is not gauge of good quality seed but a sum total of several seed attributes, needing more tests to ascertain seed quality. Germination of Jute mallow seeds Germination of Jute mallow seed from Kenya Seed Company (86%) was highest and the lowest germination was from Kapsabet farms (32%). This means that Kenya Seed Company Jute mallow seed would do better in field than all of the seeds from various sources analyzed. The North rift region results analysis showed low germination of 57%, though some of its Jute mallow seed sources like Kenya Seed Company was high (86%). Compared to ISTA, (2004), which requires Jute mallow seeds germination to have minimum germination percentage of 60%, North Rift region had its Jute mallow germination at 57%, thereby not meeting required minimum standard for it to be used as seed. Jute mallow seeds from Kenya Seed Company had high vigor (88%) in terms of emergence as compared with low Jute mallow seed emergence from other seed sources analyzed in North Rift region. This trend could be expected to perform the same in the field if these seeds were planted on the soils in farm. This low seedling emergence could be contributing to low yields and supports what was observed by Peacock and Hawkins, (1970) and Franzern et al. (1996). Electrical conductivity (seed vigor) of Jute mallow seeds Kenya Seed Company had the lower leachate conductivity of 2.69 µScm-1g-1 meaning that the seeds were of high vigor. Seeds from North Rift region had equally low leachate conductivity of 4.50µScm-1g-1. Kapsabet farms had higher Electrical conductivity of 7.09 µScm-1g-1 though the seeds were still of good quality. This means that though seeds from North Rift region were of high vigor, further tests on seeds are needed to ascertain why there were low yield in the region. This supports observations made by Amarjit, (1995) recommending several seed tests needed to be done to find out why seeds showed poor performance in field yet laboratory results showed high seed vigor. Jute mallow phenotypic characterization During phenotypic characterization four varietal clusters came out based on color, maturity and height as follows: Green Early Maturing Short, Brown Early Maturing Short, Green Late Maturing Tall and Brown Late Maturing Tall. Such observation was also found out in Nigeria by Epenhuijsen, (1974), though he only found two varieties and named them in Nigerian local language as Amugbadu, and Oniyaya and acknowledges that in reality there exists diversity in Jute mallow grown by farmers. 85 Early maturing short (GEMS and BEMS) varietal cluster characteristics The early maturing C. olitorius varietal clusters were short or dwarf (less than 50cm).This meant that the GEMS (28 cm) and BEMS (29 cm) had no significant difference in plant height and only differentiated by stem color during selection to different varieties. GEMS pod counts per plant were same for seed sources (6 pods), Kenya Seed Company had highest with 14 pods and the lowest was Kapenguria market (4 pods). This showed that Kenya Seed Company could produce more seed yield per unit area than from farm or market thereby increasing profit margins of farmers. The BEMS pod count per plant showed no significant difference between varieties, the highest was 6 pods and the lowest was 4 pods, leaving the average of the North Rift region BEMS with 6 pods per plant. Farmers could plant BEMS or GEMS as both were easily found within their areas, though if Kenya Seed Company could make their seed (GEMS) readily available to the farmers they would benefit more in terms of increased seed yield and profit from sales of Jute mallow. From the Petiole length analysis it was observed that there was no significant difference between varieties. On average GEMS and BEMS Petiole length was 1.2 cm and could only be differentiated by their stem colors. The GEMS and BEMS Leaf lamina shape was lanceolate measuring 5.5 cm length by 2.5 cm width and in alternate positions. The Leaf serration of both GEMS and BEMS were present and facing leaf apex. The leaf tip was acute in shape. The leaf color was green. The flowers were solitary, yellow, with 5 narrow sepals. Late maturing tall varietal cluster characteristics Corchorus olitorius GLMT and BLMT varietal clusters showed significant difference in plant height with 95 cm being the tallest plant height and with 75cm being shortest and an average of 85cm. There were no tall plants in Kenya Seed Company, which indicated that it had been purified to remove BLMT and GLMT varieties. The farmers could plant any of the varieties and get the required crop for sale during the planting season as both would give same output. There was no significant difference for pods between locations with GLMT and BLMT plants having pod count per plant at 18 pods, with lowest pod count per plant being 16 pods. Kenya Seed Company had no late maturing plants. Both GLMT and BLMT had leaf length at 1.7cm and could only be differentiated by their stem colors. The leaf lamina shapes for both GLMT and BLMT were lanceolate and in alternate position but varied in length sizes where GLMT was longer. Both types had leaf serration which were facing leaf apex with its leaf tips of acute shape. The flowers for both varieties were yellow, solitary, in opposite to leaves, same in size and had 5 narrow sepals. Both varieties had same sized fruit with cylindrical shape having 10 ridges in each fruit dehiscing by 5, with traverse septa between seeds. There were 100 seeds per capsule for both GLMT and BLMT. For both varieties, the seed color was dark grey with pyramidal shape. These seeds were 1 cm in length and had a seed weight of 400 seeds per gram. 86 TABLES Table 1: Germination and Emergence percentage of Jute mallow from various sources Source Location Standard germination test % Dead seeds % Abnormal seedlings % Normal seedlings % Emergence of radicle Markets Iten market 40 6 54 60 Eldoret market 29 5 66 71 Kapsabet market 42 4 54 58 Kitale market 46 5 49 54 Kapenguria market 47 5 48 53 Average of markets 41 5 54 59 Processd Kenya Seed Company 12 2 86 88 Farms Kitale farms 27 6 67 73 Eldoret farms 28 6 66 72 Iten farms 35 6 59 65 Kapsabet farms 63 5 32 37 Kapenguria farms 55 2 43 45 Average of farms 42 5 53 58 N/R region mean 57 Table 2: Mean plant height measurements (cm) at harvesting Source Location Mean plant height Mean plant height GEMS BEMS GLMT BLMT Markets Kitale 27±2.5 25±1.7 88±8 89±11.7 Kapsabet 33±2.9 35±2.6 95±13.1 91±0.6 Iten 28±1.5 26±1 83±6.7 80±8.5 Kapenguria 25±1.7 24±1 75±1 72±3.2 Eldoret 29±1.5 30±3.2 81±6.1 84±2.1 Mean of markets 28 28 84 83 Processed Kenya Seed Company 27±1 - - - Farms Kitale 30±2.5 32±3.1 82±3.1 78±7.2 Iten 26±0.6 27±0.6 85±4 81±5.3 Eldoret 30±1 31±3.8 83±7 83±5.5 Kapsabet 29±2.5 30±2.9 90±10.4 77±4 Kapenguria 27±1.5 28±1 83±3.5 82±7.6 Mean of farms 28 29 84 80 Note: (-) denotes variety not present in seed from the source 87 Table 3: Mean pod count in numbers of Jute mallow from various sources at harvesting Source Location Mean pod number Mean pod number GEMS BEMS GLMT BLMT Markets Kitale 5±1 6±1 17±0.6 17±1 Kapsabet 6±0.6 6 18±1.5 18±1.7 Iten 5±1.7 5 17±0.6 17±0.6 Kapenguria 4±0.6 4±0.6 16 16 Eldoret 6±1 6±0.6 16±0.6 16±1 Mean of markets 5 5 17 17 Processed Kenya Seed Company 14±2.1 - - - Farms Kitale 5±0.6 6 18±1.5 17±1.5 Iten 5±0.6 5±1 17±0.6 17±0.6 Eldoret 6±0.6 6±0.6 17±0.6 18±1.5 Kapsabet 6±0.6 6±1 17±1.2 17±0.6 Kapenguria 5±0.6 5±0.6 17±0.6 18±0.6 Mean of farms 5 6 17 17 Mean of North Rift region 6 6 17 17 Note: (-) denotes variety not present in seed from the source Table 4: Mean petiole length measurements (cm) of Jute mallow varieties at harvesting Mean length (cm) Mean length (cm) Source Location GEMS BEMS GLMT BLMT Markets Kitale 1.4±1 1.3±0.1 1.8±0.1 1.7±0.2 Kapsabet 1.3±0.1 1.3±0.1 1.7±0.1 1.7±0.1 Iten 1.3±0.1 1.2±0.1 1.6±0.2 1.6±0.1 Kapenguria 1.1±0.1 1.2±0.1 1.7±0.1 1.8±0.1 Eldoret 1.4±0.1 1.3±0.1 1.8±0.1 1.7±0.1 Mean of markets 1.3 1.3 1.7 1.7 Processed Kenya Seed Company 1.2±0.1 - - - Farms Kitale 1.3±0.1 1.3±0.1 1.7±0.1 1.7±0.1 Iten 1.3±0.1 1.2±0.6 1.7±0.2 1.6±0.1 Eldoret 1.3±0.1 1.2±0.6 1.8±0.1 1.7±0.1 Kapsabet 1.2±0.6 1.2±0.1 1.6±0.1 1.7±0.1 Kapenguria 1.2±0.1 1.2±0.1 1.6±0.1 1.7±0.1 Mean of farms 1.3 1.2 1.7 1.7 Mean of North Rift region 1.3 1.3 1.7 1.7 Note: (-) denotes variety not present in seed from the source 88 Table 5: Mean width and length measurements (cm) of Jute mallow leaf at harvesting Measurement Mean Mean of early maturing Mean Mean of late maturing GEMS BEMS GLMT BLMT Width (cm) 2.5±0.1 2.7±0.2 2.6 2.8±0.1 2.8±0.1 2.8 Length (cm) 5.5±0.4 6.5±0.5 6.0 8.4±1.1 8.8±1.2 8.6 Table 6: Mean width and length measurements (cm) of Jute mallow flower petals Measurement Mean Mean of early maturing Mean Mean of late maturing GEMS BEMS GLMT BLMT Width (cm) 1.0±0.1 1.0±0.1 1.0 1.0±0.1 1.0±0.1 1.0 Length (cm) 1.1±0.2 1.2±0.1 1.2 1.2±0.1 1.2±0.1 1.2 Table 7: Mean width and length measurements (cm) of Jute mallow fruit at harvesting Measurement Mean Mean of Early Maturing Mean Mean of Late Maturing GEMS BEMS GLMT BLMT Width (cm) 1.1±0.4 1.2±0.5 1.2 1.3±0.4 1.3±0.5 1.3 Length (cm) 3.5±0.5 3.8±0.6 3.7 3.8±0.4 3.9±0.6 3.9 CONCLUSIONS C. olitorius seed production status showed that most farmers in the North Rift region get their seed for planting from own saved seed or neighbors where storage was in plastic tins usually stored in the open. Most farmers were illiterate as shown by low level of education, were aged and have not been trained on seed and agronomic aspects of Jute mallow. C. olitorius seed grown by farmers in North Rift region though of high analytical purity is of poor physiological seed quality as seen from the high percentage of dead seeds. The dead seeds could be result of: - immature harvested seed, diseased seeds, poor processing methods/ storage conditions right from farms to markets. It was also concluded that seed grown by farmer’s exhibit diversity. This is true because during phenotypic characterization, four varieties came out based on color, maturity and height as follows: Green Early Maturing Short, Brown Early Maturing Short, Green Late Maturing Tall and Brown Late Maturing Tall. RECOMMENDATIONS The government of Kenya and other Stakeholders of the seed sector e.g. Universities, Seed merchants, KARI and KEPHIS to increase trainings to seed producers either formal or informal to enhance seed quality production and proper practices of post harvest handling of seeds. This could be done through farmer field schools so that the 89 farmers could be reached through group approach to improve on issues on seed quality production and their related agronomic practices. The recommended seed for use by farmers is from Kenya Seed Company, Kitale having showed high purity, germination, and vigor. The rest of seed lots from North Rift region should not be used as seed for it will lead to poor yields due to its low seed quality. It is also recommended that purification be made by breeders or seed companies to improve on genetic constitution or come-up with new varieties or improved selections. REFERENCES Abukutsa Onyango M.O., 2002 Market survey on African indigenous vegetables . Eds- JM Wesonga, T lozenge, CK Ndungu, K. Ngamau, JMB Njoroge, FK Ombwara, SG Agong, A Fricke, B Hau and H Stutzel. In: Proceedings of the second Horticultural seminar on Sustainable Horticultural Production in the Tropics at JKUAT 2002: 39-46. Abukutsa-Onyango M.O. 2004. Crotalaria Brevevidens. In: G.J.H Grubben and O.A. Denton (Eds). Plant Publishers, Leiden, Netherlands/CTA, Wageningen, Netherlands PP 229-231 AICAD, 2003. Examination of the needs through the study of poverty strategy for Kenya, Uganda and Tanzania. African Institute for Capacity Development and Japan International Cooperation Agency, PP. 1-54 Akoroda, M. O. 1995. Morphotype diversity in Nigerian Landraces of Corchorus olitorius. Journal of Horticultural Science 60 (4):557- 62. Amarjit S.B., 1995 Seed quality: Basic mechanisms and Agricultural implications, New York. Pp 135-169 Ayodele A.E. 2005. The medicinally important leafy vegetables of south western Nigeria. Available from: http://www.siu.edu/~ebl/leaflets/ayodele.htm. Beebe S., P. Skroch, J. Tohme, M. Duque, F. Pedraza and J. Nienhuis., 2000. Structure of genetic diversity among common landraces of middle American origin based on correspondence analysis of RAPD. Crop Sci 40:264– Bhattacharjee, A.K., B.N. Mittra and P.C. Mitra., 2000. Seed agronomy of Jute II. Production and quality of Jute (Corchorus olitorius) seed as influenced by nutrient management, Seed Science and Technology, V.28 (1): PP. 141-154 Bhattarai, K., B.S. Bushman, D.A. Johnson and J.G. Carman. 2010. Phenotypic and genetic characterization of western prairie clover collections from the western United States. Rangeland Ecology and Management 63:696-706. Chandha, M.L. and M.O. Oluoch., 2003. Home based vegetable gardens and other strategies to overcome micronutrient malnutrition in developing countries. Food, Nutrition and Agriculture, FAO, 32: PP. 17-23. Chweya J.A., 1985 Identification and nutritional importance of indigenous green leaf vegetables in Kenya. Acta Hort. 1985; 153: 99 - 108 90 Drossou A, A. Katsiosis, J. Leggett, M. Loukas and S. Tsakas, 2004. Genome and species relationships in genus Avena based on RAPD and AFLP molecular markers. Theory Appl. Genet. 109:48–. Ellis, R.A. and E.H. Roberts, 1981. The quantification of ageing and survival in orthodox seeds. Seed Sci. Technol., 9: 373–409 Epenhuijsen C.W. Van, 1974 Growing native vegetables in Nigeria FAO, Rome, Italy Pp 96-124 Feistrizer W.P., R. Bradley and F. Ogada, 1975 Seed production and harvesting in cereal seed technology. FAO Pub., Rome, Pp37 Franzen H.P., F.B. Begemann, J.A. Wadsack and H. Rudaf, 1996 Variety improvement in the informal sector: Aspects of a new strategy, workshop proceedings: 26-29 July 1995: IPGR, Rome. Pp 19-30. George R.A.T. 1985. Vegetable seed production, Longman, London Pp 75-102. George R.AT. 1999. Vegetable seed production. CAB International, Walingford U.K. 2nd Ed Pp 58-78. Grubben, G.J.H and O.A. Denton, 2004. Plant Resources of Tropical Africa vegetables. PROTA Foundation, Backhuys Publishers, Leiden, Netherlands/CTA, Wageningen Netherlands. International Seed Testing Association, 2004 International rules for seed testing, Edition 2004, Bassersdorf, CH- Zurich, Switzerland K’Opondo F.B.O., R.M. Muasya and O.K. Kiplagat, 2005 A review on the seed production and handling of indigenous vegetables (spider plant, Jute mallow and African nightshade complex) In: MO Abukutsa-Onyango, AN Muriithi, VE Anjichi, K Ngamau, SG Agong, A Fricke, B Hau, and H Stutzel (Eds) Proceedings of the third Horticulture Workshop on Sustainable Horticultural Production in the Tropics. Maseno University, Maseno 2005: 42-48. K’opondo, F.B.O., Muasya R.M. and O.K. Kiplagat, 2005. A review on the seed production of and handling of indeginous vegetables (spider plant, Jute Mallow and African nightshade complex) In: M.O. Abukutsa-Onyango, A.N. Muriithi, V.E. Anjichi, K.. Keller, G.B. 2004. African Nightshade, Eggplant, Spider flower et al - production and consumption aspects of traditional vegetables in Tanzania from the farmers point of view. Goettingen: Georg-August Universitaet. Kenya Agricultural Research Institute, 2009 Annual report, Government Printers, Nairobi, Kenya Kokwaro, J.O., 1990. Seasonal Traditional Vegetables of Kenya. In H.D. Ihlenfeldt (Ed). Proceedings of the Twelfth Plenary Meeting Aetfat, Symposium VII, Int. Alleg. Bot. Hamburg, Band 23b S, 1988 Hamburg (Germany). Mitteilungen Aus Dem Institut Fur Allgemeine Botanik, Humburg. PP. 911-928. Louwaars N. P., 1994 Integrated seed supply: a flexible approach. In: Hanson J. (Eds) Seed production by small holder farmers. Proceedings of the ILCA/ICARD Research Planning Workshop held in ILCA, Addis Ababa, Ethiopia, 14-15 June 1994. 58pp. 91 Maundu P.M., 1997. The status of traditional vegetable utilization in Kenya, In: L Guarino (Eds) Proceedings of IPGRI International workshop on genetic resources of traditional vegetables in Africa: conservation and use. ICRAF-HQ, Nairobi, Kenya 1997: 66-75 Ministry of Agriculture, 2009 Annual report, Government Printers, Nairobi, Kenya Ndinya C., 2005. Seed production and supply systems of three African leafy vegetables in Kakamega District. In: MO Abukutsa-Onyango, AN Muriithi, VE Anjichi, K Ngamau, SG Agong, A Fricke, B Hau, and H Stutzel (Eds). Proceedings of the third Horticulture Workshop on Sustainable Horticultural Production in the Tropics Maseno University, Maseno 2005: 60-67 Ngoze S. and N. Okoko, 2005 On farm seed production in Kisii District: An overview of recent situation. In: MO Abukutsa-Onyango, AN Muriithi, VE Anjichi, K Ngamau, SG Agong, A. Fricke, B Hau, and H Stutzel (Eds). Proceedings of the third Horticulture Workshop on Sustainable Horticultural Production in the Tropics Maseno University, Maseno 2005: 59-60 Okongo P., 2005 Community seed production of African indigenous vegetables In: MO Abukutsa-Onyango, AN Muriithi, VE Anjichi, K Ngamau, SG Agong, A Fricke, B Hau, and H Stutzel (Eds) Proceedings of the third Horticulture Workshop on Sustainable Horticultural Production in the Tropics Maseno University, Maseno 2005: 33-37 Peacock H.A. and B.S. Hawkins, 1970 Effects of seed source on seedling vigor and yield. Crops Sci.10: 667-70 Schippers R.R., 2002 African indigenous vegetables: An overview of the cultivated species. Chatham, UK. Natural Resources Institute / ACP-EU Technical Centre for Agricultural and Rural Cooperation.2002. Shiundu KM, Oniang’o RK (2007). Marketing African Leafy Vegetables: Challenges and Opportunities in the Kenyan Context. Afr. J. Food Agric. Nutr. Dev. Smith, I. F. and P. Eyzaguirre,. 2007. African leafy vegetables: Their role in the third world Health Organizations Global Fruit and Vegetables Initiative AJFAND Volume 7 No 3 African Journal of food, Agriculture, Nutrition and Development, Rural Outreach Kenya, Nairobi, Kenya. WHO, 2003. World Health Organization: Diet, Nutrition and the prevention of chronic dieases. Report of a joint FAO/WHO Expert consultation, WHO Technical Report No 916, Geneva, Switzerland. Yebpella G.G., Magomya A.M., Udiba U.U., Gandu I., Amana S.M., Ugboaja V.C., Usman N.L. (2010). In Assessment of Cd, Cu, Mn and Zn Levels in soil, water and vegetable grown in irrigated farm along river Kubani, Zaria, Nigeria. J. Appl. Environ. Biol. Sci. 1(5):84-89. 92 2.2 Morphological and Biochemical Traits of Sorghum Fostering Abiotic Stress Tolerance Superiority among Cereals: A Review Egesa A. O.1*, Mwirigi P.1, Muui W. C.1, and Kirubi, D. T1 Department of Agricultural Science and Technology, Kenyatta University, P.O. BOX 43884- 00100, Nairobi Corresponding author’s email; ayechoogola@gmail.com Abstract Sorghum (Sorghum bicolar L. moench) is very important for high tolerance to both abiotic and biotic stressors. The crop is amongst five most important cereals in the world contributing to food, fodder, fuel, ethanol and other industrial products like ropes and bags. Sorghum derived foods are special for weight management; they function in body sugar regulation through slower digestibility and longevity in the digestive system. The crop initially neglected, is gaining importance in the agricultural sector due to its ability to grow well and produce in poor precipitation, high soil salinity and in infertile soils. These traits enhance sorghum’s potential in enhancing food security and enabling sustainable crop production in face of climate change. The unique morphological and biochemical advantages of this crop are necessary for evaluation and introduction into other important cereals. This review highlights the exceptional abiotic stress management mechanisms of sorghum, illustrating why this orphan crop should be considered as a model crop for research in understanding and addressing a biotic plant stressors in cereals for better productivity. Keywords: Sorghum, a biotic stress, traits, reals, food security INTRODUCTION Sorghum is an annual cereal crop of tropical Africa origin, (Burkhart, 2007; Menz et al., 2004, Peterson et al., 2009). The crop was domesticated in Africa back in 3000 BC, (Burkhart, 2007). However, it has spread all over the world, with current production in Africa, China, Central America, South America, India, Australia and the United States. It is a staple food crop in sub-Saharan Africa and Asian countries, (Menz et al., 2004; Ratnavathi & Patil, 2013), and used as a major source of fodder in India, china, North America and Australia, (Csrl, 2013). In United States of America, sorghum introduction was mainly for syrup production; currently it is an important source of fodder and uses in ethanol production, (Peterson et al., 2009). Sorghum grows well in a wide range of agro ecological zones (high potential to low potential lands), it is explicitly known for above average growth and productivity in marginal areas characterized by prolonged droughts, (Price et al., 2005; Dalal et al., 2012). Apart from abiotic stresss, some cultivars of sorghum harbor beneficial insects including: beetles and lacewings, this offers better pest management measure, (Clark, 2007). 93 As a staple food, sorghum is used in preparation of meals such as: porridge, biscuits, cakes and bread, (Ratnavathi & Patil, 2013; Yones et al., 2016), the grains are also used in preparation of alcoholic beverages, while the other parts are used to make wall boards, and packaging materials, (Delserone, 2008). Forage sorghum cultivars are used to prepare quality silage and hay; some are chopped and fed to livestock directly or as pastures. Other sorghum cultivars are now in use for biofuel production, (Dalal et al., 2012; Csrl, 2013). Sorghum is an upright annual plant, a member of the Poaceae family. Have flat to erect leaves. The plant exhibits varying heights and maturity, this is highly dictated by the cultivar. Growth characteristics also differ due to manipulation by the environment, production resources (inputs) and cultural practices. Sorghum can be categorized into: grain sorghums (Sorghum bicolor ssp. bicolor), Sudan grass (Sorghum Bicolor ssp. drummondii) and forage sorghums (Sorghum bicolor), (Kearney & Peebles, 1969; Burkhart, 2007). Forage sorghums are more vegetative, taller with late maturity, Sudan grass is in between the forage sorghum and the shorter grain sorghum, and this is not always the case since there a number of taller grain sorghum. Both Sudan grass and forage sorghum are mainly used as livestock feeds, and do exhibit high regeneration, (Awad et al., 2013). Sorghum alongside other small grained cereals such as millet, are important hardy cereals for water use efficiency characteristics, enabling high tolerance to water limiting conditions. A number of Sorghum cultivars grow fast and mature within short rainfall seasons in semiarid areas. The crop has better regeneration ability after exposure to dire conditions of drought and high temperatures. The crops exhibits good growth and development in resource deficient soils. Accumulation of compatible solutes within the sorghum plant tissues is credited for efficient osmo-regulation system; this has been specifically attributed to survival in saline environments, (Csrl, 2013, Zhang et al., 2014). In general sorghum exhibits a complex interaction of both physiological and biochemical processes that synergistically enhance its ability to tolerate abiotic stressors, (Dalal et al., 2012; Schittenhelm & Schroetter, 2014). The current climate change consequences of elevated frequencies of heat stress, floods and droughts are destabilizing world food production, necessitating measures in understanding the stress tolerance and adaptation mechanisms in hardy crops for their incorporation and over expression in agriculturally important crops. Sorghum provides such an avenue, (Blum, 2005; Dalal, et al., 2012). Drought tolerance Climatic variability is a major contributor to inconsistency of crop yields, (Kang et al., 2009; FAO, 2015). One important crop production factor, water is highly affected by the climate variations; this is seen in drought and flood events that incapacitate crop growth, (Trenberth, 2008). Limited water for plant growth and development is a common condition in the dry lands, (Anjum et al., 2011; Dalal et al., 2012), and is recently reported to advance to highly productive agro ecological zones as a result of climate change, (Dalal et al, 2012). Today, unexpected water shortages in important crop 94 production zones is decreasing the total food production and increasing the challenge of food insecurity, (Shao et al., 2009). Sorghum possesses superior traits for tolerance to abiotic stressors, (Assaefa et al., 2010; Csrl, 2013). Among them; Drought escape (completion of life cycle ahead of water deficit onset), as seen in early anthesis of some cultivars. Dehydration avoidance, fostered by superior soil moisture absorption of the deep roots, and reduced evapo- transpiration from leaves. The crop also demonstrates dehydration tolerance, the capacity to uphold cellular processes in low tissue water potential, (Dalal et al., 2012). The leaves of sorghum have limited expansion, manipulated stomata opening, and high photosynthesis in elevated temperatures. Generally plants are directly affected by the reduction in leaf turgor potential as a result of less water, this significantly hinder the normal functioning of cells (Jones & Turner, 1977). However, some plants have elaborate survival mechanisms, as reported by Jaleel et al., (2008), upon noting increased root development of Catharanus spp under water stress. High evapo-transpiration needs by plants in elevated temperatures result to poor metabolism. Low soil fertility cause stunted growth to plants. Saline environments hinder plants ability to capture water even in moist soil. It is therefore very important to study stress tolerance mechanisms by plants and devise measures for their enhancement in important crops, this starts with understanding the behaviors of common abiotic stress tolerant plants, (Duan et al., 2007; Mickelbart et al., 2015). Sorghum root system Sorghum is significantly deep rooted, (Dalal et al., 2012; Schittenhelm & Schroetter, 2014). The extensive root system may penetrate up to eight feet into the soil and more than three feet away from the stem; extending from 1 m to 1.6 m away from the plant as described by Assefa et al., (2010). These roots enable capture of water from the subsoil (Shoemaker & Bransby, 2010). Both the seminal and nodal roots of sorghum are uniquely endowed with large vessel diameters, this is believed to affect vessel–vessel connections, and are factors in embolism and refilling as described by Johnsons et al., (2013), from their work in investigating hydraulic signaling means between roots and shoots for arid plants. Root thickness, root length density, number of thick roots and root volume are factors that contribute to both water and nutrients absorption by the plant, (Ronggai et al., 2015). Leaf and Stem based modification Sorghum leaves are usually green, glossy, flat to erect, and relatively narrow; having reduced number of stomata’s, hence less exposure to water loss. High epicuticular wax on sorghum leaves enhances efficient water conservation, by lowering water loss from the leaves by evapo-transpiration, (Dalal et al., 2012). Wax is made up of free fatty acids, n- alkanes, steroids, aldehyde, both primary and secondary alcohols and ketones. Over 30% of water conservation by sorghum is by this wax. Cameron et al., (2006), observed instances of increase in wax accumulation in tree tobacco leaves a result of inducement by dehydration stress. Similarily, Hamissou and Weibel, (2004), had initially illustrated 95 the relationship between wax cover and plant water content, asserting a correlation of high plant water content with high wax accumulation. In their work, Shepherd and Griffiths, (2006), observed that increase in wax thickness (accumulation) in cuticle was contributed by increase in temperature and light intensity, implicating wax instances in crops to be more in dry hot conditions typical of the tropical region believed to be where sorghum was first domesticated. Accumulation of wax is controlled by a type of lipid proteins, increase in associated genes in water limiting conditions were observed by Cameroon et al., (2006), these genes can be screened, isolated and studied for introgression in other crop plants. According to Shepherd & Griffiths., (2006), the wax layer on leaves creates a hydrophobic interface between the leaf tissue and the outside preventing water loss. Leaf rolling that precede the stomata closure in high temperatures as influenced by leaf water potential together with stomatal conductance (controlled by soil moisture dependent signals) these are characteristics uniquely efficient in sorghum, (Amelework, 2015). The rows of motor cells along the midrib on the upper part of sorghum leaf enable rapid roll up of leaves during moisture stress. This technic is said to be evident about twelve days of exposure to water stress, Assefa et al., (2010). The stem of sorghum plant is solid and dry, to succulent. The stems are light green, blue to whitish in color with a thick waxy layer. The dull colouration and wax layers contributes to reduced water loss. Osmoregulation In spite of the long root systems there are enzymatic mechanisms facilitated by deposition of solutes in root vacuoles, this enables attainment of an osmotic adjustment that function in increasing flow of water from the dry soils into sorghum plant. Solute accumulation and sugar storage in sorghum root and other tissues is in form of soluble solutes that are non- toxic at high cytolsolic concentration. Osmotic adjustment has a significant role in maintaining the turgor relations of plants subjected to water deficit (Jones & turner, 1977; Girma & Krieg, 1992). The lowering of osmotic potential by net solute accumulation limit water loss from leaf tissues. These solutes also function in detoxification of reactive oxygen species, stabilization of membranes and are key in maintainace of enzymes and proteins native structures. Sorghum in particular produces proline and glycinebetaine, these are osmolytes preventing cells from dehydration. Dwarf sorghum varieties are believed to synthesis more osmolytes, an indicator of higher expression of glycinebetaine and proline ecoding genes. The elevated osmotic adjustment in sorghum is believed to be acquired high selection pressure in dry areas. Sorghums, exhibiting this have slow rates of leaf senescence under water limiting condition and similarly displays, postponed leaf rolling and leaf tissue death, a contribution of the massive accumulation of compatible solutes, (Chen et al., 2007). 96 Uptake of micro elements to boost resilience to water stress in saline conditions Absorption and deposition of silicon in sorghum roots has been known to reduce Na+ toxicity in sorghum and similarly enhanced water uptake from dry soils, (Lux et al., 2003). Yin et al., (2015) found silicon amendment in soil causing reduced uptake of Na+ by the sorghum roots, this enables reduced salinity stress to the plant. There are cases of storage of active ions in vacuoles and deposition in tonoplast preventing cell toxicity. It had also been illustrated earlier by Yin et al., (2014), in his description of silicon mediated enhanced water uptake by sorghum roots in dry conditions. Uptake and deposition of silica in stems and roots of the plants in poaceae family led to development of highly branched roots, this had been outlined by Hattori et al., (2003) as one that advantageously benefited such plants promoting their survival in dry soils. Similarly, silica enhanced rapid root elongation. According to Lux et al., (2002), who observed higher aggregation of silica in drought tolerant sorghum, drought tolerance could also be contributed by the microelement. In saline environment, sorghum is able to undergo root cells modifications and increased production of energy, (Kong et al., 2013). Current research is directed at outlining the role of micro element as abiotic stress tolerance enhancers in discovery of functions of novel genes, (Yin et al., 2015), postulated a possibility of gene mediated mechanisms being in play in enhancing salinity tolerance of sorghum in silicon amended media. Secretion of anti-stress compounds Sorghum exhibits an elaborated root system this has been credited against soil water deficit through distant water sourcing, however, this doesn’t imply that the crop has high dependence to distant water sources, suppression of weeds my also function in enhancing less completion to water resources. The root exudates of sorghum have been shown to reduce the growth of weeds such as velvet leaf, thorn apple, redroot pigweed, crabgrass, yellow foxtail and barn yard grass, (Stapleton et al., 2010). Sorghum is also recommended for control of nut sedge infestations, (Clark, 2000). Sorgoleone excreted by sorghum roots suppresses the growth of many other weedy plant species, (Dayan et al., 2010). High temperature tolerance in sorghum The tolerance of sorghum to high temperature depends on; the duration of exposure of sorghum to high temperature, the activity or stage of growth of the exposed tissue and the thermal adaptation of the particular sorghum cultivar. The cutin, consisting of the uppermost structure of leaves, is made up of mononers, without double bond systems, making them poor aborbers of ultra violet radiations, this is same to the dull stems, (Hamissou and Weibel, 2004). Norcio, (1976) found a positive correlation between high photosynthesis rates at high temperatures and cellular heat tolerance of sorghums determined by the "electro conductivity" method (Sullivan, 1972). When heat stress is involved, an ability to maintain relatively high rates of photosynthesis may very well contribute to yield. A 97 series of experiments were conducted, at ICRISAT, in the laboratory, glasshouse and field studying the emergence of a number of sorghum genotypes over a range of high temperatures. The work demonstrated that some lines had ability to emerge even when soil surface temperatures were as high as 55°C , (Craufurd & Peacock, 1993, Kumar et al., 2008). In a study using 6 sorghum cultivars grown in 14 different environments, it was found that tolerance to high temperatures contributed more to yield stability. Specifically, where high temperatures occurred during panicle development, the capacity of a genotype to produce greater seed numbers per head (and hence greater yield) was important. However, the ability of genotypes to heat tolerance varies, (Craufurd & Peacock, 1993; Sing et al., 2016). Prospective advances in utilizing sorghum as a model crop for a biotic stress tolerance Varied drought stress tolerance traits conferred by sorghum may be useful to other cereals including maize and wheat, (Geleta & Labuschagne, 2003). The immense sorghum germplasm, is yet to be fully exploited in understanding of abiot stress management and very few accessions have been utilized in new varieties development. The huge pool of over 37,000 described sorghum accessions will provide a good source of abiotic stress tolerance mechanisms, (Mullet at al., 2002; Csrl, 2013), since many of the genes responsible are yet to be mapped the venture will involve gathering of the materials and subsequent screening for the traits, (Dalal et al., 2012).While testing the potential use of genes from sorghum to other crops, Mullet et al., 2003 reported success in transferring Cryogenic glycosides coding genes into Arabidopsis, enabling resistance to the insects in the family Chrysomelidae. It is advantages that sorghum genome is small 730 Mb, and has already been fully sequenced, (Peterson et al., 2008). The wide germplasm resource, lower level of gene duplication unique to sorghum makes this cereal ideal especially for grasses and plant genomics research as a whole, (Dillon, 2007; Dalal et al., 2012). Increased utilization and use of sorghum genetic materials in research will spearhead identification of novel genes, which may be of benefit in abiotic stress tolerance, this would be done screening for gene functions for the varied sorghum tolerance mechanisms. In addition such efforts will enable prevention of the increased biodiversity loss, (Geleta & Labuschagne, 2003; Peterson et al., 2008).). CONCLUSION Sorghum is a cereal crop exhibiting adaptation to varied conditions of low rainfall, low soil nutrients, high temperatures and grows well to maturity in such conditions. The experimentation and documentation of unique process and adaptation behaviors will pave way for quick gene screening, and experimentation involving gene tweaking, gene transfer, within sorghum and to other crops. The fact that sorghum has a smaller genome, it will be easier to extensively study the genes responsible for this stress management. Introgression of these superior salinity and water stress tolerance 98 mechanisms to maize wheat, rice and other important cereals promises a potential for better food production. REFERENCES Amelework, B., Shimelis, H., Tongoona, P., & Laing, M. (2015). African Journal of Agricultural Research Physiological mechanisms of drought tolerance in sorghum, genetic basis and breeding methods: A review. African Journal of Agricultural Research,10(31),3029–3040. https://doi.org/10.5897/AJAR2015.9595 Anjum, S., Xie, X., & Wang, L. (2011). Morphological, physiological and biochemical responses of plants to drought stress. African Journal of Agricultural Research, 6(9), 2026–2032. https://doi.org/10.5897/AJAR10.027 Assefa, Y., Staggenborg, S. A., & Prasad, V. P. (2010). Grain Sorghum Water Requirement and Responses to Drought Stress: A Review. cm, 9(1), 0. doi:10.1094/cm-2010-1109- 01-rv Awad, A., Hafiz, S., Hammada, M. S., El-nouby, A., & El-hendawy, S. (2013). Grain yield production of Sudan grass (Sorghum sudanense (Piper) Stapf) as influenced by cutting numbers, potassium rates, and intrarow spacing in a semiarid environment. Turkish Journal of Agriculture and Forestry, 37, 657-664. doi:10.3906/tar-1208-53 Bidinger, F. R., & Johansen, C. (1988). Drought Research Priorities for the Dryland Tropics. Screening. Blum, A. (2005). Drought resistance, water-use efficiency, and yield potential—are they compatible, dissonant, or mutually exclusive? Australian Journal of Agricultural Research, 56, 1159–1168. https://doi.org/10.1071/AR05069 Burkhart, E. P. (2007). Flora of North America: North of Mexico. Volume 24: Magnoliophyta: Commelinidae (in part): Poaceae, part 1. Edited by Mary E Barkworth , , Kathleen M Capels , , Sandy Long , , Laurel K Anderton , and , Michael B Piep ; illustrated by Cindy Talbot Roché, Linda Ann Vorobik, Sandy Long, Annaliese Miller, Bee F Gunn, and Christine Roberts. Oxford and New York: Oxford University Press. $95.00. xxviii + 911 p; ill.; nomenclatural index. ISBN: 0‐19‐531071‐3. 2007. The Quarterly Review of Biology, 82(3), 278-278. Doi:10.1086/523149 Cameron, K. D. (2005). Increased Accumulation of Cuticular Wax and Expression of Lipid Transfer Protein in Response to Periodic Drying Events in Leaves of Tree Tobacco. PLANT PHYSIOLOGY, 140(1), 176-183. doi:10.1104/pp.105.069724 Craufurd, P. Q., & Peacock, J. M. (1993). Effect of Heat and Drought Stress on Sorghum (Sorghum Bicolor). II. Grain Yield. Experimental Agriculture, 29(01), 77. doi:10.1017/s0014479700020421 Csrl, V. (2013). Proceedings of the 2013 Sorghum Improvement Conference of North America (SICNA ) Meeting Theme : “ BRIDGING RESEARCH GAPS BETWEEN PUBLIC AND PRIVATE SECTORS ,” 1–30. 99 Dalal, M., Mayandi, K., & Chinnusamy, V. (2012). Sorghum: Improvement of Abiotic Stress Tolerance. Improving Crop Resistance to Abiotic Stress, 2, 923–950. https://doi.org/10.1002/9783527632930.ch36 Dillon, S. L., Shapter, F. M., Henry, R. J., Cordeiro, G., Izquierdo, L., & Lee, L. S. (2007). Domestication to crop improvement: Genetic resources for Sorghum and Saccharum (Andropogoneae).Annals of Botany. https://doi.org/10.1093/aob/mcm192 FAO. (2015). Climate Change and Food Security: Risks and Responses. Girma, F. S., & Krieg, D. R. (1992). Osmotic Adjustment in Sorghum 1. Mechanisms of Diurnal Osmotic Potential Changes. Plant Physiol, 99, 577–582. Hamissou M & Dale E. Weibel. 2004. The Effects of Epicuticular Wax Cover on the Rate of Water Loss of Sorghum bicolor (L.) Moench. Asian Journal of Plant Sciences, 3: 742-746. Hattori, T. (2003). Silicon-Induced Changes in Viscoelastic Properties of Sorghum Root Cell Walls. Plant and Cell Physiology, 44(7), 743-749. doi:10.1093/pcp/pcg090Jaleel, C.A., P. Manivannan, A. Wahid, M. Farooq, R. Somasundaram and R. Panneerselvam, (2009). Drought stress in plants: a review on morphological characteristics and pigments composition. Int. J. Agric. Biol., 11: 100–105 Jones, M. M., & Turner, N. C. (1978). Osmotic adjustment in leaves of sorghum in response to water deficits. Plant Physiology, 61(1), 122–126. https://doi.org/10.1104/pp.61.1.122 Kang, Y., Khan, S., & Ma, X. (2009). Climate change impacts on crop yield, crop water productivity and food security – A review. Progress in Natural Science, 19(12), 1665- 1674. doi:10.1016/j.pnsc.2009.08.001 Kumar, R.S., Hammer, G. L., Broad, I., Harland, P., & McLean, G. (2009). Modelling environmental effects on phenology and canopy development of diverse sorghum genotypes. Field Crops Research, 111(1-2), 157-165. doi:10.1016/j.fcr.2008.11.010 Lux, A., Luxová, M., Abe, J., Tanimoto, E., Hattori, T., & Inanaga, S. (2003). The dynamics of silicon deposition in the sorghum root endodermis. New Phytologist, 158(3), 437– 441. https://doi.org/10.1046/j.1469-8137.2003.00764.x Lux, A., Luxova, M., Hattori, T., Inanaga, S., & Sugimoto, Y. (2002). Silicification in sorghum (Sorghum bicolor) cultivars with different drought tolerance. Physiologia Plantarum, 115(1), 87-92. doi:10.1034/j.1399-3054.2002.1150110.x Menz, M. A., Klein, R. R., Unruh, N. C., Rooney, W. L., Klein, P. E., & Mullet, J. E. (2004). Genetic diversity of public inbreds of sorghum determined by mapped AFLP and SSR markers. Crop Science, 44(4), 1236– 1244.https://doi.org/10.2135/cropsci2004.1236 Mickelbart, M. V., Hasegawa, P. M., & Bailey-Serres, J. (2015). Genetic mechanisms of abiotic stress tolerance that translate to crop yield stability. Nature Reviews Genetics, 16(4), 237–251. https://doi.org/10.1038/nrg3901 100 Mullet, J. E., Klein, R. R., & Klein, P. E. (2002). Sorghum bicolor — an important species for comparative grass genomics and a source of beneficial genes for agriculture. Current Opinion in Plant Biology, 5(2), 118-121. doi:10.1016/s1369-5266(02)00232-7Norcion. V. (1976) Effect of hlgh temperature and water stress on photosynthesis and resplratlonrates of grain sorghum Ph D Dlssertatlon. Unlverslty of Nebraska. Llncoln. Nebraska. USA Paterson, A. H. (2008). Genomics of Sorghum. International Journal of Plant Genomics, 2008, 1-6. doi:10.1155/2008/362451 Paterson, A. H., Bowers, J. E., Bruggmann, R., Dubchak, I., Grimwood, J., Gundlach, H., … Rokhsar, D. S. (2009). The Sorghum bicolor genome and the diversification of grasses. Nature, 457(7229), 551-556. doi:10.1038/nature07723 Price, H. J., Dillon, S. L., Hodnett, G., Rooney, W. L., Ross, L., & Johnston, J. S. (2005). Genome evolution in the genus Sorghum (Poaceae). Annals of Botany, 95(1), 219– 227. https://doi.org/10.1093/aob/mci015 Ratnavathi, C. V., & Patil, J. V. (2013). Sorghum Utilization as Food. Journal of Nutrition & Food Sciences, 4(1), 1–8. https://doi.org/10.4172/2155-9600.1000247 Ronggai L, Yucui H, Peng L, Ruiheng D and Guoqing L (2014). Molecular mapping of the brace root traits in sorghum (Sorghum bicolor L. Moench). Breeding Science 64: 193– 198 (2014).doi:10.1270/jsbbs.64.193 Schittenhelm, S., & Schroetter, S. (2014). Comparison of Drought Tolerance of Maize, Sweet Sorghum and Sorghum-Sudangrass Hybrids. Journal of Agronomy and Crop Science. https://doi.org/10.1111/jac.12039 Shao H.B, Chu LY, Jaleel CA, Manivannan P, Panneerselvam R, Shao MA. (2009). Understanding water deficit stress-induced changes in the basic metabolism of higher plants-biotechnologically and sustainably improving agriculture and the ecoenvironment in arid regions of the globe. Crit. Rev. Biotechnol., 29: 131-151 Shepherd, T., & Wynne Griffiths, D. (2006). The effects of stress on plant cuticular waxes. New Phytologist, 171(3), 469-499. doi:10.1111/j.1469- 8137.2006.01826.x Singh, V., Nguyen, C. T., Yang, Z., Chapman, S. C., Van Oosterom, E. J., & Hammer, G. L. (2016). Genotypic Differences in Effects of Short Episodes of High- Temperature Stress during Reproductive Development in Sorghum. Science, 56(4), 1561. doi:10.2135/cropsci2015.09.0545 Sullivanc, Y. & Ross. W. M. (1979). Selecting for drought and heat resistance in grain sorghum. Pages 263-281 in Stress phys~ology in crop plants, eds. Trenberth, K. E. (2008). The Impact of Climate Change and Variability on Heavy Precipitation, Floods, and Droughts. Encyclopedia of Hydrological Sciences. doi:10.1002/0470848944.hsa211 Yin, L., Wang, S., Liu, P., Wang, W., Cao, D., Deng, X., & Zhang, S. (2014). Silicon- mediated changes in polyamine and 1-aminocyclopropane-1-carboxylic acid are involved in silicon-induced drought resistance in Sorghum bicolor L. 101 Plant Physiology and Biochemistry, 80, 268-277. doi:10.1016/j.plaphy.2014.04.014 Yin, L., Wang, S., Tanaka, K., Fujihara, S., Itai, A., Den, X., & Zhang, S. (2015). Silicon- mediated changes in polyamines participate in silicon-induced salt tolerance in Sorghum bicolor L. Plant, Cell & Environment, 39(2), 245-258. doi:10.1111/pce.12521 Zhang, C., Hicks, G. R., & Raikhel, N. V. (2014). Plant vacuole morphology and vacuolar trafficking. Frontiers in Plant Science, 5. doi:10.3389/fpls.2014.00476 2.3 Exploration of marine microalgae as potential for biofuel production *K.K Musungu,1Dr. A.J Mgutu2, Dr.C.M Matoka3 and Dr. J.Odalo3 1School of Pure and Applied Sciences, Coast Institute of Technology P.O.BOX 34-80300,Voi-Kenya. 2School of Pure and Applied sciences, Kenyatta University. P.O.BOX 43844-00100,Nairobi-Kenya. 3School of Pure and Applied sciences, Technical University of Mombasa P.O.BOX 90420-80100, Mombasa – Kenya. *Corresponding author: katynach2@gmail.com ABSTRACT Fossil fuels are heavily straining the environment through green house effect resulting to Global warming which has led to continual depletion of the ozone layer. In addition environmental pollution due to fuel derived from fossil fuels has been on the rise. As a result biofuel derived from microalgae have been advocated in the recent years. Microalgae are considered advantageous for biofuel production because they do not compete with agricultural crops for arable land. Algal lipids have also been suggested due to their accumulation inside the cell at the end of the growth stage. The aim of this study was to identify microalgae that can be cultivated in vitro for environmental friendly biofuel. Sampling of microalgae was done at the Indian Ocean at the Kenyan coast. The objectives of the study included; to identify marine microalgae species that occur naturally, to investigate the effect of media on microalgae growth and to identify microalgae for large scale production of biofuel. The results indicated that Oscillatoria sp. exhibited 100% growth in three culture media namely; F/2 Guillard, Walne and Tris Acetate Phosphate (TAP) medium. Shannon weiner diversity indices indicated that Oscillatoria species was found to be present in large numbers in all the sampled stations. Lipid screening was done by Nile red staining. Further, lipid extraction by Bligh and Dyer method indicated that Oscillatoria sp.was able to produce 16% of lipid which could be converted to biofuel. The study therefore recommends this species for large scale cultivation for biofuel production because of its fast steady growth, ease to cultivate, media tolerance. Further, the amount of lipid content from it is moderate and can be bioprospected 102 Key words; Oscillatoria sp., F/2, Walne, Biofuel, in vitro INTRODUCTION Biofuel production from renewable sources is widely considered a sustainable energy alternative source compared to petroleum. Microalgae provide a viable means of environmental and economic sustainability [1]. Microalgae . Currently hold promise as an ideal third generation biofuel feedstock. This is supported by their rapid growth rate, carbon dioxide fixation ability and high lipid production capacity. They also do not compete with food crops for arable land and water resources for irrigation. They can potentially be produced on non arable land [1]Some microalgae strains are capable of generating 70% weight by weight (w/w) lipids in their biomass [2]However; they may yield significant lipid content under stress conditions. Some of the stress causing factors include; nutrients, light, salinity among others. Utilization of microalgae for biofuel production is meritorious over higher plants since they synthesize and accumulate large quantities of neutral lipids; they are also capable of all year round production yielding higher oil quantities[3].In addition, microalgae need less water than higher crops, hence reducing the load on freshwater sources [4] Microalgae are reported to exhibit faster growth rates besides expressing the ability to grow under saline conditions that are unsuitable for agriculture [2].Adoption of biodiesel will support environmental conservation, agricultural and economic developmental goals. However, a successful and economically viable algal based oil industry depends on the selection of appropriate algal strain. Species of microalgae should therefore be bioprospected to determine suitability for oil production [5]. In the last decade, Deoxyribonucleic Acid (DNA) sequencing and genomics have brought substantial changes to microalgae taxonomy [6]. Both the subunit ribosomal DNA and its genes have been used in studies for species identification because they include the highly conserved regions at the species level. MATERIALS AND METHODS Sampling of microalgal strains Water samples were collected in triplicates at the sampling stations for analysis. Surface water samples was collected using a bucket and 40 litres of water passed through 20µm mesh-size plankton net for concentration to 50ml. The resultant concentrated plankton was transferred to sample bottles labeled with date, sampling station and preserved in 5% Lugol’s solution Qualitative samples were stored after collection in a cooler box and transported to the laboratory for for further analysis. Water quality parameter Surface water quality parameters were measured in situ at each station; pH was measured using an electronic pH probe, temperature was measured using an YSI Model 550A, salinity was measured using a hand-held refractometer. 103 Quantitative sample analysis In the Laboratory, 1ml aliquots of samples preserved in Lugol’s iodine solution were mounted on slides and observed under an inverted microscope and the counts of all seen phytoplankton recorded. The algal species were identified using an identification manual of marine microalgae by Hasle and Syvertsen (1977). Isolation and purification The algal samples for qualitative analysis were subjected to purification by serial dilution, addition of antibiotics, and addition of enrichment media followed by culturing them in TAP, F/2 (Appendix iii) and Walne (Appendix iv) media for microalgae and incubated at 25 ± 1°C under 1.2+0.2 k lux intensity with 16:8 hours light photoperiod. The purity of the culture was ensured by repeated sub culturing in fresh media and regular observation under the inverted microscope. Identification of microalgal strains The purified monoalgal samples were observed under the inverted microscope and the morphological properties of the isolates identified based on the manuals. Culture maintenance Unialgal cultures of the microalgae strains were maintained in the F/2 and Walne culture media. A drop of the respective culture was inoculated aseptically using a sterile micropipette into 50 ml of the medium in sterile 50 ml conical flasks. The cultures were incubated in an algal growth room with constant illumination at 110 μmol.m-2/s at 25 °C. This procedure was performed on weekly basis Measurement of Growth Rate The growth rate of algae was measured by optical density at 660nm, 680nm, 760nm and780nm for 4weeks Daily measurements were also taken daily for 7 days. Identification of lipid producing microalgae strains Nile red staining was conducted to detect intracellular lipid droplets. Microalgae cells (0.5 ml) were collected by centrifugation at 1,500 rpm for 10 minutes and washed with physiological saline solution (0.5 ml) several times. Thereafter, the collected cells were then resuspended in the same solution (0.5 ml). Nile red solution (0.1 mg/ml in acetone) was added to cell suspensions (1:100 v/v) and incubated for 10mins. After washing once, stained microalgae cells were observed by Fluorescent microscopy. Microscopic photographs will be taken with a Nikon E600microscope Determination of oil content microalgae One ml of growing algae in F/2 media was collected in 3 replicates in 1.5ml tubes and centrifuged at 13000rpm for 5 minutes. The pellets were frozen in liquid nitrogen and stored at -800C.Thereafter; the pellets were measured to determine the weight. The 104 lipids were extracted as follows; in the frozen pellets, 200µl of a mixture of chloroform: isopropanol (1:1) was added and vortexed vigorously for 3 minutes. They were then centrifuged at 13,000rpm for 5minutes .The supernatant was transferred to a new tube, and then the pellets were re-extracted with 500µl of hexane and; vortexed vigorously for 3 minutes. The samples will then be centrifuged and the supernatants combined. The supernatants were dried by an evaporator and the amount of lipids measured gravimetrically .The Bligh and Dyer method of Extraction using Chloroform and Methanol was also used [7]. RESULTS Description of study site Sampling was done three times at the Tudor creek. .Tudor creek is located Latitude of 4.0000°, Longitude. 39.6500°. It is 15.5km away from Mombasa town. The study site was accessed by a fibre glass boat at the Tudor Water sports entrance. Sampling was done in triplicates; one set was preserved with 5% lugol’s iodine for quantitative diversity studies. The other set was other set was sorted, isolated and cultured, while another set was incubated at 280C for fourteen days Temperature, pH, total suspended solids and salinity measurements for all the sampled stations were recorded. Physico-chemical Characteristics The mean surface Water Temperature was 27.750C with a standard deviation of 0.5 and standard error of 0.25. The highest average temperature for the Creek studied was 310while the lowest average temperature was 210C (Table i). Analysis of variance reveals that there was no statistically significant variation in Water and Temperature between sampling stations and between the three months of sampling at (P>0.05) for the pH was observed during the study period, with the highest average value of 8 and a lowest of 7.8. The mean ± SD for pH is 7.25 ± 0.23 (Table i). This observed variation was statiscally significant .The surface Water Temperature ranged from 270C to 280.The highest temperature was recorded at Coast general station and Technical University while the lowest temperature was recorded at Mkomani station. The pH was observed during the study period, with the highest average value of 8.0and a lowest of 7.8The salinity ranged from 34-36.3 PSU. With the highest concentration at Mkomani station. Total Suspended solids ranged from 0.037-0.162g/l with the highest concentration being recorded at Fort Jesus. 105 Table 1: Physico-chemical parameters of the sampled stations Parameter TSS Salinity Temp Ph Units g/l Psu oC - Mean 0.07 35 27.75 7.85 Standard Error 0.03 0.48 0.25 0.05 Median 0.05 34.85 28 7.8 Standard Deviation 0.06 0.96 0.5 0.1 Variance 0.003 0.92 0.25 0.01 Range 0.125 2.3 1 0.2 Maximum 0.162 36.3 28 8 Minimum 0.037 34 27 7.8 Quantitative sample analysis This was determined by observing the samples preserved by lugol’s iodine under the Leica Inverted microscope. Species present The species were identified morphologically using conventional means by use of the phytoplankton guides. The species observed are shown in table 4. The species present in all the seven stations were; Oscillatoria spp., Thallasiosira sp., Coscinodiscus sp., and Nitzschia. The species present in only one station were; Monidiscus sp,Coscinodiscus sp, ,protoperidinium sp ,Dictyliosolenia sp, ,Bleaklayer sp.,Protocentrum sp.,Hemidiscus sp , Anabaena sp. , Dactyliosolen fragilissimus, Cylindrotheca closterium, Gonyaulax spinifera, Spatulodinium sp. , Nitsia longissima, Anacustis nidulans. The most abundant phytoplanktons are in the group of Bacilliarophyta(diatoms) which occupied 60% followed by the Cynanophyta(cyanobacteria) which occupied 30%and chlorophyta occupying 10%.Bacillariophyta have been reported by many authors to be dorminat in the phytoplankton composition as it is in the present study [8]. Chlorophyta was the second group after Bacillariophyta in the number of identified species, these result also agree with study on Grand River in Oklahoma by [9]. It is also in agreement with other studies in Iraq [10]. The maximum occurrence of phytoplankton was in (October 2014) and then decreased. This may be due to available nutrients and other physical and chemical factors which promote growth of phytoplankton. While the minimum total number of phytoplankton species was recorded at moroto which might be due to domestic discharge, effluents from run-off that empty into the river, this corresponds to the work of [11] .The differences in number of taxa and number of individuals between sampling stations for each class of phytoplankton may be due differences in temperatures and pH as different species obtain at different pH and temperatures[12]have suggested a relationship between 106 species diversity and pollution status of aquatic system and classified as follows; > 3 = Clean water, 1-3 = moderately-polluted < 1 = Heavily polluted. Water pollution levels can also be accurately identified by analyzing the species abundance, physiological, biological responses and residue contents [13]. However, algae may not only be significant for biomonitoring studies but could also be a useful phytoremediation technology to restore water quality due to their high bioaccumulation ability 1.3.2 Shannon Wiener Diversity indices Computation of diversity indices was also done. Resultant graphs were drawn as shown in figure 1 Phytoplankton Abundance and Diversity Diversity indices reveal that TUM station had the highest index of 16.47.For all the diversity was stable. This indicated that there is a large abundance of phytoplankton in the marine ecosystem sampled. Fig 1: Comparative diversity indices of the sampled stations 107 The highest diversity occurred during the second sampling which indicates very high diversity indices from the graphs shown. Isolation and purification of microalgal cultures This was done through isolation of the microalgae and culturing them on solid and the three liquid media namely; TAP, Walne and F/2 Liquid Media preferences Species occurrence was highly dependent on the media, the media used were TAP,Walne and F/2.Plots of media and the species obtained are shown in figure ii Culture of microalgae in liquid media The isolated microalgae were then cultured in test tubes and conical flasks of 250 ml with the suitable media being supplied to each of them. Continuous and vigorous aeration was provided to culture which keeps the culture in suspension. In addition continuous aeration is also helpful for uniform distribution of nutrients. Figure ii:Line graph showing a graph of media against media From the graph it shows that F/2 and Walne Media are the most prefered media for growth of most microalga Total Isolates obtained At the end of the isolation process, there were various species obtained. Isolation was done based on stations, and species. Liquid Media preferences Species occurrence was highly dependent on the media, the media used were TAP,Walne and F/2.Plots of media and the species obtained are shown in figure i 108 The methods used for isolation included; serial dilution, culturing in liquid and solid media. The isolation protocol developed in-house resulted to be an efficient method for the microalgae isolation and transfer from the natural environment in to laboratory conditions. The streak plate method for microalgae enrichment, although slow, proved to be an excellent approach for the isolation of green phototrophic microorganisms. With regards to the physical properties of the water samples, they remained fairly constant all throughout the different depths assessed as well as in the different locations. Temperatures fluctuated between 20.5 and 23.50C, pH between 6.9 and 8.1, and salinity between 32 and 37.8 psu.Regardless of the stations the microalgae colour appearance on the solid media included; 15% green, 16% brown, 8% pink, 23%cream.This results corresponds with the study from [12].Isolation procedures using solid media allowed for the selection of microorganisms with colors besides green; culture attempts of the same in liquid media were utterly unsuccessful. None of the solid or liquid media were prepared with an external carbon source, with the purpose of selecting phototrophic organisms. However, solid media was gellied with agar, which is a polymer of the sugar galactose. However, certain marine organisms have the capacity of producing agarose and liquefy solid media efficiently [13].These enzymes are responsible for allowing them to use agar as their primary carbon source and enables their ability to thrive in the ocean .In this case, it might have happened that the organisms isolated, other than the green ones, had this innate ability to metabolize agar and thus developed efficiently in solid media; interestingly, all of them showed growth close to the agar surface. As for the liquid Media preference, F/2 Media with silicate seemed to be the most preferred because there was a large growth of the species.TAP media was the least preferred probably due to the fact that it forms a cloudy appearance after a short while. These results are consistent with observations made by [9] and many others who noticed the zonation of pigments on marine organisms depending on depth. 1.5.5 Measurement of growth Optical density was taken at intervals of one week. The absorbance readings were taken at three optical densities namely; Optical densities of; 620nm, 650nm and 750 nm. The results are shown in figure 2. Figure 2 :Graph showing Optical densities on progressive days 109 Total lipid content of selected isolates using conventional extraction methods. The results revealed that Oscllatoria spp.had the highest percentage of total lipid of 16+ 0.5 these results were in harmony with the results of [14] who stated that oscillatoria has total lipid content 16.5%. Two diatoms species were investigated in this study Nitzschia linearis and Navicula cuspidata. The two species showed relatively low percentage of oil 6.1% and 10.2% for Nitzschia linearis and Navicula cuspidata respectively. Nitzschia sp. was investigated by [4] and found to have lipid percentage of 45-47%. On the other hand [15] found that Nitzschia frustulum has a total lipid percentage of 13.9%. This percentage near to that presented in this study in table ii Table 2:Percentages of total oil content in the microalgae studied Microalgae isolate Total lipid content % Oscillatoria sp 1 11+ 0.4 Oscillatoria sp.2 16+0.5 Nitzschia lineaus 6 + 0.1 Navicula cuspulata 10+ 0.2 Intracellular lipid droplets of were observed by Nile Red staining under fluorescent microscope with excitation at 450–490-nm and emission at 515-nm [16].Neutral lipid or triglycerides appeared as yellow dots, whereas polar lipid and chlorophyll were stained in red colour cells were observed by Nile Red staining under fluorescent microscope with excitation at 450–490-nm and emission at 515-nm.Nile red staining: Nile red (9- (Diethylamino) -5H benzo [α] phenoxazin- 5-one) staining is specifically used to identify and confirm the intracellular lipid droplets from the biological samples [17]. 110 The results indicated that not all algal species could be affected by Nile red staining since oil droplets were not clear and the whole cells were stained in red. This was the situation with green isolates, while certain blue green isolates were affected by the dye where the yellow stained parts were clear. Referring to Oscillatoria sp 1 and Oscillatoria sp 2 the cells showed yellow florescent color under florescent microscope even without adding the dye. So this gives false results when referring to lipid content. Diatoms were stained well Diatoms were stained well with the dye and the oil drops were clear. Since the staining method may not be accurate The florescent method has been successfully applied to the determination of lipids in certain microalgae, but has been unsuccessful in many others, particularly those with thick, rigid cell walls that prevent the penetration of the dye [18].Since Nile red method was not accurate in determining lipid content in microalgal cells so lipid content was determined using conventional extraction method using two organic solvents. DISCUSSIONS AND CONCLUSION It was therefore observed that physic-chemical parameters have a major role to play in species occurrence.The method of lipid extraction, type of media and level of axenicity of the microalgal species determines the percentage of lipid extracted .The Oscillatoria species seems to be a hardy species and therefore still survives when subjected to stress it is therefore recommended for large scale cultivation in photobioreactors for biofuel production.Stress conditions such as nutrient stress results to production of more lipids from the microalgae.Nile red staining is not so accurate therefore should be supplemented by other methods of lipid determnination.The choice of algal strain,the method used for culture and the location of sampling highly determines the amount of lipid produced by microalgae. The strains used for large-scale algal biofuel production need to be improved through selection and genetic approaches. Breakthroughs and innovations in areas such as increasing the capability of algae to use nutrients efficiently or engineering designs to reduce processing requirements have the potential to greatly improve the energy balance and enhance the overall sustainability of algal biofuels. Acknowledgement This work was funded by the National council of science and technology.We thanks the management of Kenya Marine fisheries research headquarters in Mombasa for allowing the research to be carried out in their laboratories.we would also acknowledge Mr mwendwa Mbaluku for the technical assistance he provided. 111 References D.Giuliano,F.Bruno,A.António,andA.José,and A.Texeina. “Third generation biofuel from microalgae,”Journal of Biotechnology letters, Vol.22,2010,pp.2014-2020. A.Alcaine, A.Anna, andT.Kughiga, “Biodiesel from algae,” Journal of Chemical engineering. Vol.10,2010,pp.2713-2716. L.U.Abubakar, A.M.Mutie, E.U.Kenya, and A.Muhoho “Characterization of Algae Oil (Oilgae) and its Potential as Biofuel in Kenya,” Journal of Applied Phytotechnology, Vol.4, 2012, pp.147-153. Y.Chisti, “Biodiesel from Microalgae, “Journal of Biotechnology Advances, vol.20, 2007, pp.294-306. G.S.Arujo,L.J.Matos,L.R.Gonçalves,F.A.Fernandes, and W.R.Fatias,” Bioprospecting for oil producing microalgal strains: Evaluation of oil Biomass production for ten microalgal strains,”Journal of BioresourcesTechnology,Vol.102,2011pp.5248- 5255 R.A.Abou-Shanab, J.H.Hwang, Y.Cho, B.Min, and B.H.Jeon, “Characterization of microalgal species isolated from freshwater bodies as a potential source for biodiesel production,” Journal of Applied energy, Vol.88, 2011pp.3300-3306. E.G. Bligh and W.J.Dyer, “A rapid method for total lipid extraction and purification.” Journal of Biochemical Physiology, Vol.37, 1959 pp. 911–917 S. Polat and A. Aka, “Total and size fractionated phytoplankton biomass off Karatas, northeastern Mediterranean coast of turkey,” Journal of Black Sea/ Mediterranean Environment,”Vol 13, 2007pp. 191-202 L.A. Pfiester, R. Lynch and T. L. Wright, “Species composition and diversity of Phytoplankton in the grand River dam area Oklahoma,” Proceedings of the Oklahoma Academy of. Science,Vol 60, 1980, pp. 63-68. A.Y. Al-Handal, S. A. Al-Assa and M. A. Al-Mukhtar, Journal of Marine Science Center University. Of Basrah, Vol 4 ,1989, pp.67. M. H. Fikrat, F. K. Nuha and H. H. Falah, “Effect of Chemical and Physical Properties of River Water in Shatt Al-Hilla on Phytoplankton Communities,” E-Journal of Chemistry, Vol5, 2008, pp. 323-330. J.L. Wilm and T.C. Dorris, “ Species Diversity of Benthic Macroinvertebrates. In: A Stream Receiving Domestic and Oil Refinery Effluents. In: Islam, S.M. Phytoplankton diversity index with reference to mucalinda serovar Bodh- Gaya,”12th World Lack Conference., Taal, India, 2007. W.Chen, M.Sommerfeld, andQ.Hu, “Microwave-assisted Nile red method for in vivo quantification of neutral lipids in microalgae,” Journal of Bioresources. Technology, Vol.102, 2011, pp.135–14. Y.Chisti,“Feedstocks for second generation biodiesel,” Journal of Microbiology and biotechnology,”Vol.20,2010pp.3135-3142. M.Sayeda,AEntesar,S.Abo,R.ElEnin,D.S.ElDin,G.El Diwani, and A.Gamila, “Quantitative and qualitative determination of lipid content in Microalgae 112 for biofuel production,” Journal of Algal biomass 2014,Vol.5(3),pp.23- 28. T.Matsunaga,M.Matsumoto,Y.Maeda,H. Sugiyama, R. Sato,and T. Tanaka(2009). “Characterization of marine microalgae, Scenedesmus sp. Strain JPCCGA0024 toward biofuel production,”Journal Biotechnology Letters.Vol 31,pp.1367-1372. P.Greenspan, E.P.Mayer,and S.D.Fowler, (1985). “Nile Red" a selective fluorescent stain for intracellular lipid droplets,” Journal of cell biology vol.100, 1985.pp. 965- 973. P.Held, “Monitoring Algal Cell Growth Using Absorbance and Fluorescence,”BioTek Application Note, www.biotek.com.2011. S.M.Renaud,L.V.Thinh,and D.L.Parry,“The gross chemical composition and fatty acid composition of 18 species of tropical Australian microalgae for possible use in mariculture Aquaculture,”Vol.170,1999,pp.147-59. P.Cho, “Valuable products from Biotechnology of microalgae,” Journal of Applied Microbial Biotechnology,Vol.65,2004,pp.635-648. L.T.Dinh,Y.Guo,and M.S.Mannan, “Sustainability evaluation of biodiesel production using multicriteria decision-making,” Environmental. Programme for sustainable Energy,vol.28,2009, 38–43. M.L.Eltgroth,R.Watwood, and G.V.Wolfe, “Production and cellular localization of neutral long-chain lipids in the haptophyte algae Isochrysis galbana and Emiliania huxleyi”Journal of Phycology. Vol.41,2005,pp.1000–1009. G.Hu,MSo.mmerfeld,M. Ghiradi,M.Posewitz, M.Seibert, and A. Darzins, “Microalgal triacylglycerols as feedstocks for biofuel production: Perspectives and Advances,”The Plant Journal,vol.54,2008.pp.621-639. G.Khola,and B.Ghazala,“Biodiesel production from Algae,”Journal of Biotechnology letters,vol. 44,2012.pp.379-381. Y.Liu,,X.Yeusong,,P.Ying,,Y.Rongqing,,Q.Dairong, C.Yi, and C.Yu,(2014). “The biodiversity of oleaginous microalgae in Nothern Qinghai-TibetPlateau,”Journal of Microbiology Research,Vol 8,2014,pp. 66-74. 113 3.Parallel Session 2 : Technology diffusion impact and sustainable Agribusiness value chains 3.1 Effect of audience simulation on the knowledge of dieback disease management among cocoa farmers in Ife East Local Government Area of Osun state, Nigeria Olajide Bamidele. Rasak1 and Oladipupo Olasunbo Kafila2 1&2Department of Agricultural Extension and Rural Development, University of Ibadan, Corresponding Address: r.olajide@gmail.com ABSTRACT Tonnes of annual cocoa production are lost to pest infestation. Several interventions have been employed to promote alternative pest management for sustainable pest management. This paper examines the potency of audience simulation experimentation on management of dieback diseases of cocoa in Ife East Local Government Area of Osun State, Southwestern, Nigeria. A multistage sampling procedure was used to select 120 cocoa farmers. Data was obtained on farmers’ socioeconomic and enterprise characteristics as well as knowledge of dieback disease management before and after exposure to audience simulation. Results show that farmers averaged 60 years old, mostly male (65.0%) and with no formal education (51.7%). Most respondents (66.7%)) were favourably disposed to the use of audience simulation in pest management information dissemination. There was a general improvement in respondents’ knowledge level of issues on traditional management of dieback. In addition, more farmers correctly answered questions on environmental friendly practices such as prevention through keeping cocoa seedling in the nursery (75.8%). It is concluded that communication intervention to promote environmental friendly pest and disease management practices could be one of the drivers of transition to alternative pest and disease management. Development agencies should establish community viewing and broadcast centers where extension clientele can be reached in groups. Keywords: Audience simulation, dieback diseases, Pest and disease management BACKGROUND AND PROBLEM STATEMENT Pest and disease management has played its role in doubling food production in the last 40 years, but pathogens still claim 10–16% of the global harvest (Charkraborty and Newton, 2011). According to the Food and Agriculture Organization of the United Nations (2001), 20 to 40 percent of the world's potential crop production is already lost annually because of the effects of weeds, pests and diseases. Specifically, tonnes of potential annual cocoa production are lost to pest infestation. In the view of Padi and Owusu (2001), cocoa is perhaps the most important export crop for countries in the forest zone of West Africa and is the main source of foreign exchange for some of them. The cocoa industry in West Africa is dominated by a large 114 number of peasants who cultivate small farms of about 0.5 to 5.0 hectares and who lack the resources (including access to information) to expand or improve their farms. This and other factors combine to plague the industry with serious problems leading to dwindling production levels. Since the conscious effort of agricultural research is to move productivity from a lower level to a higher level in order to increase the income of farmers and enhance agricultural development (Farinde and Ajayi, 2005), it is therefore important to constantly inform the farming populace about the importance of pest and diseases management. Just as information is vital in every other facet of life, agricultural information is a key component in improving small-scale agricultural production and linking increased production to remunerative markets (Yahya, 2009), thus leading to improved rural livelihoods, food security and national economies. According to (Munyua 2000) cited in Obidike (2011) the rural urban drift in recent decades in sub- Saharan Africa could be attributed to rural farmers’ lack of access to information that would help them achieve maximum agricultural yield, hence they are frustrated. Obidike (2011) points out that the least expensive input for improved rural agricultural development is adequate access to knowledge and information in areas of new agricultural technologies, early warning systems (against drought, pests and diseases outbreak) and improved seedlings. According to Phrampus, Metro and Williams (2009), human patient simulation; a subset of radio and animation has emerged as a powerful educational modality across the continuum of learners. Simulation is used to demonstrate physiologic principles, physical examination skills, along with bringing topics to life with regards to disease process or variants of physiology. Like other methods that work well with adult learners, it is typically used and most effective in very small groups. A simulation of a system is the operation of a model of the system. In its broadest sense, simulation is a tool to evaluate the performance of a system, existing or proposed, under different configurations of interest and over long periods of real time (Anu, 1997). Many radio agricultural programmes targeted at farmers are regularly and continuously aired to bring farmers up to beat on new innovation and technologies (Olajide and Amusat, 2012) to bridge information gap occasioned by traditional prints and library based information sources. These radio programme have been evaluated to generate feedback on effectiveness or otherwise of such (Yahaya and Badiru, 2002; Olajide 2011; Olajide and Amusat 2012). However, previous attempts of evaluation of radio broadcast were based on the assumption that farmers are regular listeners of these radio programmes. This has treated farmers as homogenous groups with no difference in needs and has left them with no option than to make do with whatever information they gather from around them; adequate or otherwise. Little wonder, the desirable impact for which information are disseminated are mostly not achieved. In view of the position above and considering that much as information is an essential ingredient in agricultural development programmes, farmers seldom feel the impact of agricultural innovations either because they have no access to such vital information or 115 because it is poorly disseminated occasioned by wrong choice of channels. For a significant level of intended change to be achieved, therefore, access to required information and ability to utilise such places a high premium on the channel used. It is against the aforesaid that audience simulation experimentation with information on pests and diseases that affect cocoa as observed in the study area were of interest in this study. The study was premised on the following specific objectives: Identify the socio-economic characteristics of cocoa farmers in the study area; Examine the enterprise characteristics of cocoa farmers in the study area; Ascertain cocoa farmers’ attitudes to the concept of audience simulation for information on disease management dissemination; and To determine changes in knowledge of dieback disease management of cocoa farmers before and after exposure to audience simulation. METHODOLOGY The study was undertaken in Ife East Local Government Area (LGA) of Osun state, souhwestern, Nigeria with an area of 9,251 square kilometer and a population size of 34,137,627 ( NPC, 2006). The population for this study was cocoa farmers in the LGA. A multistage sampling procedure was used for selection of respondents. In the first stage, purposive sampling technique was used to select Ife-East LGA because cocoa farming is prominent in the LGA. In the second stage, one ward known for highest cocoa output in the LGA was chosen. In the third stage, three farming communities were selected randomly from the selected ward. In the fourth stage, from each community, 40 cocoa farmers were systematically sampled from every other household. From each selected households, one person who has been in cocoa production in the last ten years was interviewed resulting in 120 respondents from the three communities The study used a quasi experimental research design. The study was undertaken in three phases with the first phase involving collection of baseline quantitative data on farmers’ knowledge of cocoa dieback disease management. Thereafter, farmers were exposed to audience simulation through a tape recording of management practices of the cocoa dieback disease. This was carried out by developing messages on management practices of dieback disease of cocoa in form of audio and selected farmers were organized in groups to listen continuously to the recorded message for three weeks. In the third phase, post-exposure quantitative data on the dieback disease management was collected after three weeks of continuous exposure to messages from the same groups to determine any change in the knowledge of respondents in management of dieback disease of cocoa. This was to ascertain the impact of the intervention. The dependent variable of this study was the change in knowledge of dieback diseases management among cocoa farmers. The change in knowledge was generated from difference between mean scores of the pre and post knowledge (measured with 20-item knowledge statements) scores of the farmers. Response options for the questions were Yes and No. An increase in knowledge was expressed with higher mean score of post intervention compared to pre-intervention mean score. Data obtained were analysed 116 using both descriptive (frequencies, percentages and mean) and inferential (t-test) statistics at p=0.05. RESULTS AND DISCUSSION Respondents’ socioeconomic and enterprise characteristics The age distribution of the respondents ranged between 18 to 88 years with the mean age at 59.01±16. More than one-third (35.0%) of the respondents fell between the ages of 50-65 years, while only about 6.7% were less than 33 years of age; as shown on Table I. This shows that majority of the farmers are old and less energetic which explains the reason for hired labour by respondents. This is in line with Gul Unal (2008) that old age might pose a disadvantage in agriculture because most of the work is physically demanding and older folks might be too conservative to try new and more efficient technologies. Data further reveals that while males were 65.0% of the respondents, females were 35.0%. This suggests that more males than females are involved in cocoa production in the study area. This supports the findings of Durodola (2014) who reported that there are more males than females in agriculture. This also agreed with the findings of Olujide and Adeogun (2006) in assessment of cocoa growers’ farm management practices in Ondo state, Nigeria which confirms that cocoa farmers are mostly male. . Information on respondents’ educational level reveals that only 17.5% of respondents had up to secondary education. A higher percentage of the respondents (51.7%) had no formal education, while only 2.5% of respondents had tertiary education. This implies that majority of the respondents had no form of education. This corroborates the view of Kumar and Quisumbing (2010) that in sub-Saharan Africa many farmers have low level of education and that improving their education would probably increase agricultural productivity and reduce poverty. Table I further show that 45.9% of the respondents had been engaged in agricultural production for more than 23 years, while 29.2% of the respondents had been in farming for 23years or less. This supports the findings of Ugwuja et al (2011) who found that majority of farmers have more than 15 years of farming experience. Furthermore, result shows that majority (66.7%) of the respondents had less than 5 hectares while less than 1.0% had up to 16 hectares of land for cocoa production. This is in agreement with Padi and Owusu (2001) who pointed out that majority of cocoa farms in West Africa are small holdings owned by a large number of peasant farmers. The result on farm outputs shows that 40.8% of the respondents recovered up to 10 bags of cocoa and 49.2% of the respondent made output of between 11 and 20 bags in the season preceding the year of this survey and only 9.2% produced 21 to 30 bags of cocoa. This implies that cocoa production is affected greatly by a number of challenges which include dieback disease especially if one considers size of farm holdings of these farmers. Result on number of farm locations shows that 51.7% of the respondents had two farm locations and only 25.0% had more than three farm locations. Results on sources of farm labour show that 93.3 % of the respondents engaged hired labourers for farming while 117 6.7% use family members as their source of labour. This is possible because higher percentages of the respondents are old and perhaps are less active. Also, 95.0% of the respondents’ source of finance was self while 1.7 % of respondents got financed by family and only 0.8% source finance from cooperatives. Table 1: Distribution of cocoa farmers based on their socioeconomic and enterprise characteristics Variable Categories Frequency Percentage Age (years) 18-33 years 1 6.7 34-49 years 46 20.8 50-65 years 40 35.8 66-81years 22 31.7 >81 13 5.0 Sex Male 78 65 Female 42 35 Education No formal education 62 51.7 Primary education 34 28.3 Secondary education 21 17.5 Tertiary education 3 2.5 Farming experience (Years) 2-23 years 35 29.2 24-45 years 30 25.0 46-67 years 38 31.7 68-89 years 17 14.2 Farm size Less than 5ha 6-10ha 11-15ha 16-20ha 80 34 5 1 66.7 28.3 4.2 0.8 Number of farm location 1 2 3 More than 3 25 62 3 30 20.8 51.7 2.5 25.0 Last season output Less than 10bags 11-20bags 21-30 bags More than 30 bags 59 49 11 1 49.2 40.8 9.2 0.8 Source: Field survey, 2014 118 Attitude of respondents towards the concept of audience simulation The favourable disposition of farmers to use of audience simulation approach is not in doubt as they responded positively to most of the issues raised in their attitudinal disposition to the broadcast method. For instance, while majority (75.8%) were affirmative on social desirability of audience simulation, 65.0% also agreed that messages from audience simulation may be more easily understood (Table II). The farmers could be coming from the background of what they stand to gain in listening to a broadcast that addresses their problem in groups. It might also be a forum to proffer solution to common problem that confronts them as well as another opportunity to create an alliance to combat the disease frontally in group. From the position of majority who opined that the message could be easily understood, it is given that debate and discussion will ensue soon after the broadcast is over. This will no doubt engender better understanding as individual farmer seeks clarifications (Adesina and Zinnah (1993). . Furthermore, while a reasonable proportion (53.3%) were in doubt about whether the simulation broadcast methodology is for small scale farmers, 50.8% debunked the fact that it offers no room for feedback while more than two-third (69.2) equally disagreed that they will easily forget message received from audience simulation channel. The farmers’ position of doubt of audience simulation being for them is understood considering neglect they experience on regular basis from several facets of their livelihood. The latter optimism expressed in respect of feedback opportunity as well as retention of messages thereof, perhaps, was reinforced by the conviction of 59.2% of the respondents that audience simulation provides direct contact to object of dissemination, therefore makes it sound more like live broadcast. Again, while 59.0% believed that audience members’ composition especially mixed audience will impact the use of audience simulation, 73.3% were not in any doubt that participation in audience simulation will help in better management of pest and diseases on their farms. The positive rating of audience simulation by this set of respondents was reinforced by 76.6% who unequivocally asserted that audience simulation is a good communication approach and that its message could enhance their understanding of dieback disease. In all, it can be safely concluded that most cocoa farmers (66.7%) are favourably disposed to the use of audience simulation in disseminating diseases and general cocoa management information. 119 Table II: Distribution of respondents based on their attitude towards the concept of audience simulation n=120 Statement SA A U D SD 1 Audience simulation is socially desirable 75.8 7.5 15.8 0 .0 0.8 2 Message is easy to understand with audience simulation 65.0 20.0 15 0 .0 0 .0 3 Audience simulation is not for small scale farmers 1.7 14.2 53.3 15 15.8 4 Audience simulation does not give room for feedback 5.0 11.7 32.5 38.3 12.5 5 I easily forget messages from audio audience simulation 1(0.8) 3.3 26.7 26.7 42.5 6 Audience simulation provides direct contact to object of dissemination, therefore makes it sound more like live broadcast 25.0 34.2 39.2 1.7 0.0 7 Audience members’ composition especially mixed audience will impact the use of audience simulation. 59.0 22.5 18.3 0.0 0.0 8 My participation in audio simulation will help in managing pest and diseases on my farm better 73.3 0.0 9.2 15.8 10 9 Audio audience simulation is not a good communication approach 2.5 2.5 18.3 8.3 68.3 10 Audience simulation further compounds problem of understanding messages. 1.7 0.0 15.0 11.7 71.7 Overall attitudinal disposition Favourable 66.7 Unfavourable 33.3 Source: Field survey, 2014 Figures are percentages Farmers’ knowledge of dieback disease management practices From available data in Table III, there was a general remarkable improvement in respondents’ knowledge level in the post intervention period of simulation. Notable among items that recorded at least two-third improvement in the post intervention performance of the farmers were the realisation that dieback disease affects cocoa from seedling (94.2%), the effect is felt during dry season (75.0%), dieback kills cocoa stems (90.8%), internal discoloration of stems (91.7%) and prevention through keeping cocoa seedling in the nursery and free of moisture (90.%) as against 36.7%, 27.5%, 29.2%, 38.3% and 38.3% of respondents who got these correctly in the pre intervention period. However, two items recorded a drop in the performance of farmers in the post test period. Though insignificant, more respondents (63.3%) and (39.2%) respectively, correctly answered the questions on potency of dieback disease both at the nursery and field levels as well as prevention of dieback through proper 120 spacing during the pre-simulation intervention compared to 62.5% and 37.5%, respectively, in the post intervention phase. The result on proper spacing leaves so much to be desired as proportion that got it right at both pre and post intervention era were abysmally low. This might be due to deep rooted belief in the traditional practices that the farmers are familiar with and had stuck to. While the result above may leave one in doubt about the effect or impact of the simulation on respondents’ knowledge of dieback disease, the picture painted by the set of data on Table IV reflects the significance of the simulation intervention. While less than a quarter of the respondents (22.5%) had high knowledge of the dieback disease in the pre-intervention phase, the proportion of respondents with high knowledge (52.5%) was more than double those that had high knowledge in the pre-intervention period. In concrete term, whereas, mean score was 10.8 in the pre-intervention phase, a mean score of 15.1 was recorded in the post-intervention stage. 121 Table III: Distribution of respondents by pre and post audience simulation knowledge of dieback disease management Item Before % After % The mortality rate of dieback on infected farm could be total. 76.7 95.8 Dieback does not affect cocoa tree at seedling stage of development. 36.7 94.2 The younger the seedling at the time of infection, the greater its chance of being killed. 71.7 83.3 Dieback can affect trees both in the nursery and on the field. 63.3 62.5 Dieback only affects older cocoa trees. 31.7 68.3 An infected tree starts by showing some little yellow spots on the leaf at the tip of the tree. 59.2 70.8 Dieback affects cocoa in all seasons. 56.7 77.5 The effect of dieback is not felt during dry season. 27.5 75.0 Colour changes of cocoa leaves from green to yellow are not dieback symptoms to look out for. 70.0 75.8 Spotted leaves are one of the critical symptoms of dieback. 54.2 92.5 Dieback cannot kill cocoa stem. 29.2 90.8 Dropped leaves signify advanced dieback infection. 90.8 92.5 Internal discoloration of stem occurs in trees affected by dieback? 38.3 91.7 Planting of disease free seedlings cannot prevent dieback. 48.3 85.0 Ensuring that prepared field for cocoa planting is disease free is a means to prevent dieback outbreak. 58.3 89.2 Keeping cocoa seedling in the nursery and free of moisture can help prevent dieback. 38.3 90.0 Mealy bug is a suspected vector of dieback. 52.5 90.0 Ants crawling up cocoa tree and building soil tunnels may play host to sources of diseases. 50.8 92.5 Adequate shading in the nursery is a measure to prevent dieback. 45.8 75.8 Adequate or proper spacing will not prevent dieback. 39.2 37.5 Source: Field survey, 2014 Table IV: Categorization of respondent’s level of knowledge before and after the simulation (n=120) Level of knowledge High Low Mea score Before 27 (22.5) 93 (77.5) 10.1 After 63 (52.5) 57 (47.5) 15.1 Change 66 (55.5) 54 (45.0) 122 Test of difference between knowledge of dieback disease in the pre and post simulation intervention The result of the inferential statistical analysis of test of difference between the pre and post intervention phases of the simulation (Table V) indicates that there was a significant difference between knowledge possessed by respondents in the two periods. This is in tandem with the findings of Olajide and Thomas (2011) and Olajide (2011) that establish a positive impact of radio entertainment – education program on sexual practices of selected farm families in Nigeria. It is also in agreement with the position of Gershon, Rinku and Jaime (2004) in their study of the acquisition and diffusion of knowledge in pest management in Indonesia. Table V: Paired sampled t-test between the knowledge before stimulation and the knowledge after stimulation. Hypothesis Mean SD Df t-value p-value Decision Difference in knowledge before and after 14.50 4.36 119 -6.25 0.000 Significant difference Level of Significance=0.05 CONCLUSION AND RECOMMENDATION Evidences from results of this study suggest that cocoa farmers in the study areas are mostly male, old, vastly experienced in cocoa farming but with small holdings in scattered locations. They were mostly favourably disposed to the audience simulation broadcast method and benefited greatly as the audience simulation recording enhanced their knowledge of the dieback disease. It is recommended that all stakeholders, government agencies and non-governmental establishments that are reaching out to farmers should put the audience simulation to use. This can be achieved by establishing community viewing and broadcast centers where extension clientele can be reached in groups. REFERENCES Adesina A.A and Zinnah M.M., (1993). Technology characteristics, farmers’ perceptions, adoption decisions: A tobit model application in sierria leone.Agricultural Economics 9.297-311. Anu, M. (1997). Introduction to modelling and simulation. In S. Andradóttir, K. J. Healy, D. H. Withers, and B. L. Nelson( Eds), Proceedings of the 1997 Winter Simulation Conference (pp 7-8). Binghamton, NY 13902-6000, U.S.A. Crop life America (2013) Increasing food production. http:www.croplifeamerica.org/crop- protection/benefits/increase-food-production. Retrieved June 21, 2014 123 Chakraborty, S. and Newton, A.C, (2011) Plant pathology, climate change, plant disease and food security. An overview. Retrieved September 23, 2013 from http://onlinelibrary.wiley.com/doi/10.1111/j.1365- 3059.2010.02411.x/abstract Durodola, M. I. (2014) Farmers’ level of utilisation of agricultural information from Agbe afokosoro radio farm broadcast in Ogun state. Unpublished Masters Dissertation, Department of Agricultural Extension and Rural Development, University of Ibadan, Ibadan, Nigeria 119pp Farinde and A.O. Ajayi, (2005). Training Need of Women Farmers in Livestock Production Implication for Rural Development in Oyo State of Nigeria. Journal of scientific Research.,10:159-164. Food and agricultural Organization / Technical centre for Agriculture (FAO/CTA). 2001. First International workshop on farm Radio broadcasting; Rome, 19–22 February, 2001. Information and communication technologies Servicing farm radio: New contents, new partnerships. Final Report. Rome: FAO. Gershon F., Rinku M. and Jaime B. Q (2004) The Acquisition and Diffusion of Knowledge: The Case of Pest Management Training in Farmer Field Schools, Indonesia; Journal of Agricultural Economics Volume 55(2) pp221-243 Gul Unal, F (2008). Small is Beautiful; Evidence of an Inverse Relationship between Farm Size and yield in Turkey. Working paper No.551, The Levy Economics Institute P.O.BOX5000 Annandale-on-Hudson, NY 12504-5000 Munyua, H. (2000). Application of information communication technologies in the agricultural sector in Africa: a gender perspective. In: Rathgeber, E, & Adera, E.O. (Eds.) Gender and information Revolution in Africa IDRC/ECA. Pp. 85- 123. Obidike .N.A. (2011). Rural farmers’ problem accessing agricultural information http://www.webpages.uidaho.edu/~mbolin/obidike.htm Retrieved 12th July, 2013 Olajide B. R. and Thomas K. A (2011) “The Impact of Radio Entertainment – Education Program on sexual practices of selected farm families in Ebonyi State’’ Ozean Journal of Social Sciences Vol. 4(1) 23-35pp Olajide B. R. (2011) Assessment of Farmers’ Access to Agricultural Information on Selected Food Crops in Iddo District of Oyo State, Nigeria. Journal of Agricultural and Food Information Vol. 12 (3-4) 354-363 Olajide B. R. and Amusat A. S (2012) Perceived Efficacy of Radio Agricultural Commodities Trend Programme among Farmers in Oyo State, Nigeria. Journal of Media and Communication Studies Vol. 4 (3) 46-51 Olujide M.G. and Adeogun S.O. (2006) Assessment of cocoa growers’ farm management practices in Ondo State, Nigeria. Spanish Journal of Agricultural Research (2006) 4(2), 173-179 Padi . B and Owusu. G.K.(2001) Towards an integrated pest management for sustainable cocoa production in Ghana. Nationalzoo.si.edu/scbi/migratorybirds/research/cocoa/padi.cfm retrieved on November 13, 2014 124 Phrampus, E., David, M. & John, .W. (2009). Engaging encounters: keeping everyone in the game by combining audience response technology and simulation based group Education. University of Pittsburgh School of Medicine, Pittsburgh, PA. Ugwuja V.C., Adesope O.M.,Odeyemi T.J., Matthews-Njoku E.C., Olatunji S.O., Ifeanyi- Obi C.C. and Nwakwasi R. (2011) Socioeconomic Characteristics of Farmers as Correlates of Fertilizer Demand in Ekiti State, Southwest Nigeria: Implications for agricultural extension. Greener Journal of Agricultural Sciences Vol. 1 (1), pp. 048-054 Kumar N. and Quisumbing A. R. (2010) Inheritance, Legal Reform, and Gender Differences in Poverty and Well-Being in Rural Ethiopia. CPRC/ODI Roundtable on Inheritance and the Intergenerational Transmission of Poverty, ODI, London, 11th October, 2010. International Food Policy Research Institute pp 1-21 Yahya I. H. (2009). Information services for rural community development in Nigeria. http://www.webpages.uidaho.edu/~mbolin/harande.htm Retrieved on July 13 2013. onlinelibrary.wiley.com/doi/10.../pdf Yahaya, M. K., and Badiru, O. I. (2002). Measuring the impact on farmers of agricultural radio and television programs in Southwest Nigeria. Journal of Applied Communications, 86 (3), 24–36. 3.2 Analysis of spillover effects of Integrated Pest Management strategy for suppression of Mango fruit flies in Meru County, Kenya Caroline Githiomi¹, Beatrice Muriithi²*, Patrick Irungu¹, Chris Miyinzi, Hippolyte Affognon³ Gracious Diiro². ¹University of Nairobi, P.O Box 29053, Nairobi, Kenya; ²International Centre of Insect Physiology and Ecology (ICIPE), P. O. Box 30772-00100 Nairobi, Kenya International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Bamako, BP 320, Mali Correspondence: *bmuriithi@icipe.org; Tel: +254 720 663 800 ABSTRACT Horticultural sub sector in Kenya experiences many challenges among them insect pests like the fruit flies. An integrated pest management (IPM) strategy for suppression of fruit flies among mango growing communities in Africa has recently been developed and disseminated by International Centre of Insect Physiology and Ecology (Icipe) and partners. While the economic benefits of the strategy in mango production has been established, the cross-commodity benefits have not been assessed, yet the target invasive fruit fly species Bactrocera invadens has a wide range of cultivated host plants. Using propensity score matching analytical technique, we analyze the spillover effect of 125 adoption of IPM strategy for fruit fly control on farm incomes; utilizing cross-sectional data collected using a household survey of 374 farmers in Meru County, Kenya. We focus on four alternative cultivated hosts- avocado, pawpaw, oranges and bananas -of this major quarantine pest that are predominantly grown in Meru, one of the counties in Kenya where the fly population has been observed to occur in large numbers. Our findings suggest that cross-commodity spillover effect in the use of the IPM strategy exerts a positive and significant impact on gross margins of pawpaw and oranges by Ksh. 28,154 and 11,114 with kernel based matching and Ksh 27,249 and 11,025 with radius matching per acre per year respectively. The results of this study suggest that there is a wide scope of IPM investment in Kenya, and thus up scaling this type of technology at large scale, throughout Africa would be expected to substantially increase household income and boost productivity of existing cultivated hosts of B. invadens. Key words; mango fruit flies, cross-commodity spillover effects, propensity score matching, Kenya INTRODUCTION Horticultural sector plays a major role in the economy of many African countries. It is among the leading foreign exchange earners, and contributes to food security and employment, especially among smallholder farmers (Salami et al., 2010). In Africa for example, in 2007, fruit and vegetable exports generated over US$ 16 billion in foreign exchange from exported commodities employing over 40 million people directly and indirectly (Billah et al., 2015). Between 2001 and 2011, Kenya’s horticultural sector’s exports grew at an average annual rate of 16 percent. While in 2012, the total value of horticultural exports was Ksh 87 billion having exported 380,000MT of produce. This was a 4% decline in quantity exported as compared to 2011. The decline in quantity and value was due to the reduction in quantities of nuts and processed fruits exports (Kibira et al., 2015). Despite the importance of the subsector to the economy, it experiences many challenges among them the reduction of both quality and quantity of marketable produce which is mainly attributed to attack by insect pests such as fruit flies (Ekesi & Billah, 2007). Detected in Kenya in 2003, Bactrocera invadens (B.invadens), an invasive fruit fly species is one of the most destructive insect pests in horticultural production (Drew et al., 2005). B.invadens is ubiquitous, highly polyphagous attacking 40 host fruit and vegetable crops in 22 families in Africa among them mango, pawpaw, avocado, citrus, guava, and banana (Billah et al., 2015). Although the primary host of B. invadens is mangoes, the preferred hosts vary according to the region, climate, and host availability (Cugala et al., 2014). Past studies show up to 12–60% loss of mango and pawpaw attributed to B. invadens ( Danjuma et al.,2014). Fruit fly infestation cause approximately US$ 1 billion annual economic losses from mangoes globally and more than US$ 42 million in Africa (Lam et al., 2012). In 2008, Kenya lost US$ 1.9 million due to B. invadens quarantine restriction for avocadoes imposed by South Africa (Ekesi, 2010). 126 In Africa, the International Centre of Insect Physiology and Ecology (ICIPE) have developed and disseminated an integrated pest management (IPM) strategy to help cope with B. invadens challenge in mango production. The ICIPE-IPM package consists of five components (see Muriithi et al., 2016). Past studies have shown that the use of IPM reduces mango damage caused by fruit flies by a bigger percentage compared to use of synthetic pesticides for instance lower fruit damage of less than 14% was recorded by mango farmers participating in the application of IPM components in Embu, Kenya (Kibira et al., 2015). Muriithi et al. (2016) also found that application of the IPM strategy resulted in a 48% average increase in mango net income while mango yield losses due to fruit fly infestation reduced by an average of 19% among the IPM users. Although the economic benefits of the IPM strategy on the primary target crop – mango- are clearly demonstrated as highlighted above, the potential gains of the strategy on other cultivated hosts of the invasive fruit fly species within a farm have not been quantified in the previous studies. These past studies fail to capture the widespread diffusion of the technology to other host cultivated plants which may essentially under estimate the actual impact of IPM on farm income. This study seeks to fill this gap by analyzing the unintended (spillover) effects of the strategy using specific crops enterprises within the same farm as mango using one of the ICIPE-IPM action sites, Meru County in Kenya. The study adopt an impact evaluation approach by fitting a propensity score matching regression model to assess the spillover effect of IPM strategy for control of mango fruit fly on the farm income generated from avocados, pawpaw, orange and banana. METHODOLOGY Data The data used in this study was collected through a household level survey in three sub-counties in Meru County where mango production is predominant; Meru Central, North Imenti and Tigania West. Three hundred and seventy four (374) of the 1200 households that were surveyed in 2013 by ICIPE were randomly selected and revisited in November and December 2014. The sample consisted of 210 IPM adopters and 164 non-IPM mango farmers from the control area, computed following (Bartlett et al., 2001). The survey was conducted using questionnaires administered by trained enumerators and supervised by one of the authors of this study. The data collected included information on input use, yields and output prices, farm-level and socioeconomic characteristics of the households. Theoretical review Adoption of IPM technology for suppression of mango fruit- fly Farm level economic analysis of IPM technology investigates whether when the technology is disseminated, adopted and spills results in higher farm profits (Kibira et al., 2015). Farmers benefit from externalities derived from sustained IPM practice. 127 Adoption of IPM technology is assumed to be a dichotomous choice that is, the new technology is adopted when the net benefits from using that technology outweigh those of not adopting the IPM technology. This study assumes that the use of IPM technology is expected to affect the demand of other inputs such as pesticides and fertilizers as well as yields and income. Following Ariane & Guthiga (2012) linking adoption decisions with potential outcomes, this study considers a risk-neutral farmer who minimizes the total cost of production which comprises traditional/conventional costs (input costs), subject to conventional constraints. The farmer chooses IPM technology I to minimize the conventional costs. Algebraically this can be expressed as; Min C (WX) (1) Subject to a production function written as: Y(X) = Y (V, I (T), K, Z) (2) where C is the total input cost, W is the vector of input prices, X is vector of all production inputs, Y is the output produced, consumed and sold (as a result of using IPM technology and other inputs), V is a vector of conventional variable inputs such as fertilizer used by the farmer, I is conventional pesticides whose use embodies the use of IPM technology T, K and Z are fixed and quasi-fixed capital inputs and institutional factors respectively. The farmers’ optimization problem is therefore to choose I that minimize the total cost of production subject to a given quantity of output Y0 as shown below. When stated differently, the farmer will decide to adopt IPM technology for suppression of mango fruit fly if doing so minimizes the total cost of production subject to a target output level. To simplify this, when two inputs are used, the production function is assumed to take the following functional form; Y= f [V I (T)] = Y0 = [V I (T)] Min C = W1V + W2 1(T) (3) Subject to: 𝑓[𝑉, 𝐼(𝑇)] ≥ 𝑌0 (4) a lagrangian function is written for this problem as shown; ℓ = W1V + W2I(T) + λ(Y0 − f(V, I(T)) (5), after which we obtain conditional factor demand for using IPM technology. The solution of the lagrangian function which is associated with the cost minimization problem yields, among others, I* which is conditional input demand equation (associated with IPM technology) as functions of output Y, input prices W, conventional variable inputs V, fixed factors K and institutional factors Z. This is expressed as; I*=I*(W, Y, V, K, Z) (6) Equation (6) indicates that adoption of IPM technology for suppression of mango fruit- fly I which is affected by, among others, factors prices (incentives), the fixed and quasi- fixed capital (capacity) and institutional factors. Some of the capacity variables could be farm and farmer specific. 128 Estimation strategy for spillover effect of IPM technology for suppression of mango fruit- fly Following Ariane & Guthiga (2012),we model the indirect impact (spillover) of the adoption of IPM technology by small-holder farmers on household income Y, as a linear function of explanatory variables (Xi), an adoption dummy variable ( Ri) = 1 if the farmer adopts the technology and 0 otherwise, μi is the error term, β and A are coefficients. The linear regression model for assessing the indirect impact of IPM technology for suppression of mango fruit fly on farm income can be specified as; Y = βXi + ARi + μi (7) Whether farmers adopts the IPM technology or not is dependent on the farm and farmer characteristics, hence the decision by a farmer to adopt is based on each farmer’s self-selection instead of random assignment. Assuming a risk-neutral farmer, the index function to estimate adoption is expressed as; Ri ∗ = γXi + ei (8), where R𝑖 ∗ is a latent variable denoting the difference between utility from adopting the IPM technology U𝑖𝐴 and the utility from not adopting the technologyU𝑖𝑁. The farmer will adopt the IPM technology if R𝑖 ∗ = U𝑖𝐴 − U𝑖𝑁 > 0 . The term 𝛾𝑋𝑖 provides an estimate of the difference in utility from adopting the technology (U𝑖𝐴 − U𝑖𝑁), using the household and farm-level characteristics, as explanatory variables, while 𝑒𝑖is an error term. In estimating equations (7) and (8), it needs to be noted that the relationship between IPM technology and outcome (farm income) could be interdependent. Thus, technology can help increase output and as such richer households may be better disposed toward the adoption of new technologies. Thus, treatment assignment is not random, with the group of adopters being systematically different, resulting in selection bias problem. To address the problem of self- selection we adopt a propensity score matching (PSM) technique, which involves finding a group of treated individuals similar to the control group in all pre-treatment observed characteristics, such that the only difference is that one group use IPM technology while the other does not ( Hirano & Imbens, 2004). PSM starts by fitting a discreet choice model to estimate the propensity scores and in this case a logit model was used: 𝑃(𝑋) = 𝛽0 + 𝛽1𝑎𝑔𝑒 + 𝛽2𝑎𝑔𝑒𝑠𝑞𝑢𝑎𝑟𝑒𝑑 + 𝛽3𝑠𝑒𝑥 + 𝛽4𝑒𝑑𝑢𝑐𝑎𝑡𝑖𝑜𝑛 + 𝛽5𝑙𝑎𝑛𝑑𝑠𝑖𝑧𝑒 + 𝛽6𝑛𝑜. 𝑜𝑓 𝑚𝑎𝑛𝑔𝑜 𝑡𝑟𝑒𝑒𝑠 𝑜𝑤𝑛𝑒𝑑 + 𝛽7𝑤𝑒𝑎𝑙𝑡ℎ 𝑐𝑎𝑡𝑒𝑔𝑜𝑟𝑦 + 𝛽7𝑎𝑔𝑟𝑖𝑐𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝑒𝑥𝑡𝑒𝑛𝑠𝑖𝑜𝑛 𝑠𝑒𝑟𝑣𝑖𝑐𝑒𝑠 + 𝛽8𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑡𝑜 𝑡𝑎𝑟𝑚𝑎𝑐 𝑟𝑜𝑎𝑑 + 𝛽9𝑎𝑐𝑐𝑒𝑠𝑠 𝑡𝑜 𝑖𝑟𝑟𝑖𝑔𝑎𝑡𝑖𝑜𝑛 𝑤𝑎𝑡𝑒𝑟 + 𝛽10 𝑔𝑟𝑜𝑢𝑝 𝑚𝑒𝑚𝑏𝑒𝑟𝑠ℎ𝑖𝑝 + 𝛽11𝑡𝑟𝑎𝑖𝑛𝑖𝑛𝑔 + 𝜇𝑖 ( 9) In PSM the variable of interest is the average treatment of the treated (ATT). The spillover effects of IPM are as a result of impact on host crops when farmers adopt the IPM strategy on mangoes. Therefore, the set up for estimating the spillover effects was not different from estimation of direct impact of the IPM technology. 𝐴𝑇𝑇𝑖 = 𝐸(𝑌𝑖|T = 1) = E(𝑌1𝑖(1)|T = 1) − E(𝑌0𝑖(0)|T = 1) (10) ATTi is also called conditional mean impact. Y1i is the gross margins for pawpaws, avocado, oranges and bananas of the 𝑖 -ith farmer who participated in the program 129 while Y0i is the gross margins of the 𝑖 -ith farmer who had not participated in the program. RESULTS AND DISCUSSION Descriptive statistics Three households were dropped from the selected sample due to missing data and apparent enumerator errors, leaving a sample of 371 households that is utilized in this article for analysis. The mean age of the household heads for the IPM users was 55 years while for the control was 56 years. Age of the household head was expected to have negative influence on technology, that is, the more the years a farmer had the less the likelihood of applying a technology This was not the case because 208 farmers (56%) who had applied IPM had a mean age of 55.3. This would be because only 8.4% of the interviewed household heads were youths. It would also be because the older farmers were the land owners and made all the decisions to be carried out in the farm. The average farming experience was 28 years for both IPM users and non-users. The mean cultivatable land size was 5.1.acres for the IPM users and 3.9 acres for the IPM non- users Empirical results Table 1 presents results of the logit model employed to predict the probability of applying IPM strategy. The dependent variable is a binary IPM adoption. Prior to running the model, a test is conducted to detect the problem of multicollinearity between the variables included in the analysis (Molefe & Hosmane, 2007). The mean VIF was 1.36; all the variables in the model had a VIF far less than 10 which satisfy the rule of thumb. Further, link test was done to show whether the model in this study is properly specified and the _hat was found to be statistically significant with a p-value of 0.0000. On the other hand the _hatsq was insignificant at p (0.811) which was in line with other authors (ibid). The logit results are only discussed briefly as our main objective is to evaluate the indirect impact of IPM technology on farm income. To calculate asset index, we followed the recommendation made by Filmer & Pritchett, (2001) to use principle components analysis (PCA) to aggregate several binary asset ownership variables into a single dimension. PCA is relatively easy to compute and understand, and provides more accurate weights than simple summation. Farmers with bigger land sizes are more likely to apply the IPM strategy. Our results in table 1 also indicate that the probability of the decision to apply IPM strategy increased with distance from the tarmac road. A possible explanation stated by different authors is that households closer to market towns which are characterized by presence of tarmac roads tend to engage more heavily in non-farm activities such as small businesses compared to households that are further away from market centers and whose livelihood opportunities are limited to farm enterprises (Muriithi & Matz, 2014). 130 Asset index was used in this study to categorize the households according to wealth status with respect to the amount of assets a household has or its wealth. It has a positive influence on the adoption of IPM strategy meaning that the more assets a household has the higher the chances of adoption of IPM strategy. Finally, farmers who attended trainings on mango production were more likely to apply IPM strategy since they were more knowledgeable on the advantages to damage control measures. Table 2 presents the estimates of ATT by the Radius matching (RM) and Kernel based matching (KBM) methods. The matching results in table 2 indicate that application of IPM has positive and significant effect on gross margins of pawpaw and oranges. Applying the IPM strategy increased the average gross margins of pawpaws and oranges by a range of KSh. 28,154 - 11,114 and KSh. 27,249 - 11,025 per year per acre respectively. The negative causal effect for avocado and bananas implying the users of IPM suppression strategy had lower gross margins per acre in the two crops than the non- users. In their study on seasonality and host utilization of the invasive fruit fly, B. invadens in central Tanzania, Mwatawala et al. (2006) observed that avocado contained 0.8 flies per kilogram compared to other fruits like oranges and mangoes which were 2.9 and 149.3 fruit flies respectively. Several authors have also noted that infestation of bananas by B.invadens is limited in the “all yellow fruit stage” and not mature green bananas (Cugala et al.,2014). From the last column in table 2, the level of gamma (critical level of hidden bias) is defined as the odds ratio of differential treatment assignment due to an unobserved covariate. The sensitivity analysis for the indirect effect of mango fruit fly IPM control technology on gross margin of oranges for example suggest that at the level of γ = 3.05 the causal inference of the significant effect would have to be viewed critically. For a gamma level of 2.35 it implies that if individuals who have the same characteristics differs in their odds ratio of participation by a factor of 135 percent then the significance of the estimated participation effect on gross margins would be questionable. Therefore in conclusion the gamma results shows that even slightly bigger amount of unobserved heterogeneity would not alter the inference about the estimated indirect effects of IPM strategy but it would be imperative if the study is done after several years of application of IPM to ascertain the inference. Tables Table 1: Propensity score estimates: (logit estimates) Explanatory variables Coefficient Std Error P-value Gender of the household head 0.41 0.64 0.521 Age of the household head (years) -0.098 0.15 0.517 Age of the household head squared 0.0007 0.001 0.577 Number of years household head completed in school -0.027 0.55 0.621 Log of agricultural cultivatable land in acres 0.83 0.315 0.008** Access to irrigation water(dummy) 1.10 0.49 0.027** 131 Distance to the nearest tarmac (km) 0.52 0.07 0.000*** Distance to the extension provider (km) -0.03 0.04 0.379 Asset index (proxy for wealth category) 0.05 0.02 0.009** Training(dummy) 1.89 0.52 0.000*** Constant -5.46 4.05 0.178 Number of observations Pseudo R2 =0.7261 371 Notes: *** Significance at 1% level, ** significance at 5% level, *significance at 10% Source: Field survey data Table 2: Average treatment effect results Crop enterprise Matching algorithm Average gross margin based on matched observations ATT t-values Critical level IPM users IPM non- users Avocado KBM 12659 25219 -12560 -3.58*** 3.40-3.45 RM 12659 25506 -12847 -3.68*** 3.40-3.45 Orange KBM 21154 10040 11114 3.07*** 3.05-3.10 RM 21154 10129 11025 2.83*** 3.05-3.10 Pawpaw KBM 64272 36118 28154 2.61*** 2.35-2.40 RM 64272 37023 27249 2.54*** 2.35-2.40 Bananas KBM 18357 39473 - 21,116 -4.64*** 2.00-2.05 RM 18357 39985 - 21,268 -4.77*** 2.00-2.05 Notes: *** Significance at 1% level, ** significance at 5% level, *significance at 10% KBM with band width 0.03 and common support: RM with 0.03 caliper and common support Source: Author’s computation CONCLUSIONS We have analyzed the spillover effects of IPM strategy adoption on farm income of smallholder farmers in Meru County, Kenya. A non-parametric econometric technique is used to mitigate selection bias stemming from observed heterogeneity and to test robustness of results. This involves the application of propensity score matching technique to measure the effects of adoption of the IPM strategy for suppression of mango fruit fly on gross margins of avocado, oranges, pawpaws and bananas commonly grown in Meru County and which are also hosts of B. invadens. Results indicate that applying IPM strategy has generated a positive and significant effect on gross margins of pawpaws and oranges at the household level. Applying the IPM strategy increased the average gross margins of pawpaws and oranges by a range of KSh. 28,154 - 11,114 and KSh. 27,249 - 11,025 per year per acre respectively. The study therefore concludes that IPM strategy for suppression of mango fruit fly increases the farmers’ farm income by reducing crop damage to other host crops. 132 The findings are important for policies that focus on introduction of new technologies targeting small-scale farmers to enable them increase their farm incomes and consequently reduce poverty. Further research using panel data would help in addressing the effects of unobserved heterogeneity, which was not possible to account in the PSM approach. Acknowledgements The authors would like to thank DFID for the financial support. The authors would also wish to thank the African Fruit Fly Programme (AFFP) of the International Centre of Insect Physiology and Ecology (icipe) for field logistics support, farmers, contact persons and the enumerators for their efforts. References Ariane, Z., & Guthiga, P. (2012). An Assessment of the Impact of Rainwater Harvesting Ponds on Farm Income in Rwanda. Presented at the 2012 Eighth AFMA Congress, November 25-29, 2012, Nairobi, Kenya, African Farm Management Association (AFMA). Baartlett, J. E., Kotrlik, J. W., & CCHCC, C. H. (2001). Organizational research: Determining appropriate sample size in survey research appropriate sample size in survey research. Information Technology, Learning, and Performance Journal, 19(1), 43. Billah, M., Afreh-Nuamah, K., Obeng-Ofori, D., & Nyarko, G. (2015). Review of the pest status, economic impact and management of fruit-infesting flies (Diptera: Tephritidae) in Africa. African Journal of Agricultural Research, 10(12), 1488–1498. Cugala, D., Ekesi, S., Ambasse, D., Adamu, R., & Mohamed, S. (2014). Assessment of ripening stages of Cavendish dwarf bananas as host or non‐host to Bactrocera invadens. Journal of Applied Entomology, 138(6), 449–457. Danjuma, S., Boonrotpong, S., Thaochan, N., Permkam, S., & Satasook, C. (2014). Journal of Advances in Agricultural Science and Technology. Drew, R. A. I., Tsuruta, K., & White, I. (2005). A new species of pest fruit fly (Diptera: Tephritidae: Dacinae) from Sri Lanka and Africa. African Entomology, 13(1), p– 149. Ekesi, S. (2010). Combating fruit flies in eastern and southern Africa (COFESA): elements of a strategy and action plan for a regional cooperation program. Fly Issue Paper- World Bank. Ekesi, S., & Billah, M. (2007). A field guide to the management of economically important tephritid fruit flies in Africa. ICIPE Science, Nairobi. Ekesi, S., Mohamed, S., & Tanga, C. M. (2014). Comparison of Food-Based Attractants for Bactrocera invadens (Diptera: Tephritidae) and Evaluation of Mazoferm– Spinosad Bait Spray for Field Suppression in Mango. Journal of Economic Entomology, 107(1), 299–309. 133 Filmer, D., & Pritchett, L. H. (2001). Estimating wealth effects without expenditure data— or tears: an application to educational enrollments in states of India. Demography, 38(1), 115–132. Hirano, K., & Imbens, G. W. (2004). The propensity score with continuous treatments. Applied Bayesian Modeling and Causal Inference from Incomplete-Data Perspectives, 226164, 73–84. Kibira, M., Affognon, H., Njehia, B., Muriithi, B., Mohamed, S., & Ekesi, S. (2015). Economic evaluation of integrated management of fruit fly in mango production in Embu County, Kenya. African Journal of Agricultural and Resource Economics Volume, 10(4), 343–353. Lam, H., Edewa, A., & Kleih, U. (2012). Impact of SPS standards on agri-food trade: A case study of the invasive fruit fly (Bactrocera invadens) in Kenya. Food Chain, 2(1), 86– 103. Molefe, A. C., & Hosmane, B. (2007). Test for link misspecification in dependent binary regression using generalized estimating equations. Journal of Statistical Computation and Simulation, 77(2), 95–107. Muriithi, B. W., Affognon, H. D., Diiro, G. M., Kingori, S. W., Tanga, C. M., Nderitu, P. W., Ekesi, S. (2016). Impact assessment of Integrated Pest Management (IPM) strategy for suppression of mango-infesting fruit flies in Kenya. Crop Protection, 81, 20–29. Muriithi, B. W., & Matz, J. A. (2014). Smallholder participation in the commercialization of vegetables: evidence from Kenyan Panel Data. Available at SSRN 2405462. Mwatawala, M., De Meyer, M., Makundi, R., & Maerere, A. (2006). Seasonality and host utilization of the invasive fruit fly, Bactrocera invadens (Dipt. Tephritidae) in central Tanzania. Journal of Applied Entomology, 130(9‐10), 530–537. Salami, A., Kamara, A. B., & Brixiova, Z. (2010). Smallholder agriculture in East Africa: trends, constraints and opportunities. African Development Bank Tunis, Tunisia. 3.3 Challenges facing strawberry farming in central Kenya Egesa A.O.1*, Njeri N.1, Matheri F.2, Mwirigi P.1 and Mwangi M.1 1Department of Agricultural Science and Technology, Kenyatta University, P.O. Box 43844 Post Code 00100 Nairobi, Kenya, 2Plant Health Division, International Centre of Insect Physiology and Ecology (ICIPE), P.O Box 30772 Post Code 00100 Nairobi, Kenya *Corresponding author’s email: ayechoogola@gmail.com ABSTRACT Fruits are important for food and nutrition security. Among fruits, strawberries have high nutritive value and enhance human wellbeing through provision of large quantities of essential nutrients and antioxidants. The berry contains phytochemicals 134 such as anthocyanins and ellegitannins which produce the anti-cancer agent, ellagic acid, in addition to having alkalizing and anti-diuretic properties that prevent kidney diseases. The fruit is rich in vitamin C, iron, calcium, phosphorus, magnesium and manganese. Increasing demand for the fruit has increased cultivation and consumption in rural and peri- urban regions, especially in Nairobi, Nyeri and Kiambu Counties. Despite increased interest by growers in peri-urban areas, production is still low due to a number of challenges facing farmers. A field survey was carried out in Nyeri, Nairobi and Kiambu Counties to ascertain these challenges and provide insights to possible solutions. Among the respondents, insufficiency of high quality planting material was most significant (90%), while lack of production knowledge and skills (30%), diseases (19%), insect pests (17%), birds (17%) and poor market chain (17%) were major. Other significant challenges included insufficient postharvest storage and handling systems (<10%). These findings show that provision of quality planting materials, educating farmers on strawberry production and increasing availability of effective crop protection measures are critical to increasing production. Further research is needed to evaluate inputs use (fertilizer regimes, pesticides and water) and their impact on productivity of strawberry in Kenya. Key words: Challenges, Improvement, increased, Production, Strawberry INTRODUCTION In Kenya, Strawberry farming is described as a highly profitable emerging venture for small scale horticultural farmers (Maina, 2010). However adoption of this crop by Kenyan vegetables and fruit farmers in has been slow. The crop requires intensive management including good land preparation, well timed agronomic management practices, proper plant nutrition and excellent harvest and post-harvest handling, (Mwangi and Mwaura, 2009). Strawberry has been consumed by man from as early as 2000 BC as a wild fruit (Vaughan D. 1987). This wild or Wooden strawberry (Fragaria vesca) was first described by Vilmorin-Andrieux (1885; 2011), attributing it to perfume like aroma and delicacy of flavor with about 2,500 seeds to the gramme. He differentiated Wooden strawberry from Alpine Strawberry, describing Alpine ones as larger with fewer seeds (1500), and of continuous flowering in summer. Ottoman Strawberry is another unique type of strawberry which has been in production since 1900. Strawberries are fruits from the family Rosaceae, which comprises of: peaches, nectarines, apples, pears, quinces, apricots, plums, cherries, raspberries, loquats, blackberry, apricots and almonds (Harris and Mitcham, 2007). These fruits are widely accepted for their distinctive high quality aroma, juicy texture, and sweetness. Strawberries are characteristically bright red in color, although some new varieties have been bred to yellow and white amongst other colors. The fruits are consumed as fresh and in processed solid food and drinks, milkshakes, and chocolates, (Maina, 2010). The fruit is highly sought for by food processing industries, juice, yogurt, jams, cakes and 135 sauce making. Products constituting of fresh strawberries such as cakes are most of the time for the high end market, fetching high prices (Rahman, 2015). A number of strawberry varieties are categorized into June-bearing varieties: Chandler, Honeyoye, Allstar, Eclair; ever bearing varieties: Quinault, Ozark Beauty and Fort Laramie; and day neutral varieties: Tristar, Selva and Veestar. June bearing varieties yield a big harvest in one month while everbearing varieties strawberries bear twice per year while, day neutral varieties flower and fruit continuously. The country’s most common varieties are: Pajero, Chandler, Tri-Star, Tioga, Domanil, Douglas, Selva, Rabunda and Tribute (Waithaka and Ngugi, 1985). According to Maina, (2010) some varieties have been extensively spread in Kenyan horticultural crops production areas covering, Limuru, Nyeri, Kirinyaga, Kajiado, Molo, Kinangop, Nakuru, Eldoret, Kericho, Meru, Embu, Mt. Elgon, Kitale and Bugoma. Current varieties under production include: Chandler, Rotunda, Douglas, Pajaro and Aiko, (Waithaka and Ngugi, 1985; Mwangi and Mwaura, 2009). During production and post-harvest handling, Strawberries are susceptible to attacks by several insect pests and diseases leading to huge crop losses. The pests include insects, birds and snails. Weed control in strawberry field is sometimes difficult due to the plant architecture which constitutes of shallow root systems that makes uprooting of weeds near roots difficult. This necessitates the need for chemical herbicides which are deleterious to the environment. Plastic mulch and natural mulch, like rice straw and grasses are potential (cheap) means of weed control and soil conservation (Mwangi and Mwaura, 2009). METHODOLOGY A field survey was conducted in selected counties in Kenya covering; peri-urban areas in Nairobi, Kiambu and Nyeri Counties in February, 2016 using a stratified sampling procedure. The area was first stratified in terms of geographical distance to cover the approximate ecological range of strawberries growing in these counties. Counties and sub counties’ agricultural officers were used as key informants for accurate identification of strawberry farmers. Data was collected through personal interviews with individual farmers and labourers involved in management of strawberry orchards using structured and semi-structured questionnaires. Characteristics of strawberry production under small holdings was recorded which included: details of years in strawberry production, farm sizes, sources of planting materials and production level per week of the farmers interviewed. Information on challenges in production ranging from inputs to other agronomic and market related issues were gathered. The data collected was subjected to descriptive analysis, mainly to obtain frequencies and valid percentages of occurrences of testable variables using the SPSS data analysis software version 20.1. 136 3.0. RESULTS 3.1. Production area and time The study confirmed a higher concentration of smallholdings strawberry farming in Nyeri County The villages with highest percentage of respondent farmers growing strawberry were Kwa wambui, Karia, and Iganjo at 20% each, while, Sagana and Kiamariga, Mutaaga, Karandi and scheme, each had 10% of the strawberry farmers interviewed, (figure 1). Figure 1: Strawberry most intensive production villages under smallholdings in Nyeri County A significant proportion of farmers had small portions of land allocated to strawberry production, despite having experience of strawberry farming for over three years. All the farmers interviewed were small scale farmers, cultivating from less than1/8 to ¾ acres, (figure 2). 0 5 10 15 20 25 karia kiariariga kwa wambui kanyo mutaaga sagana p e rc e n ta ge o f re sp [o n d e n t st ra w b e rr y fa rm e rs Villages 137 Figure 2: Average farm sizes of strawberry production Majority (50%) of these famers had been growing strawberries for about 5 years while 20% of the farmers had been in strawberry enterprise for over 6 years. The remaining (30%) had been growing the berries for one to four years, (figure 3). Figure 3: Resilience in strawberry production by smallholder farmers in Nyeri County Source of planting materials The predominant strawberry variety grown was Chandler and Pajero. The common form of propagation was splits (vegetative materials) which were obtained from 0 10 20 30 40 50 60 1-3 years 4-5 years 6-10 years % r e sp o n d e n t fa rm e rs Time period 138 runners with majority of the farmers (70%) acquiring them from their neighbours, (figure 4). Figure 4: Sources of strawberry plantings materials among smallholder farmers in selected counties in Kenya Strawberry yields The average yields obtained by farmers were low, the yield per week during the peak harvest season were said to range from, 36kg (50% of farmers), 16kg (30% farmers) with some few farmers (20%) recording as low as 6kg per week as their highest yields, (figure 5) Figure 5: Strawberry production in Nyeri 0 10 20 30 40 50 60 70 80 brokers commercial neighbours own % o f R e sp o n d e n t fa rm e rs Source of planting materials 0 10 20 30 40 50 60 36kg 6kg 16kg % s tr aw b e rr y fa rm s Mass in kilograms 139 Pests, diseases and other production challenges At the field insect damage was reported (17%), birds (17%), and fungal diseases (19%). Insufficient production knowledge, poor soil fertility, insufficiency of cold storage facilities and excess rainfall constituted 30% of the challenges faced by farmers. Challenges in handling fruits and timely access to good markets was difficult resulting to poor marketing prices (17%), (figure 6) Figure 6: Other challenges in strawberry production Open field cultivation was practiced using unskilled labour. Farm input was a mixture of both of inorganic and organic compounds. Inorganic fertilizers in use included: a compound fertilizer Nitrogen Phosphorus and Potassium (NPK 17:17:17) and calcium ammonium nitrate. The bio stimulants in use included: D.I. Grow® and OMEX® (foliar macro and micro –supplements). Bestock® was used in control of insect pests while Oshothane® and Ridomil® were the main fungicides for control of fungal pathogens Brokers were the main target market at farm gate. Generally it was noted that with marketing, there were quite a number of challenges, due to the high perishability of the fruit and insufficiency of cold storage facilities. These factors contributed to low pricing of the fruit. DISCUSSION The area under study (Central Kenya) is characterized by well drained, loam soil which is preferable for strawberry production (Taneyhill, L, 2010; Rowley and Drost, 2010). 140 Allocation of small land sizes (1/8 acres) to strawberry production, could be attributed to the rapid land subdivision in Kenyan peri-urban areas The difference in the number of farmers growing strawberry in the villages under study could have been as a result of variation in uptake of strawberries among the villages. This may also be contributed to by the variation in farmer perception on the production, and fear of inability to handle challenges accompanied. Use of unskilled labour in production as observed in the study area is well explained by both the small scale of production and uneconomically high cost of production inputs known to affect smaller enterprises. Most of Kenyan strawberry produced in large scale is for export market. The study has identified that majority of the smallholder farmers growing these fruits lack the required production knowledge hence depend on their neighbours for information on production and other aspects of strawberry management. In 2012, Kenya produced 239 tons of strawberry, compared to 1,366,850 produced by USA in the same year (FAOSTAT, 2012). Of the 239 tons produced in Kenya, 90% is likely to be exclusively from commercial export farms. The smallholders with farm sizes of 1/8 to ¾ acres, contribute to low yields which never get formal market access. The yield volume and cost of production was greatly influenced by the size of land under cultivation as can be confirmed by similar studies by Woods and Thornsbury, (2005). Unavailability of formal seed/seedlings system to smallholder strawberries farmers is a barrier to production of a number of varieties by farmers apart from Chandler and Pajero which are obtained from neighbours. This limited choice of variety may not be a choice but rather the only choice of planting strawberry to smallholders without their access to export companies controlled strawberry nurseries, (Lantz, et al., 2016). Majority of farmers sourced their material from neighbours and as such picking on the available variety which was probably Chandler. In their earlier survey, Mwangi and Sarah, (2009) had mentioned poor seedling production systems for strawberry, this was found to be still the case 7 years later, as exhibited by the high number of farmers obtaining their seed material from neighbors The variation in the amount of produce could have been due to factors such as differences in agronomic practices among the respondents, size of land under strawberry production or even the experience of the respondent farmers in production. Size of land as well as difference in agronomic practice has also been attributed as a cause in variation of yield by Woods and Thornsbury (2005) Challenges gathered in this survey such as, poor soil fertility (fertilizer requirements), and pests in strawberry production pests are similar to findings by Paparozzi, (2013). The choice of the preferred chemicals could have been due to their readily availability in the local market, their perceived efficiency against the respective pests and, or their lower cost compared to similar products from other manufacturers. 141 REFERENCES Lantz, W., Swartz, H., Demchak, K. and Frick, S. 2016. Season-Long Strawberry Production with Ever- bearers for Northeastern Producers EB 401. Retrievedfrom,http://www.sare.org/content/download/65370/918209/Everbe aring_Strawberry_Guide.pd Accessed on.18/07/2016 Maina, W. 2010. Strawberries. Trawling into the exciting horticultural news. September- october, 2010.the east Africa fresh produce journal. Issue no 13. September October, 2010. Mwangi, M and Sarah, M. 2009. Challenges in strawberry seedling production in Kenya. Journal of Applied Biosciences, 23: 1403 – 1405 Paparozzi, T.E. 2013. The Challenges of Growing Strawberries in the Greenhouse. Forum proceedings, ‘‘Creative Thinking, Creative Funding: Research, Extension, and Teaching Programs and Consortiums’’ pp 800-802 Rahman, M. M., Moshiur. R.M.M, Khaleque M.AM. and Khaliq Q A. 2015. Field performance and fruit quality of strawberry genotypes under subtropical climate. ISSN 0258-7122 Bangladesh J. Agril. Res. 40(1): 137-151, march 2015 Rowley D and Drost, D. 2010. High Tunnel Strawberry Production. Utah State University. Taneyhill L. 2010. Strawberry Cultivation in Semi-arid Climates – A Manual. The Pennsylvania State University. Pp 1-11 U.S. Department of Agriculture-National Agricultural Statistics Service. 2000. 2002 Census of Agriculture. Volume 1, Chapter 2, Table 33 “Berries Harvested for Sale: 2002 and 1997”. Available from http://www.nass.usda.gov/census/census02/volume1/us/index2.htm (accessed February 24, 2016). Vaughan D. 1987. The plant macrofossils. 233-38. In: Balaam N D (et al.). Prehistoric and Romano-British sites at Westward Ho!, Devon, 1983-1984. In: Balaam N D, Levitan B and Straker S (eds). Studies in palaeoeconomy and environment in SW England. BAR BS 181. England, Devon Vilmorin, Andrieux. 2012. Perennial stoloniferous herbs, rooting at the nodes and forming new. plants. Vegetable garden diary 2012: album vilmorin (taschen diaries).published by taschen gmbh (2011-08-18) ISBN 10: 3836530252 Waithaka, K. and Ngugi, C.W. (1985). Growth Performances of Everbearing Strawberries in Kenya. Acta Hortic. 153, 85-90. DOI: 10.17660/ActaHortic.1985.153.10 http://dx.doi.org/10.17660/ActaHortic.1985.153.10 Woods M and Thornsbury S. 2005. Costs of Adopting Good Agricultural Practices (GAPs) to EnsureFood Safety in Fresh Strawberries.Agriculturaleconomicsreportdepartment of agricultural economics michigan state university. pp: 2-20 142 3.4 Evaluating health and environmental effects of mango integrated fruit fly management technologies in Meru County, Kenya. Chris Miyinzi1, Beatrice Muriithi²*, Vincent Ngeno ¹ Hippolyte Affognon³ Caroline Githiomi² Gracious Diiro². ¹ Department of Agricultural Economics and Resource Management, Moi University, P. O. Box 3900-30100, Eldoret, Kenya. ² International Centre of Insect Physiology and Ecology, P. O. Box 30772-00100, Nairobi, Kenya. ³International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Bamako, BP 320, Mali *Corresponding author; E-mail: bmuriithi@icipe.org; Tel: +254 720 663 800 Abstract Integrated pest management (IPM) has been promoted as an effective alternative means of controlling pests and diseases in agricultural production since it is environmentally friendly and economically feasible. Ideally, the adoption of IPM technologies allows farmers to reduce pesticide use and their harmful effects to health and the environment. This research work investigated the health and environmental effects of IPM technologies introduced recently by the International Centre for Insect Physiology and Ecology (icipe) and partners among Mango farmers in Meru County, Kenya. Environmental Impact Quotient (EIQ) was used to compare the effects of IPM technologies on health and environment with other common fruit fly control methods. The dummy variable for IPM use was included in order to analyze the effect of the adoption of IPM technologies on EIQ using Endogenous Switching Regression (ESR). The results showed that over 20 pesticides were used during mango production to control fruit flies; however, only 9 of over 20 pesticides were commonly used by majority of the mango farmers. Significant factors determining the magnitude of EIQ were farming experience, access to credit, labour management, use of protective clothing and number of production trees. The empirical results show that the use of IPM technologies reduces EIQ field use by 6.81 %, and this is statistically significant. This implies that IPM technologies have positive benefits on health and environment. Key words: Integrated pest management, environmental impact quotient, endogenous switching regression, Kenya INTRODUCTION Mango is the third most important fruit crop in Kenya after bananas and pineapples respectively in terms of production and area under coverage (Kibira et al., 2015). In Kenya, mango production in 2013 was under 46,980 ha which produced 581,290 tonnes (HCDA, 2013) accumulating Kenyan shillings 7.4 billion in totality. Mangoes processing earns foreign exchange through exportation, making a potential source of employment. Inclusive is farmer’s wellbeing both financially and nutritionally. However, Mango production suffers various challenges, among them two main pests (fruit flies, mango 143 seed weevil) and two main diseases (powdery mildew and anthracnose) (Griesbach, 2003). To counter decreased quality and quantity of mango products, farmers have resulted to using pesticides as a remedy. Sometimes limited effectiveness of synthetic pesticides and farmers level of knowledge prompt misuse of pesticides (Fan et al., 2015). Very few studies have been conducted to assess the sustainability of pesticides use in developing countries despite the increasing concern about their impacts on health and environment as is documented (Chitra et al, 2013; Edwards, 2013; London, 2013). Among the effects include environmental degradation, loss of biodiversity and irreparable changes to the eco-system (Relyea, 2005; Gill et al., 2014). High pesticide residue levels in fruits products restrict entry to international markets due to tight trade regulations as they pose health hazards. This has led to seizure and devastation of consignments and possibly outright bans on for the exporting countries. IPM is a chemical-free protection strategy that is less harmful to environmental components and human health as it offers safe and cost-friendly protection of crops against pests and diseases. IPM decreases the net pesticide used in fruit fly control hence environmental and health benefits. (Colette et al., 2001) reports that IPM reduces production cost by more than $173 Million and environmental costs by $99 million per year in USA. Kibira et al., (2015) and Muriithi et al., (2016) evaluated the impact of the IPM strategy on pesticide expenditure, mango fruit yield loss and profit in Embu and Meru Counties respectively. There’s possible gains underestimation as human and environmental effects have not been quantified despite the studied positive economic impacts. This study, addresses the health and environmental effects of the IPM strategy among mango producers in Meru County, in Kenya. METHODOLOGY 2.1. Study area and sampling technique This study was conducted in Meru County, Kenya where ICIPE has disseminated the strategy. A multi-stage sampling procedure had been applied in selecting three sub- counties, two were provided with IPM materials (Meru central and Imenti North), while a third wasn’t (Tigania West). 371 of the 1200 mango producers surveyed earlier by ICIPE were randomly selected and revisited in the October 2014. Probability Proportional to Size (PPS) sampling technique was used to select 206 adopters and 165 non-adopter farmers. METHODS AND DATA ANALYSIS Environmental Impact Quotient Descriptive statistics were used together with the EIQ and endogenous switching regression models to analyze the data in STATA version 12. To assess the relative difference adopters and non-adopters, the main descriptive analyses used were two- group mean comparison t-test and the two-sample test. EIQ model was used to determine the magnitude of health and environmental effects. It combines the pesticide 144 hazard posed to farm workers (applicators and pickers), consumers (farm products and ground water) and the local environment (aquatic and terrestrial) into one numerical value (Pradel et al.,2008; Macharia, 2009). EIQ model estimates pesticides risks on a three point scale of 1, 3, and 5 in accordance to the hazard of various pesticides; 1= lowest, 5=highest, (Kovach et al, 1992). The EIQ Formula is as stated below. 3/)]}5**()3**()3*)2/)((*( )*[()]()*)2/)((*[)]*()5*[({ PBPZPSD RFLSYPSCPDTDTCEIQ   (1) Where: C = chronic toxicity; DT = dermal toxicity P = plant surface residue half-life S = soil residues half-life SY= systemicity; L=leaching potential; F=fish toxicity R = surface loss D = bird toxicity Z = bee toxicity B = beneficial arthropods activity For comparison of pest control strategies, we computed EIQ field use by finding the product of EIQ, pesticide dose, percentage active ingredient and the frequency of pesticide application as shown in the formulae below. napplicatiooffrequencyingredientactiveEIQratingusefieldEIQ *%* ..... (2) Endogenous Switching Regression. According to Lokshin et al (2004) ,the probability of a farmer adopting IPM technologies is given by the expected benefits *sI against the expected effects of not adopting IPM practices, *tI . * sI and * tI are latent variables hence observed is the actual adoption of IPM technologies, I, I=1 if *tI  * sI and I=0 if * tI  * sI . Adoption of IPM can be denoted as:   ZI (3) Where; I = Dummy variable for adoption, Z= independent variables, α= vector of parameters and μ=error term. The general probit model of adopting IPM technologies is therefore specified as:       )()(ln)( )()()(ln)(ln)(ln )(ln)(ln(exp)ln)(ln)( 141312 11109876 543210 irrwatlabmngtcredit grpmempclothextenmtreesfsize hhsizeageeducgenderIPM jjj jjjjjj jjjjjj …… (4) Where; IPM= adoption status of the farmer. This variable takes a value of 1 if the farmer has adopted IPM practices and 0 otherwise. Full Information Maximum Likelihood (FIML) is an efficient method to estimate ESR because it calculates the probit equation and the outcome equation simultaneously to yield consistent standard errors (Alene et al., 2007). The general representation of the outcome equation is as shown:       )()(ln)( )()()(ln)(ln)(ln )(ln)(ln(exp)ln)(ln)( 141312 11109876 543210 irrwatlabmngtcredit grpmempclothextenmtreesfsize hhsizeageeducgenderEIQ jjj jjjjjj jjjjjj (5) Where; EIQ= the environmental field use representing impact of IPM technologies on health and environment. To successfully achieve this we need to estimate the conditional expectation of effects that participants would have without adoption of IPM 145 technologies (Lokshin et al 2004). For a farmer with characteristics X and Z who adopt IPM technologies, the expected effects can be defined as: SSVSs XIyE   )1/( (6) Where; ( SSV ) denotes sample selectivity, i.e. that a farmer who adopts IPM may be different from an average farmer with characteristics X and Z due to unobserved factors. For the same IPM adopter, the anticipated health/environmental effects would be, Stvtt XIyE   )1/( (7) The change in health and environmental effects due to adoption of IPM is calculated as: ssvtvtsts XIyEIyE  )()()1/()1/(  (8) This is the average treatment effect on the treated (ATT) which compares the magnitude of health and environmental effects for adopters with and without IPM technologies while average treatment on the untreated (ATU) compares the same effects for the non- adopters. This treatment effect on the treated is due to the differences in the coefficients in Equations. (7) and (8). RESULTS AND DISCUSSION Farmers perception on impacts of pesticides on health and environment Both IPM participants and non-participants ranked the risks associated with pesticide use to health and environmental indicators (table i). Both reported that pesticides use poses risk on their health and environment. On health and environment, IPM participants reported larger risk as compared to non-participants who reported large risk on pesticides. 3.2. EIQ for commonly used pesticides to control mango fruit flies in Meru County A total of 65 pesticides comprising of 41 active ingredients were used in mango production. Bayleton and Bulldock were the most used pesticides in mango production representing 34.50% and 33.96% of the total pesticide used respectively (table v). The two recorded 15.43% and 24.47% of the total pesticide usage by the non-participants and 22.78% and 22.42% among participants respectively. The average EIQ values for farm workers, consumers and environmental component of IPM non-participants were 24, 10 and 75, respectively while the mean components of IPM participants were 25, 10, and 73, respectively. Basing on EIQ Classification rule (Macharia, 2009), values for all the mango pesticides used in mango production showed that 30%, 25% and 45% of those pesticides were rated as low (EIQ = 0 to 20), medium (EIQ = 21 to 40) and high (EIQ ≥ 41), respectively. Results showed that the environmental component of the EIQ was high for both participants and non-participants, a significant difference was observable between the two groups. The total EIQ field use in Meru County was 4049.67, 84% was obtained from the non-participants. The total field-use EIQ rating per individual pesticide ranged from 0.58 to 946.16, being lowest for Deltamethrin (0.58) and highest for Dimethoate (946.16). 146 ECONOMETRIC RESULTS Determinants of adoption of IPM technologies The first stage of the endogenous switching regression model was a probit model that evaluated factors that influenced adoption of IPM technologies (table ii). The variables in the table represent factors determining adoption of IPM technologies. Adoption is high in large scale production as it implies higher revenue generation from mango production. Irrigation water access was significant at 5% with a coefficient of 0.326 as it boosts production. Farmer group members unexpectedly, are less likely to adopt IPM technologies despite access to information. Farmers with larger farm sizes are likely to adopt IPM technologies as they could be used as collateral to access credit. With large farm sizes, production economies of scale are achieved hence higher income generation. This makes it affordable to procure technologies. Household heads age is negatively related to adoption of IPM. Determinants of health and environmental effects. The estimation results show a structural difference between the two groups because most of the variables that influenced health and environmental effects in the two regions were distinct (table iii). From the results, if the labor management ratio reduces by 1% the value of EIQ field use will reduce by 2.1 % implying a positive impact on the environment. Also, an increase in the number of mango trees by 1% will reduce environmental impact by 73% in the participant’s region. Household head age was also significant by increasing the negative health and environmental effects of pesticide use. For non-participants, household size and number of productive trees were significant in determining the magnitude of EIQ field use. However positive health and environmental impact of number of mango trees was higher for IPM participants. Health and environmental effects of IPM Adoption The predicted health and environmental effects were compared for the sub-sample with and without IPM; other factors held constant (Table iv). The ATU was computed, which indicated the difference between the mean health and environmental effects if they had adopted IPM technologies or not. IPM non-adopters would have a positive environmental impact from adoption. IPM adopters have reduced EIQ field use, health and environmental effects. In the second scenario, we assumed that all farmers have access to credit, which would reduce the value of EIQ field use by 28.38% among the adopters. The results showed that access to credit does have an impact on health and environmental effects since all adopters gained from strategy whether they had access to credit or not. Use of protective clothes reduced health and environment effects for IPM adopters. 147 TABLES Table i: Farmers perceptions on pesticide risks (Source: Field survey, 2014) Perceptions Overall (n=371) Non-IPM participants(n=165) IPM Participants (n=206) Health Large Risk Medium Risk Small Risk Very Toxic No risk at all Freq. % 150 39.4 98 22.4 57 17.6 42 13.9 24 6.7 Freq. % 65 39.4 37 22.4 29 17.6 23 13.9 11 6.7 Freq. % 85 41.3 61 29.6 28 13.6 19 9.2 13 6.3 Environment Large Risk Medium Risk Small Risk Very Toxic No risk at all Freq. % 106 25.5 95 26.1 97 23.1 12 4.9 61 20.6 Freq. % 42 25.5 43 26.1 38 23.1 8 4.9 34 20.6 Freq. % 64 31.1 52 25.2 59 28.6 4 1.9 27 13.1 Source: field survey data 2014 Table ii: Probit models for determinants for IPM adoption (Source: Field survey, 2014) Independent probit Joint estimated probit Variables Coefficien t SE Coefficient SE Household size (count) Years in Mango farming (years) Household head’s age (years) Agricultural land owned (acres) Household head’s gender (dummy) Extension contact (dummy) Protective clothe usage (dummy) Group membership (dummy) Credit services access (dummy) Irrigation water access (dummy) Labour management (ratio) Mango production trees (count) Monthly income (Kshs) Educational level (school years) Constant Total observations -0.236 0.214*** -0.000** 0.226** 0.050 -0.071 0.179 -0.479*** 0.001 0.326** 0.002 0.505*** -0.010 -0.004 -1.067 0.152 0.075 0.000 0.104 0.207 0.156 0.241 0.169 0.177- 0.153 0.005 0.741` 0.018 0.019 0.609 370 -0.128 0.131* -0.002*** 0.064 -0.022 -0.070 0.167 -0.486** 0.043 0.143 0.001 0.516*** -0.000 0.003 -1.167** 0.148 0.068 0.000 0,122 0.182 0.163 0.142 0.156 0.161 0.138 0.005 0.738 0.016 0.014 0.574 370 NB: Statistical significance at 0.01(***), 0.05(**) and 0.1 (*) Source: field survey data 2014 148 Table iii: Full information maximum likelihood parameter estimates for health and environmental effects (Source: Field survey, 2014) IPM Participants IPM Non- participants Variable Coefficient S.E Coefficient S.E Household size (count) 0.389 0.264 -0.435* 0.254 Years in mango farming (years) -0.019 0.125 0.012 0.105 Household head age (years) 0.000*** 0.000 -0.000 0.000 Agricultural land owned (acres) -0.182 0.233 0.135 0.194 Household head age(dummy) 0.435 0.327 -0.319 0.288 Extension officers access (dummy) 0.177 0.299 0.286 0.267 Protective cloth usage (dummy) 0.009 0.262 -0.379 0.239 Group Membership (dummy) 0.562** 0.269 0.083 0.291 Credit services access (dummy) 0.070 0.278 0.244 0.265 Irrigation water access(dummy) -0.411 0.258 0.206 0.243 Labor Management(ratio) -0.021* 0.012 0.002 0.005 Mango production trees (count) -0.730** 0.131 -0.378** 0.167 Constant 2.717 0.972 4.995 0.937 lnϭ1, lnϭ0 0.693(0.059)*** 0.355(0.061)*** ρ1ρ0 -1.89(0.45)*** -0.179(0.268) LR test of independent questions 35.25** Total observations 371 Log likelihood -335.287 ***, **, * mean values are significant at 1%, 5% and 10% respectively Source: field survey data 2014 149 Table iv: Treatment effects of IPM technologies (Source: Field Survey, 2014) Observations Adopters Non- Adopters Net change (%) EIQ Field use EIQ Field use All mango farmers 371 5.65 47.12 -6.81*** By land holding Mango farmers<1 acres 18 0.10 74.17 -160.45*** Mango farmers 1-2 acres 100 11.64 45.39 -14.17*** Mango farmers>2 acres 253 5.21 43.09 -31.65*** Protective Clothes use Yes 205 6.06 44.16 -27.21*** No 165 5.23 49.62 -31.53*** Credit Facilities Access Yes 95 5.81 57.19 -28.38*** No 276 5.75 43.92 -28.67*** Attend ICIPE’s Training Yes 234 5.27 0 -31.29*** No 137 7.36 35.244 -22.40*** ***, **, * mean values are significant at 1%, 5% and 10% respectively Source: field survey data 2014 150 Table v: Environmental Impact Quotient Field Use (Source: Field survey, 2014) Active Ingredient Trade Name EIQ F EIQ C EIQ E EIQ T Rate (kg/acr e) EIQ Field use overa ll EIQ Field non- parti cipan ts EIQ Field use particip ants T value P value % of farmer s using Vol.kg Thiamethoxam (U) Actara 10.3 5 12.03 77.5 2 33.3 0.12 5.75 3.55 2.19 0.7891 0.2341 0.81 1.14 Methomyl (IB) Agrinate 6 11 75 31 0.28 277.38 221.1 2 56.27 0.2389 0.8805 6.20 452.31 Propineb (1II) Antracol 6 5.78 14 18.34 0.27 30.93 0.10 30.82 1.2615 0.5807 6.20 534.42 Triadimefon (III) Bayleton 12.1 5 15.15 53.5 7 33.3 0.25 277.31 105.0 4 172.27 - 2.9994*** 0.0033 34.50 134040 .8 Beta-Cyfluthrin (II) Bulldock 9 4 69 27 0.21 634.28 550.4 2 83.86 2.6674*** 0.0087 33.96 69418. 38 Copper Oxychloride (III) Copper 108 19 76 67.7 0.20 557.61 512.4 9 45.12 3.0990*** 0.0035 12.12 9444.3 2 Cypermethrin (II) Cyclone 9 4 69 27 0.14 22.75 12.14 10.61 -1.5302 0.1482 4.58 173.30 91 Dimethoate (II) Danadi m 72 9 141 74 0.22 76.91 46.79 30.11 -0.5301 0.6104 2.67 35.25 Deltamethrin (II) Decis 6 3 68 26 0.26 0.58 0.21 0.37 0.2350 0.4325 2.43 101.10 Dimethoate (II) Twigath oate 72 9 141 74 0.21 946.16 794.1 5 152.01 1.3425 0.7856 8.63 2855.7 65 Mancozeb (U) Dithane 12 3 29 44 0.40 532.53 515.9 5 16.56 0.0955 0.9256 6 2.43 90.54 Lambda Cyhalothrin (II) Karate 21 3 106 44.17 0.09 7.96 7.12 0.84 0.1144 0.9100 6.20 84.38 Methomyl (IB) Weiling 6 11 75 31 0.31 56.45 38.27 18.18 -1.9925* 0.0866 2.42 69.92 Propineb (III) Milraz 6 6 14 9 0.17 1.54 0.00 1.54 0.8745* 0.0534 1.35 5.07 151 Acephate (III) Orthene 15 12.5 47.1 5 24.88 0.19 16.81 12.59 4.22 0.7131 0.4875 4.58 230.25 Carbendazim (U) Rodazim 25 40.5 86 50.5 0.26 490.33 485.5 4 4.79 -0.7419 0.4752 3.23 28.57 Alpha- cypermethrin (II) Tata alpha 21 3 106 44 0.30 39.62 39.62 0.00 0.1451 0.8862 5.39 34.38 Sulphur (U) Thiovit 10 6 120 45.5 0.32 24.09 14.12 9.97 0.7117 0.4835 7.00 1427.0 1 Imidacloprid (II) Thunder 6.9 10.35 92.8 8 36.71 0.21 36.36 35.51 0.85 -1.1227 0.2722 7.27 435.39 Thiophanate- Methyl (U) Topsin 16.2 15.3 39.9 5 23.83 0.10 14.32 11.30 3.02 -0.58546 0.5796 2.16 2.87 Total 4049.6 7 3410. 27 639.40 - 7.7660*** 0.000 100 219465 .20 NB: Statistical significance at 0.01(***), 0.05(**) and 0.1 (*) 152 CONCLUSIONS The need for sustainable and environmental friendly agriculture has promoted invention of new techniques leading to increased mango production and access to lucrative markets. There is no study that has estimated health and environmental effects of IPM technologies. In regards, this gap can be bridged by estimating health and environmental effects of IPM technologies. Based on primary survey data of mango farmers in Meru County, we find that farmers with less than one acre of land had more benefits in adopting IPM technologies. The results indicated that IPM technologies have positive benefits on health and environment. ACKNOWLEDGEMENTS Special thanks the Department of International Development (DFID) for the financial support. Chris Miyinzi acknowledges the opportunity for research fellowship offered by the African Fruit Fly Programme (AFFP) of the International Centre of Insect Physiology and Ecology (ICIPE). REFERENCES Alene, A. D., & Manyong, V. (2007). The effects of education on agricultural productivity under traditional and improved technology in northern Nigeria: an endogenous switching regression analysis. Empirical Economics, 32(1), 141–159. Chitra, G. A., Muraleedharan, V., Swaminathan, T., & Veeraraghavan, D. (2013). Use of pesticides and its impact on health of farmers in South India. International Journal of Occupational and Environmental Health. Colette, W. A., Almas, L. K., & Schuster, G. L. (2001). Evaluating the impact of integrated pest management on agriculture and the environment in the Texas panhandle. Presented at the WAEA annual meeting, Logan, UT. Edwards, C. (2013). Environmental pollution by pesticides (Vol. 3). Springer Science & Business Media. Fan, L., Niu, H., Yang, X., Qin, W., Bento, C. P., Ritsema, C. J., & Geissen, V. (2015). Factors affecting farmers’ behaviour in pesticide use: Insights from a field study in northern China. Science of the Total Environment, 537, 360–368. Griesbach, J. (2003). Mango growing in Kenya. World Agroforestry Centre. HCDA. (2013). Horticultural Crops Production Report. Government Printers, Nairobi. Kibira, M., Affognon, H., Njehia, B., Muriithi, B., Mohamed, S., & Ekesi, S. (2015). Economic evaluation of integrated management of fruit fly in mango production in Embu County, Kenya. African Journal of Agricultural and Resource Economics Volume, 10(4), 343–353. Klungness, L., Jang, E. B., Mau, R. F., Vargas, R. I., Sugano, J. S., & Fujitani, E. (2005). New sanitation techniques for controlling tephritid fruit flies (Diptera: Tephritidae) in Hawaii. Kovach, J., Petzoldt, C., Degni, J., & Tette, J. (1992). A method to measure the environmental impact of pesticides. 153 Lokshin, M., & Sajaia, Z. (2004). Maximum likelihood estimation of endogenous switching regression models. Stata Journal, 4, 282–289. Macharia, I. N., Mithöfer, M., & Waibel, H. (2009). Potential environmental impacts of pesticides use in the vegetable sub-sector in Kenya. African Journal of Horticultural Science, 2. Muriithi, B. W., Affognon, H. D., Diiro, G. M., Kingori, S. W., Tanga, C. M., Nderitu, P. W., Muhamed S., Ekesi, S. (2016). Impact assessment of Integrated Pest Management (IPM) strategy for suppression of mango-infesting fruit flies in Kenya. Crop Protection, 81, 20–29. Pradel, W., Forbes, G., Ortiz, O., Cole, D., Wanigarantne, S., & Maldonado, L. (2008). Use of the Environmental Impact Quotient to estimate impacts of pesticide usage in three Peruvian potato production areas. Sin Publicar, 32. Relyea, R. A. (2005). The impact of insecticides and herbicides on the biodiversity and productivity of aquatic communities. Ecological Applications, 15(2), 618–627. 154 4. Parallel Session 3 : Crop Management, Biotechnology and Climate Change 4.1 Bostrichidae beetles associated with Acacia xanthoploea at Kenyatta University and Mitaboni in Nairobi and Machakos Counties, Kenya Kahuthia-Gathu, R.1, Kirubi, D. T.1, Wangu, L.1 and Kimani R.1 1 Kenyatta University, Department of Agricultural Science and Technology, PO Box 43844-00100 Nairobi Corresponding author; rkahuthia@gmail.com ABSTRACT The Naivasha thorn tree Acacia xanthoploea (Fabaceae: Mimosoideae) is an important tree found in areas with high water tables in Kenya. They are grown for foliage, timber, shade and rehabilitation of soils. Its production is threatened by both abiotic and biotic factors especially the insect pests which are a major cause of losses. The invasive powderpost beetle is one of the insect pests posing a major threat to acacia and other forest resources in the Country. Despite this, very little is documented on the powderpost beetles, their damage and natural enemies in Kenya. The study was conducted to evaluate the occurrence and damage of powderpost beetles in the family Bostrichidae on Acacia xanthoploea. Infested wood samples of A. xanthoploea with fresh exit holes were collected from three sites in Kenyatta University (KU) and Mitaboni Kenya. The samples were placed in clear plastic buckets and taken to the laboratory where they were kept at ambient temperatures and observed daily for emergence of powderpost beetle. The beetles were collected every three days, identified and recorded. The experiment was replicated four times and data collected twice a week for 6 months. Data on abundance was subjected to analysis of variance using SAS. Over 5,850 and 4,691 insects were recovered where 2,187 and 3,097 were Bostrichidae which accounted for 37% and 66% in KU and Mitaboni, respectively. A total of 12 species of powderpost beetle in the Family Bostrichidae were recovered which included Sinoxylon ruficorne, S. doliolum, Xylion adustus, Xyloperthodes nitidipennis, Xyloperthella picea, Xylopsocus castanoptera, Lyctus brunneus, Heterbostrychus brunneus, Xylopsocus sp., and Dinoderus gabonicus. The most abundant species were Xylion adustus in KU with 1,915 beetles accounting for 88.4% and Sinoxylon ruficorne in Mitaboni with 1,050 beetles comprising 33.9%. Sinoxylon ruficorne was only recorded in Mitaboni while 2 specimens of D. gabonicus were found in KU only. The mean number of exit holes on A. xanthoploea differed significantly between sites which coincided with the damage by the beetles. Therefore, there is urgent need to document the abundance of powderpost beetles in Kenya and quantify the damage caused. Since a number of parasitoids were recovered, there is need to conduct more studies and document the species diversity, parasitism rates and efficacy trials as possible bio-control agents. Key words: Bostrichidae, powderpost beetles, species, abundance, damage 155 INTRODUCTION Acacia xanthophloea belonging to the pod bearing plant family Fabaceae, subfamily Mimosoideae is commonly known as sulphur bark, Naivasha thorn or fever tree in English, Mgunga in Swahili, and Murera in the Kikuyu dialect of Kenya. There are over 40 species, subspecies and varieties of Acacia represented in Africa. The fever tree is an attractive, semi-deciduous to deciduous tree approximately 15 to 25 meters tall, with open rounded to spreading or flattish crown which is sparsely foliated. They have characteristic lime green to greenish-yellow bark is smooth, slightly flaking and coated in a yellow powdery substance with straight white paired thorns, up to 7 cm, white and clustered leaves on short lateral shoots of 4-6 pairs of pinnae. Flowers are axillary clusters of pale yellow spherical heads while the pods are straight, papery with a wavy margin and indehiscent (Dharani et al., 2006; Gathuru, 2011). Acacia xanthophloea is native to Botswana, Kenya, Malawi, Mozambique, Somalia, South Africa, Swaziland, Tanzania, Zambia, Zimbabwe (Orwa et al., 2009). It grows in semi- evergreen bush land and woodland in areas with a high groundwater table near swamps, riverine forests or at lakesides and often forms dense stands in seasonally flooded areas. It is able to tolerate several degrees of frost. Other than growing naturally, the species is grown for foliage and pods provide food for livestock, young branches and leaves are eaten by elephants, leaves and pods by giraffes and velvet monkeys (Dharani et al., 2006). It has been used in apiculture as the trees produce good bee forage, source of firewood, timber, shade, nitrogen fixing, acts as live fences, provide nesting sites for birds, (Orwa et al., 2009) and for charcoal production especially in areas of its widespread occurrence (Okello et al., 2001). Today, the species is also being grown as an ornamental tree. The roots and the bark of stem have been used as medicine against malaria due to their pharmacological properties (Orwa et al., 2009). They have also been used in rehabilitation of limestone quarries in Athi River, Kenya (Gathuru, 2011) and to curb soil erosion. This results in the increase of biodiversity of plants, animals and restoration of nature. The species are faced by a number of biotic and abiotic constraints that interfere with the good prospects from the tree. They are always under the influence of biotic factors such as pests and diseases and abiotic factors such as drought, water-logging (Wargo, 1996; Hakeem et al., 2012). This makes them more susceptible to attack by wood boring insects (Thomas et al., 2002) which are generally xylophagous. Many xylophagous insects have symbiotic protozoa and/or bacteria in their digestive system which assists in the breakdown of cellulose. Though the vast majority of powderpost beetles are ecologically important and economically benign, some species are economic pests attacking relatively healthy trees, untreated wood, and downed trees in lumber yards. However, some are invasive species that threaten natural forest ecosystems. Plant-insect interactions are quite common in nature and arthropod pests cause direct damage to the plants or their products. Some of the key pests include wood boring beetles in the family Lyctidae or true powder post beetles, Anobiidae or deathwatch beetles, and Bostrichidae or false powder post beetles among others. There are approximately 570 species in 89 genera worldwide (Lawrence, 2010), with over 40 156 species of insects documented to be associated with Acacia species. Bostrychidae is the most notorious Family in locating and infesting freshly cut wood (Orwa et al., 2009). They are often referred to as powderpost beetles because of their ability to reduce wood to a thin external shell covering the frass produced by the boring activities of the larvae and adults (Beaver et al., 2011). Generally, Bostrichid beetles prefer to infest dry wood with relatively low moisture content between 8 and 30% (Lek-Uthai et al., 1982; Ivie, 2002). Despite the risk of pest introduction from traded agricultural products being manageable, many biological invasions continue to occur worldwide EEA (2012). Some species have been dispersed around the world by trade in timber and timber products (Beaver et al., 2011) and many invasions are unavoidable due to the pest characteristics (Lopes-da-Silva et al., 2014). The powderpost beetles attack the wood at all stages from log to seasoned wood and finished products (Findlay, 1985). They have been recorded in forests, timber depots, sawmills and furniture industries, buildings, seasoned timber, wood artefacts, boxes and packing cases (Ivie, 2002). They obtain nutrition from starch, enabling many species to utilize almost any dry wood material from enormous host range including trees, shrubs, herbaceous plants and bamboo (Ivie, 2005; Fihlo et al., 2006). Most species are polyphagous and attack a wide variety of host plants in many different families (Beaver et al., 2011). They prefer wood with high starch content (Creffield, 1991; Peters et al., 2002; Akhter, 2005; Allen, 2005) with relatively low moisture, or dried wood (Gerberg, 1957; Cookson, 2004; Ivie, 2002). However, some species are specialist and tend to be confined to a single genus or species of a host plant. The powderpost beetles, the auger beetles or the horned powderpost beetles (Coleoptera: Bostrichidae) are serious xylophagous pests of trees, forest products, silviculture, agricultural crops and stored vegetable products in most regions of the world, especially in tropical countries (Nair, 2007; Liu et al., 2008; Lawrence, 2010). Polyphagous pests inflict greater damage on plants in mixed vegetation systems compared to monophagous pests. Some species have also become pests of stored grain and root crops (Fisher, 1950; Lawrence, 2010). However, studies of their ecology are rare. Most of these pests have been introduced accidentally from their native countries possibly via wooden crates or pallets for cargo shipment. The genus Sinoxylon is native to Europe, the Afro-tropica; region, the Australian region, the Near East, the Neartic, the Neoptropical region, North Africa and the Oriental region (Borowski and Wegrzynowicz, 2007; Beaver et al., 2012) with over 52 species in within the genera. It is one the most intercepted species in many countries (Price et al., 2011; Kevin et al., 2012). It also commonly found in tropical areas, especially in the arid regions, where they damage wide range of trees (Bushara, 1981). The powderpost beetles have a very wide host range and attack any kind of wood both softwood and hardwoods with suitable moisture and starch content. However, they have preference for hardwoods due to presence of pores that enable easy penetration and by larvae and laying of eggs by adults. Some of the families include Ullmaceae, Euphorbiaceae, Lauraceae, Legumnoceae, Mimosaceae, Dipterocarpaceae, Anacardiaceae, Rubiaceae and Poaceae Ulma-Euphorbiaceae, Lauraceae, 157 Dipterocarpaceae, minosaceaae, Anacardiaceae, Rubiaceae among others (Fisher, 1950; Poggie et al., 1994; Peres Filho et al., 2006). Sinoxylon species are particularly important pests of trees, wood and bamboos (Fisher, 1950; Teixeira et al., 2002). Sinoxylon conigerum has been recorded attacking rubber wood Hevea brasiliensis Müll., pigeon pea Cajanus cajan (L.), mango tree Mangifera indica L., cassava stem Manihot esculenta Crantz, fallen trunk of flamboyant Delonix regia (Bojer), cotton dry roots Gossypium hirsutum L., branches of guava Psidium guajava L., and teak (Tectona. grandis) (Balasubramanya et al., 1990; Binda and Joly, 1991; Tomimura, 1993). Fischer (1950), recorded S. conigerum attacking Adina cordifolia Benth., and Hook., Albizzia amra (Roxb.) Boiv., Holoptelea integrifolia (Roxb.) Planch., M. indica, Shorea robusta Gaertn., Terminalia bialata (Roxb.) Steud., Terminalia myriocarpa Van Heurck & Mull. Arg., Grewia tiliaefolia Vahl,. Sinoxylon senegalense is the most notorious pest of the white thorn acacia Acacia seyal (Fabaceae Mimosoideae) native to Kenya (Orwa et al., 2009). Acacia tortilis is also infested by Sinoxylon crissum and auger beetle S. anale. Sinoxylon anale and S. unidentatum are the most destructive powder post beetles of rubber sawn wood in southern Thailand (Halperin and Geis, 1999; Liu et al., 2008). Sinoxylon unidentatum has also been recorded attacking Ulma-Euphorbiaceae, Lauraceae, Dipterocarpaceae, minosaceaae, Leguminoseae, Anacardiaceae, Rubiaceae (Fisher, 1950, Poggie et al., 1994, Peres Filho et al., 2006). Sinoxylon anale is a significant pest of agricultural, forestry and forest product industries as they are very harmful to trees, bamboos and wood. The false powderpost beetle Xylopsocus species has been recorded in the wood of Morus alba L. trunks of “Kimoungoue” in east Africa, bamboo, litchi Litchi chinensis Sonn., and grape Vitis sp. in Brazil (Lesne, 1924; Fisher, 1950). Miller (1934) recorded X. capucinus attacking piquia (Caryocar villosum Pers.), rambutun Nephelium lappaceum L., rubber Hevea brasiliensis (Willd. ex A. Juss) Muell. Arg.), Dipterocarpus sp., Derris sp., guava Psidium guajava L., Casuarina equisetifolia J.R. Forst. & G. Forst., Syzygium malaccensis (L.) Merrill & L.M. Perry, cinnamon Cinnamomum zeylanicum Blume, Indigofera anil L., cashew nut Anacardium occidentale L., durian Durio zibethinus J. Murr., Tephrosia candida DC., Gardenia jasminoides Ellis, tamarind Tamarindus indica L., Hibiscus Rosa sinensis L. Because of the ability of the larvae and some adults to reduce wood into a powdery frass, the beetles are of considerable economic importance to forestry and the wood- using industries, especially in tropical countries (Liu et al., 2008; Beaver et al., 2011). The damage is typically caused by the larvae and adults boring in the stems, branches, twigs of dead, damaged or stressed host (Halperin and Geis, 1999; Peters et al., 2002; Nair, 2007; Liu et al., 2008)). One of the most characteristic signs of infestation are the exit holes produced by the adults beetles at emergence which are perfectly circular and the diameter depends on the wood-boring species (Herald et al., 2009). Other signs include death, dieback, exudation of gummy sap or resin and early branching depending on the infested trees, degree of resistance and the beetle species. They infest weakened or dead trees and branches and gradually render them into fine dust. The Bostrichid beetles prefer to infest dry wood with relatively low moisture content between 8 and 30% (Lek- Uthai et al., 1982; Ivie, 2002). 158 The female adults deposit eggs in cracks, crevices, pores or old emergence holes in wood, or in tunnels made by the females. The eggs takes 1 to 2 weeks to hatch into a tiny larva that burrows and tunnel through the host tissues making extensive frass- filled galleries (Obeid, 1998). The larva undergoes a number of instars, tunnel through the host tissues making extensive frass-filled galleries (Beaver et al., 2011) and burrows toward the surface to pupate. The larval period lasts between 7 to 10 weeks while the pupa period can last 3 to 4 weeks. After the adult emergence, it continues the tunnel to the surface making the exit holes, mates and the females return to the wood to lay eggs. The females are usually larger than the males (Bushara, 1979; El Obied, 1998). The life cycle may range from 3 months to years depending temperatures, relative humidity, type and species of the powder post beetle. In the tropics, there are usually one to four overlapping generations per year (Liu et al., 2008a). According to Monks (2000), there are two practical usefulness of the study of local species that have unique properties. They are first for the protection and use of biodiversity, and secondly for assessment of the impacts that may arise due to changes in the environment. Some components that determine a species as indicators are the diversity of taxonomy, abundance, distribution, the role of these types, and the extent to which these types are easily recognizable (Monks 2000). Some families also play an important ecological role in decomposition and nutrient cycling in forest communities (Harmon et al., 1986). A number of appropriate management strategies have been used in the mitigation of the powderpost beetles. They include delimitation of infested areas, exclusion and prevention through regulatory restriction of movement of pest infested wood or plant materials, moisture control, wood replacement, surface insecticide treatment, physical destruction of infested trees, and fumigation. Biological control via introduction and release of natural enemies collected from native ranges (Liu et al., 2003; Taylor et al., 2011). The selection of management strategies depends on a number of factors such as severity of infestation, area being attacked, potential for re-infestation and treatment costs. Predators in the family Cleridae, Histeridae, Melyridae, Trogossitidae) (Lesne, 1906, 1924; Bahillo de la Puebla et al., 2007; Kolibáč, 2013, Lawrence and Slipinski, 2013). Predation by birds also occurs (Bahillo de la Puebla et al., 2007). Hymenopteran parasitoids especially Pteromalidae and Braconidae are known to parasitize beetles in the Family Bostrichidae (Gerstmaier et al., 1999; Bahillo de la Puebla et al., 2007; Yu et al., 2012; May, 2015; Noyes, 2015). According to Deepthi and Ramavedi (2012), ectoparasitoids in the family Passandridae are known to attack the powderpost beetles. Few studies have been conducted on biological control because predators and parasitoids are not known to have a significant effect on beetle populations in natural conditions (Katiyar and Sharma, 1987; Edde, 2012; May, 2015). By contrast, ecosystem services related to biotic interactions, such as pollination or pest control, could be predominantly driven by species richness (Galiballdi et al., 2013). Scanty information is available about the insects associated with the multipurpose trees and shrubs that are gaining greater economic importance as components of agroforestry 159 systems. Therefore, the study aims to investigate the abundance, species composition, diversity of the powderpost beetles and their associated natural enemies both predators and parasitoids from A. xanthoploea growing around Kenyatta University and Mitaboni. The beetles were chosen because they are the most abundant and diverse group of insects in the environment as observed by Speight et al. (2008). In addition, the dynamics of the powderpost beetles and their natural enemies are governed by the complexity and composition of the agroforestry system. MATERIAL AND METHODS Study sites The survey was conducted in May 2016 from Kenyatta University (KU) and Mitaboni in Nairobi and Machakos County. Kenyatta University is located 15 Kilometers North of Nairobi at 1°11´ S and 36°55´ E at an altitude of 1,600 meters above sea level. The area experiences bimodal rainfall between 1,000-1,400 mm annually with average temperature range of 18-24°C and has clay murram soils. Mitaboni is 17 km North west of Machakos town at 1°21´ S and 37°14´ E at 1,525 m asl. The area receives annual rainfall between 600-900 mm, average temperatures of 27°C and the soils are comprised of clay and murram. Evaluation on abundance of powderpost beetles of Acacia xanthoploea Three sites from KU and Mitaboni where Acacia xanthoploea was growing were identified for the studies. Purposive sampling was used to collect the dry wood with inactive exit holes, active or freshly bored holes by the powderpost beetles and those showing damage symptoms (sawdust coming out). The wood was transversely cut into pieces measuring 15cm long 10 cm diameter or equivalent volume as describe by Sittichaya and Beaver (2009) and Kangkamanee (2011) and placed in 20 L plastic containers (Plate 2). The samples were taken to Kenyatta university Biotechnology laboratories and kept at ambient room temperatures of 23±1ºC, 60±10% RH, and 12 L: 12 D photoperiod and observed for a period of 8 months until no more beetles emerged. The emerging and positively phototropic adults of the powderpost beetles were collected every two days, placed in petri dishes observed under the dissecting microscope where the individual families and species of the powderpost beetles were separated using morphological and taxonomic characteristics, identified and numbers recorded. The Bostrichidae were identified as described by Wisut et al. (2009) and Liu et al. (2008a; 2016). The recent taxonomic nomenclature was adopted in the identification and confirmation of the jewel beetles in the family Buprestidae (Bellamy, 2013). The specimens were preserved in 95% ethanol and taken to the National Museums of Kenya (NMK) for confirmation while Bostrichidae were sent to Taiwan for confirmation by Dr. Liu lan-Yu. 160 Plate 1: Naivasha thorn tree Acacia xanthoploea (source Kahuthia R.) Evaluation of number of exit holes of the powderpost The cut pieces of A. xanthoploea were placed in well ventilated plastic buckets with lids and taken to Kenyatta University laboratories where they were kept at ambient room temperatures at 23±2C° at 50-65% RH. The number of active exit holes or freshly bored holes by powderpost beetles was counted and recorded (Plate 1). Plate 2: Exit holes of powderpost beetles on Acacia xanthoploea branch (source Kahuthia R.) Data analysis The number of beetles was subjected to one way Analysis of Variance (ANOVA) to estimate the mean infestation. The infestation density on acacia from each region at different collection dates will be computed. The data will be compared using the Mann– Whitney U test and Kruskal–Wallis test as described by Zar (1984) to determine any difference in infestation densities within time and regions. Results Families of powderpost beetles recorded on Acacia xanthoploea 161 Over 5,850 and 4,691 powderpost beetles were collected and examined from A. xanthoploea collected from KU and Mitaboni, respectively. A total of 17 and 13 families of the powderpost, wood boring beetles in the order Coleoptera and Sub-order Polyphaga were recovered from Nairobi and Mitaboni, respectively. Other than the powderpost beetles, predators and parasitoids were also recorded from the study sites. The families included Bostrichidae, Buprestidae, Bothridiidae, Curculionidae, Cleridae, Cerambycidae, Chrysomelidae, Ciidae, Colydiidae, Dermestidae, Histeridae, Lyctidae, Tenebrionidae, Staphylinidae, Scolytinae, Silvanidae, Laemophloeidae, and Elateridae (Tables 2 and 3). Most of the powderpost beetles were mainly from the Families Bostrichidae, Buprestidae, Curculionidae, Lyctidae and Histeridae. Kenyatta University recorded 16, 6 and 6 species of Bostrichidae, Curculionidae and Lyctidae, respectively. However, 14, 6 and 5 species of the families Bostrichidae, Lyctidae and Curculionidae were obtained from Mitaboni, respectively. Bostrichidae was by far the most rich-species family in the order Coleoptera followed by Curculionidae. A total of 2,187 and 3,097 powderpost beetles were Bostrichidae accounting for 37% and 66% in KU and Mitaboni, respectively (Table 1). At least 16 and 14 species of Bostrichidae powderpost beetles were recorded from KU and Mitaboni accounting for 51.79% and 73.49%, respectively (Tables 2 and 3). Other dominant families in KU included Lyctidae, Staphylinidae and Buprestidae. Table 1: Percent contribution of the family Bostrichidae Study sites Beetles collected Bostrichidae Percent contribution KU 5850 2187 37.4 Mitaboni 4691 3097 66 Table 2: Powderpost beetle families recovered from Acacia xanthoploea at Kenyatta University Families No of species Abundance Percent (%) Bostrichidae 16 2,189 51.79 Curculionidae 6 863 20.42 Lyctidae 6 307 7.98 Staphylinidae 2 353 8.35 Buprestidae 4 201 4.76 Cerambycidae 3 127 3.00 Anobidae 2 109 1.86 Laemophloeidae 2 27 0.64 Dermestidae 2 2 0.04 Silvanidae 1 1 0.02 Scolytidae 1 1 0.02 Bothrideridae 2 16 0.38 Colydiidae 1 23 0.54 Others 4 13 0.31 162 Table 3: Powderpost beetle families recorded from Acacia xanthoploea in Mitaboni Families No of species Abundance Percent (%) Bostrichidae 14 3,097 73.49 Curculionidae 5 898 21.31 Lyctidae 6 135 3.20 Buprestidae 1 2 0.05 Anobidae 1 1 0.02 Cerambycidae 2 11 0.26 Dermestidae 1 33 0.76 Silvanidae 1 2 0.05 Staphylinidae 3 21 0.5 Bothrideridae 1 2 0.05 Others 5 5 0.17 Abundance of Bostrichidae species recovered from A. xanthoploea from KU and Mitaboni Bostrichidae species included Sinoxylon ruficorne Fåharaeus, S. doliolum Lesne, Xylion adustus Fåharaeus, Xyloperthodes nitidipennis Murray, Xyloperthella picea Oliver, Xylopsocus castanoptera Fairmaire, Lyctus brunneus Stephens, Heterbostrychus brunneus Murray, Xylopsocus sp., and Dinoderus gabonicus Lesne. The most abundant species in KU were X. adustus followed by S. doliolum with 1,915 and 137 beetles accounting for 88.4% and 6.23%, respectively (Table 4). In Mitaboni, the abundant species were S. ruficorne, S. doliolum, X. picea and X. nitidipennis accounting for 33.9%, 19.9%, 15.5 and 10.3%, respectively. The species Sinoxylon ruficorne was only recorded in Mitaboni while D. gabonicus in KU. Table 4: Abundance of Bostrichidae species recorded from Acacia xanthoploea at Kenyatta University and Mitaboni No of species Kenyatta University Mitaboni No. beetles Percent (%) No. beetles Percent (%) Sinoxylon ruficorne - - 1,050 33.9 Xylion adustus 1,915 88.4 108 3.5 Sinoxylon doliolum 137 6.23 615 19.9 Xyloperthella picea 12 0.3 480 15.5 Xyloperthodes nitidipennis 98 2.2 319 10.3 Xylopsocus castanoptera 2 0.05 6 0.2 Lyctus brunneis 93 1.6 20 0.7 Dinoderus gabonicus - - 2 0.06 163 Powderpost beetle damage on Acacia xanthoploea There was significant difference (df=5,122; f=6.32; P<0.0001) on the mean number of exit holes made by the powderpost beetles on the pieces of wood of A. xanthoploea. It was also observed that samples from Mitaboni had significantly higher number of exit holes compared those recorded at Kenyatta University (Table 6). The mean number of exit holes was comparable in all the three sites at KU. In Mitaboni, site 2 had the highest number of exit holes while in KU, site 1 had the lowest of 66.4 and 31.1, respectively. The young adults remain tunneling and feeding in the wood for several days before emerging through small circular holes they make in the wood. Upon removal of the bark, the damage was exhibited as observed on plates 3, 4 and 5. Most of the pupae and adults are found just below the bark and cause significant damage on the wood reducing it to a powder. Table 6: Mean ±SE number of exit holes of powderpost beetles on Acacia xanthoploea at Kenyatta University and Mitaboni Region Site N Number of exit holes Kenyatta University 1 19 31.1±4.7c* 2 15 39.9±6.1bc 3 20 32.3±4.9c Mitaboni 1 33 41.7±3.7bc 2 17 66.4±7.1a 3 24 59.3±6.7ab *Means followed by the same letter in the same column are not significantly different at P≤0.05 Plate 3: Pupae of powderpost beetles on Acacia xanthoploea (source Kahuthia R.) 164 Figure 4: Adult powderpost beetles on Acacia xanthoploea (Source Kahuthia R.) Plate 5: Damage on A. xanthoploea by powderpost beetles DISCUSSION A number of families were recovered from wood samples of Acacia xanthoploea in KU and Mitaboni. They included Bostrichidae, Buprestidae, Histeridae, Curculionidae, Lyctidae, Chrysomelidae, Dermestidae, Tenebrionidae, Staphylinidae, Cleridae, Cerambycidae, Silvanidae, Laemophloeidae, Ciidae and Elateridae. Similar observations were made by Kuria et al. (2010) who recorded rich beetle communities from Acacia drepanolobium in Kenya which comprised of 13 beetle families mainly Curculionidae, Anthicidae, Cleridae, Buprestidae, Cerambycidae, Bostrichidae. These results are also comparable with those of Speight et al. (2008) who found high species richness with Bostrichidae being the most abundant and diverse group of insects in the environment. The abundance and diversity of the families on the A. xanthoploea were well adapted to the prevailing climatic and edaphic conditions in both KU and Mitaboni where the temperatures range between 21°C to 30°C. The diverse families and species could be related to high nutritional value from the species. Ivie (2005) and Fihlo et al. (2006) reported that the powderpost beetles obtain nutrition from starch, enabling many species to utilize almost any dry wood material from enormous host range including trees, shrubs, herbaceous plants and bamboo. 165 The preliminary study shows that powderpost beetles in the order Coleoptera suborder Polyphaga were the most abundant and diverse group of insects recorded in wood of Acacia xanthoploea from both Kenyatta University and Mitaboni. Xylion adustus was the most dominant species at Kenyatta University accounting for 52.3% of the total beetle recovered during the study. The species has also been reported in Tanzania, Malawi, Southern Africa and South Africa. However, there is no documentation on their abundance and species they infest in Kenya. Other than Bostrichidae, Buprestidae beetles were recorded with more abundant Agrilus species being recorded in KU accounting for 5.5% in KU while only 4 specimens were recorded in Mitaboni. Studies conducted by Vladimir and Georgi (2013) indicated that the Buprestidae fauna has over 429 species in Kenya. This additional information on Buprestidae is useful as per findings of Salikan and Georgi (2013) who also suggested the need for further detailed studies in order to enrich the knowledge on numbers and distribution of the buprestid species in Kenya It was observed that S. ruficorne and S. doloilum were the most abundant species in Mitaboni and accounted for over 47.8% and 17%, respectively. Singh (1995) also recorded Sinoxylon ruficorne on Acacia mearnsii in Kenya. However, its abundance was not quantified. The auguer beetle S. ruficorne were also found infesting cassava roots in Mozambiques, Nigeria and South Africa while S. doliolum is distributed in Tanzania in Africa (Hagstrum, 2009). In addition, X. castanoptera has also been recorded in East Africa, Madagascar and Mauritius. Dinoderus gabonicus which is usually distributed in West Africa was recovered from the A. xanthoploea at Kenyatta University. The powderpost beetles caused so much damage on A. xanthoploea which is one of the major trees which local population utilizes for timber, firewood, fencing and apiculture. There is need to conduct more studies to evaluate the damage of powderpost beetles can cause especially where the population use the timber for building with very little or no preservation. Similar findings were observed on Beetles belonging to the family Lyctidae are amongst the most important degraders of sapwood from seasoned hardwoods worldwide. With impending timber shortages and concerns over depleting forest resources in many countries, intense efforts have been made to maximize utilization of forest resources including lesser known or underutilized species, lower grade timbers and also to encourage a more widespread use of sapwood in timber products. Utilization of such materials which are often of lower resistance to wood degrading insects has led to greater prominence of M. rugicollis. CONCLUSION AND RECOMMENDATIONS In conclusion, these preliminary studies show that there are numerous families and species diversity of the powderpost beetles infesting the A. xanthoploea at Kenyatta University and Mitaboni areas of Kenya. Given the distances between these two sites, there is indication that the problem could widespread in the country. There is thus an urgent need to conduct more research on the biology, ecology, distribution, abundance and species diversity of powderpost beetles in areas of Kenya where A. xanthoploea are found. Since a number of predators and parasitoids families from were also recovered 166 in the study, more investigations on host specificity and efficacy of these parasitoids needs to be investigated so as to determine species that could be used in classical biological control or augmentation in order to keep the powderpost beetles below economic threshold level (ETL). ACKNOWLEDGMENTS We acknowledge all those who made this study a success. Special thanks go to Mr Matthew Sila Mbuko for allowing us to collect samples and conduct some preliminary studies from his farm. We also appreciate the management of Kenyatta University for allowing us to use the laboratories and all the technical staff for their assistance during processing of the materials. Sincere thanks to those who assisted in the confirmation of the species more so Dr. Liu Lan-Yu from Taiwan and Mr. Morris Mutua from the National Museums of Kenya (NMK) for assisting the identification and confirming of the powderpost beetles and natural enemies recovered from A. xanthoploea. Our appreciation goes to Dr. Nicholas Korir for his continuous assistance in data analysis and for his positive criticism during reviewing of the manuscript. REFERENCES Akhter, K. (2005). Preservative treatment of rubber wood (Hevea brasiliensis) to increase its service life. The International Research Group on Wood Protection, the 36th annual meeting, Bangalore, India, April 2005, 24-28. Allen, D.C. (2005). Wood Destroying Insects: The Powder Post Beetles. The New York Forest Owner. 38, 1. Bahillo de la Puebla, P., López-Colón, J. I. and Baena, M. (2007). Los Bostrichidae Latreille, 1802 de la fauna íbero-balear (Coleoptera). Heteropterus Revista de Entomologia 7: 147–227. Balasubramanya, R. H., Shaikh, A. J., Paralikar, K. M. and Sundaram, V. (1990). Spoilage of cotton stalks during storage and suggestions for its prevention. Journal of Indian Society of Cotton Improvement, 15: 34-39 Beaver, R. A., Sittichaya, W. and Liu L. Y. (2011). A Review of the Powder-Post Beetles of Thailand (Coleoptera: Bostrichidae) Tropical Natural History 11(2): 135-158. Bellamy C. L. (2013). The world of Jewel Beetles (Insecta: Coleoptera: Buprestoidea). Last updated: 18 January 2013. URL:http://www.fond4beetles.com/Buprestidae/index.html. Bellamy C. L. (2008-2009). A world catalog and bibliography of the Jewel Beetles (Coleoptera: Buprestoidea). Pensoft, Sofia-Moscow, vol. 1-5, 3263 pp. Bellamy, C.L. (2002). Coleoptera. Volume 29: Buprestoidea. Collingwood, Victoria: CSIRO Publishing. pg. 347. ISBN 9780643069008. Belokobylskij, S.A. (1992). On the classification and phylogeny of the braconid wasps of subfamilies Doryctinae and Exothecinae (Hymenoptera, Braconidae). Part I. On the classification 1. Entomologicheskoe Obozrenie, 71 (4), 900–928. 167 Belokobylskij, S.A., Yurchenko, G.I., Strazanac, J.S., Zaldívar-Riverón, A. and Mastro, V. (2012). A new emerald ash borer (Coleoptera: Buprestidae) parasitoid species of Spathius Nees (Hymenoptera: Braconidae: Doryctinae) from the Russian Far East and South Korea. Annals of the Entomological Society of America, 105 (2), 165– 178. Binda, F. and Joly, L. J. (1991). Los Bostrichidae (Coleoptera) de Venezuela. Boletin de Entomologia Venezolana. 6: 83-133. Borowski, J. and Wegrzynowicz, P. (2007). World Catalogue of Bostrichidae (Coleoptera). Wydawnictwo Mantis, Olsztyn, Poland. 247 pp. Bushara, M. G. (1979). Unpublished some aspects of the biology of the Acacia seyal woodborer Sinoxylon senegalense (Karsch). (Bostrychidae,Coleoptera), M.Sc. Thesis faculty of science University of Khartoum. Bushara, M. G. (1981). Short notes on the biology of the talih wood-borer Sinoxylon senegalense Karsch (Bostrychidae) (Coleoptera). Sudan- Silva, 4, 25–33. Cookson, L. J. (2004). Treatment Methods for the Protection of Hardwood Sapwood from Lyctine Borers. Forest & Wood Products Research and Development Corporation, Australia. Creffield, J. W. (1991). Wood Destroying Insects, Wood Borers and Termites. CSIRO Publications, East Melbourne,Australia, 44 pp. Curletti, G. and Sakalian, V. (2009). Sixteen new species of the genus Agrilus Curtis, 1825 from East Africa (Coleoptera: Buprestidae) ZooKeys, 24: 1-29. David Hagstrum 2009 Stored-Product Insect Resource Deepthi, T. R. and Remavedi, O. K. (2012). New report of Aulonosoma insignis Grouvelle (Coleoptera: Passandridae) as a larval ectoparasitoid of Sinoxylon anale Lesne (Coleoptera: Bostrichidae). Journal of Biological Control 26: 75–76. Dharani, N., Kinyamario, J.I. and Onyari. J.M. (2006). Structure and Composition of Acacia xanthophloea woodland in Lake Nakuru National Park, Kenya. Africa Journal of Ecology. 44 (4):523-530. EEA. (2012). The Impacts of Invasive Species in Europe. EEA Technical Report 16. El Obied, A. E. (1998). The Impact of Talh Wood-Borer Sinoxylon senegalense, Karsch Coleoptera, (Bostrychidae) on The Production of Charcoal Quantitatively and Qualitatively. M.Sc. thesis U. of Khartoum. Fihlo, O. P., Teixiera, E. P., Bezerra, M. L.M., Dorval A., and Berti Fihlo, E. (2006). First record of Sinoxylon conigerum Gerstacker (Coleoptera: Bostrichidae) In Brazil. Neotropical Entomology. 35 (5): 712-713. Findlay, W.P.K. (1985). Preservation of Timber in the Tropics, Martinus Nijhoff/W. Junk Publishers, Dordrecht, The Netherlands, 273 pp. Fisher, W. S. (1950). A Revision of the North American Species of Beetles Belonging to the Family Bostrichidae. USDA Miscellaneous Publication. 698: 1–157. Gathuru, G (2011). The performance of selected tree species in the rehabilitation of a limestone quarry at East African Portland Cement Company land Athi River, Kenya. Kenyatta University PhD Thesis 168 Gerberg, E.J. (1957). A revision of the New World species of powder-post beetles belonging to the family Lyctidae. United States Department of Agriculture Technical Bulletin. 1157, 1-55. Gerstmaier, R., Halperin, J. and Chekatunov, V. (1999). An annotated list of Cleridae and Thanerocleridae (Coleoptera) of Israel. Phytoparasitica 27: 26–33. Hakeem, K.R., Chandna, R., Parvaiz, A., Muhammad, I., and Munir, O. (2012) OMICS. A Journal of Integrated Biology. 16(11):621–663 Halperin, J. and Geis, K. U. (1999). Lyctidae (Coleoptera) of Israel, their damage and its prevention. Phytoparasitica. Israel Journal of Plant Protection Science, 27:257–262 Harmon, M. E., Franklin, J. F., Swanson, F. J., Sollins, P., Gregory, S. V., Lattin, J. D., Anderson, N. J., Cline, S. P., Aumen, N. G., Sedell, J. R., Lienkaemper, G. W., Cromack, K. and Cummins, K. W. (1986). Ecology of coarse woody debris in temperate ecosystems. Advances in Ecological Research, 15: 133–299. Ivie, M.A. (2002). Bostrichidae Latreille 1802. In American Beetles, R.H. Arnett, Jr., M.C. Thomas, P.E. Skelley and J.H. Frank, Editor. Volume 2, CRC Press, Boca Raton, pp. 233-244. Kangkamanee, T., Sittichaya, W., Ngampongsai, A., Permkam, S. and Beaver, R. A. (2011). Wood-boring beetles (Coleoptera: Bostrichidae, Curculionidae; Platypodinae and Scolytinae) infesting rubberwood sawn timber in southern Thailand. Journal of Forest Research16:302–308 Katiyar, K.N. and Sharma, V. (1987). Seasonal mortality of Prolatus aciculata Sharma, a larval parasitoid of Sinoxylon anale, due to hyperparasitoids. Transactions of the Indian Society of Desert Technology and University Centre of Desert Studies. 12: 27–28. Kevin, D. C., Schiefer, T. L. and Riggins, J. J. (2012). First Incidence of Sinoxylon Indicum and Sinoxylon Sudanicum (Coleoptera: Bostrichidae) in Mississippi. Florida Entomologist. 95(3):767-770. Kuria, S.K., Villet, M.H., Palmer, T.M. and Stanton, M.L. (2010). A comparison of two sampling methods for surveying mammalian herbivore impacts on beetle communities in the canopy of Acacia drepanolobium in Kenya. African Entomology Vol. 18 (1): 87-98. Lawrence, J. F. (2010). Bostrichidae Latreille 1802, pp. 209–217 In R. A. B. Leschen, R. G. Beutel and J. F. Lawrence [eds.], Handbook of Zoology, Arthropoda: Insecta, Coleoptera, Beetles. Vol. 2: Morphology and Systematics (Elateroidea, Bostrichiformia, Cucujiformia partim). Walter de Gruyter, Berlin, xiii+786 pp. Lawrence, J. F. and Ślipinśki, A. (2013). Australian Beetles. Volume 1: Morphology, Classification and Keys. CSIRO, Collingwood., 561pp. Lek-Uthai, P. (1982). Wood-boring insects in Thailand. In: Wood product section, forest research conference 1981, Bangkok, Thailand, pp 1–5 Lesne, P. (1906). Revision des coléoptères de la famille des Bostrychides. 5e Mémoire. Sinoxylinae. Annales de la Société Entomologique de France. 75: 445–561. Lesne, P. (1924). Les coléoptères bostrychides de l’Afrique tropicale française. Encyclopédie Entomologique, Series A, 3: 1–301. 169 Liu, H., Bauer, L.S., Gao, R., Zhao, T., Petrice, T.R. and Haack, R.A. (2003). Exploratory survey for the emerald ash borer, Agrilus planipennis (Coleoptera: Buprestidae), and its natural enemies in China. Great Lakes Entomologist. 116, 147–152. Liu, H., Petrice, T. R., Haack, R. A., Zhao, T., Bauer, L. S. and Gao, R. (2003). “Exploratory survey for the emerald ash borer, Agrilus planipennis (Coleoptera: Buprestidae), and its natural enemies in China,” Great Lakes Entomologist, Vol. 36 (3-4): 191– 204. Liu, L. Y., Schönitzer, K. and Yang, J. T. (2008a). A review of the literature on the life history of the Bostrichidae (Coleoptera). Mitteilungen der Münchener Entomologischer Gesellschaft, 98: 91- 97. Liu, L.-Y., Beaver, R.A. and Sanguansub, S. (2016). A new Oriental genus of Bostrichid beetle (Coleoptera: Bostrichidae:Xyloperthini), a new synonym and a lectotype designation for Octodesmus episternalis (Lesne, 1901). European Journal of Taxonomy 189: 1–12. http://dx.doi.org/10.5852/ejt.2016.189 Liu, L.-Y., Ghahari, H. and Beaver, R. A. (2016). An annotated synopsis of the powder post beetles of Iran (Coleoptera: Bostrichoidea: Bostrichidae). Lopes-da-Silva, M., Sanches, M. M., Stancioli, A. R., Alves, G. and Sugayama, R. (2014). The Role of Natural and Human-Mediated Pathways for Invasive Agricultural Pests: A Historical Analysis of Cases from Brazil. Agricultural Sciences, 5, 634- 646. May, R. (2015). Lyctus brunneus (Stephens, 1830), Pest insects of our cultural heritage- Insects known to cause damage to heritage works. http://www1.montpellier.inra.fr/CBGP/insectes-du-patrimoine/?q=en/fiche- insecte/lyctus-brunneus (Accessed 25 March 2016). Monk, K.A., De Fretes, Y. and dan Lelley, G.R. (2000). Ekologi Nusa Tenggara dan Maluku. Alih Bahasa Kartikasari, S. N. Jakarta: Prenhallindo. Nair, K. S. S. (2007). Tropical Forest Insect Pests: Ecology, Impact, and Management. Cambridge University Press, New York. pp 404. Noyes, J. S. (2015). Universal Chalcidoidea Database. Natural History Museum, London. http://www.nhm.ac.uk/jdsml/research- curation/projects/chalcidoids/indexValidName.dsml (Accessed 27 March 2016) Orwa, C, Mutua, A., Kindt, R., Jamnadass, R, and Anthony, S. (2009). Agroforestree Database: a tree reference and selection guide version 4.0 (http://www.worldagroforestry.org/sites/treedbs/treedatabases.asp) Peres Filho, O., Teixeira, E.P., Bezerra, M.L.M., Dorval, A. and Berti Filho, E. (2006). First record of Sinoxylon conigerum Gerstäcker (Coleoptera: Bostrichidae) in Brazil. Neotropical Entomology. 35, 712-713. Peters, B. C., Creffield, J. W. and Eldridge, R. H. (2002). Lyctine (Coleoptera: Bostrichidae) pests of timber in Australia: A literature review and susceptibility testing protocol. Australia Forestry 65: 107–119. Price, T., Brownell, K. A., Raines, M., Smith, C. L., and Gandhi, K. J. K. (2011). Multiple detections of two exotic Auger Beetles of the genus Sinoxylon (Coleoptera: Bostrichidae) in Georgia, USA. Florida Entomology. 94: 354–355. 170 Sakalian, V. and, Georgi G. (2013). New data about the diversity of Jewel Beetles (Coleoptera: Buprestidae) in Kenya. Acta Zoologica Bulgaria., 65 (4): 457-460. SAS Institute. 2009. Base SASVR 9.2 procedures guide. SAS Institute Inc., Cary, NC. Singh, R.M.P. (1995). Insect pests in Agroferestry. Report of a GTZ Fellowship, International Centre for Research in Agroforestry Nairobi, Kenya Working Paper No. 70: 1-70 Sittichaya, W. and Beaver, R. (2009). Rubberwood-destroying beetles in the eastern and gulf areas of Thailand (Coleoptera: Bostrichidae, Curculionidae: Scolytinae and Platypodinae). Songklanakarin Journal of Science and Technology 31:381–387 Sittichaya, W., Beaver, R. A., Liu, L.Y. and Ngampongsai, A. (2009). An illustrated key to powder post beetles (Coleoptera, Bostrichidae) associated with rubberwood in Thailand, with new records and a checklist of species found in southern Thailand. Zookeys 26:33–51 Speight, M.R., Hunter, M.D. and Watt, A.D. (2008). Ecology of Insects: Concepts and Applications. Wiley- Blackwell, Oxford. Taylor, P.B., Duan, J. J., Fuester, R.W., Hoddle, M. and Van Driesche, R. (2012). Parasitoid Guilds of Agrilus Woodborers (Coleoptera: Buprestidae): Their Diversity and Potential for Use in Biological Control. Hindawi Publishing Corporation Psyche Volume 2012, Article ID 813929, 10 pages doi:10.1155/2012/813929. Teixeira E. P., Novo J. P. S. and Berti Filho E. (2002). First record of Sinoxylon anale Lesne and Sinoxylon senegalensis (Karsch) (Coleoptera: Bostrychidae) in Brazil. Neotropical Entomology, 31(4), 651–652. Okello, B.D., O’connor, T.G. and Young, T.P. (2001). Growth, biomass estimates, and charcoal production of Acacia drepanolobium in Laikipia, Kenya. Forest Ecology and Management 142: 143–153. Thomas, F.M., Blank, R. and Hartmann, G. (2002). Abiotic and biotic factors and their interaction as causes of oak decline in central Europe. Forest Pathology 32,277– 307. Tomimura, Y. (1993). Chemical characteristics of rubberwood damaged by Sinoxylon conigerum Gerstäecker. Bulletin of the Forestry and Forest Products Research Institute. 365: 33-43. Vladimir, S. and Georgi, G. (2013). New Data About the Diversity of Jewel Beetles (Coleoptera: Buprestidae) in Kenya Acta Zoologica Bulgarica, 65 (4): 457-460 Wargo, P.M. (1996). Consequences of environmental stress on oak: predisposition to pathogens. Annales des Sciences Forestières, 53, 359–368. Wisut, S., Roger, A.B., Lan-Yu, L. and Aran, N. (2009). An illustrated key to powder post beetles (Coleoptera, Bostrichidae) associated with rubber wood in Thailand, with new records and a checklist of species found in Southern Thailand ZooKeys 26: 33–51 Yu D. S., Achterberg, C. van and Horstmann K. (2012a). World Ichneumonoidea 2011. Taxonomy, biology, morphology and distribution. http://www.taxapad.com/taxapadmain.php (Accessed 25 March 2016) 171 Yu, D.S., van Achterberg, C. and Horstman, K. (2012b) Taxapad 2012, Ichneumonoidea 2011. Database on flash-drive. Ottawa, Ontario, Canada. Available from: www.taxapad.com (accessed on 28 Nov 2014) Yu D.S.K. (2015). Taxapad Ichneumonoidea (version May 2009). In: Species 2000 & ITIS Catalogue of Life, 26th August 2015 (Roskov Y., Abucay L., Orrell T., Nicolson D., Kunze T., Flann C., Bailly N., Kirk P., Bourgoin T., DeWalt R.E., Decock W., De Wever A., eds). Digital resource at www.catalogueoflife.org/col. Species 2000: Naturalis, Leiden, the Netherlands. ISSN 2405-8858. Zar, J. H. (1984). Biostatistical analysis, 2nd edition. Prentice-Hall, Englewood Cliffs Zhao, T., Gao, R., Liu, H.H., Bauer, L.S. and Sun, L. (2005). Host range of the emerald ash borer, Agrilus planipennis Fairmaire, its damage and counter measures. Acta Entomology Sinica 48, 594–599. 4.2 Effect of Poly culture cropping on infestation of maize by the vectors of Maize lethal necrosis viruses in Bomet county of Kenya Namikoye, E. S.1, 2* Kariuki, G. 1Kinyua, Z.M. 2Githendu, M.W. 1 and Kasina, M. 2* 1Department of Agricultural Science and Technology, Kenyatta University, P.O. Box 43844-00100, Nairobi. 2Kenya Agricultural and Livestock Research Organization, P.O. Box 14733-00800, Nairobi *Correspondence: everlynekakai@yahoo.com,namikoye.samita@ku.ac.ke, 0716050008 ABSTRACT The presence of Maize lethal necrosis disease (MLND) has constrained maize production due to a co-infection of maize by the two viruses. This study was carried out to determine the potential of various cropping systems in managing vectors of maize lethal necrosis disease causing viruses. Treatments included companion crops elephant grass, coriander, pearl millet and sorghum grown as intercrops and border crops with maize. A randomized complete block design was used with four replicates for two seasons in Bomet County. Mean vector counts was analysed using ANOVA and means separated by LSD. The cropping system had no effect on thrips and aphids infestation as well as MLND severity on maize. Maize grown as a monocrop registered the highest significant disease incidence (66%). The one surrounded by sorghum border significantly registered higher cob percentage fill (79%), cob weight (126g) number of good seeds (213) as well as fresh weight (59g). Corn thrips mean infestation on companion crops levels was significantly higher on the control (maize) (46) followed by coriander border (16). Aphid infestation on coriander was significantly higher (22) as well as maize (15). Therefore, coriander has the potential to trap the corn leaf aphids which are vectors of sugarcane mosaic virus while sorghum border has a positive effect on the yield performance of maize. 172 Key words: Maize, corn thrips, corn leaf aphids, Maize lethal necrosis disease, Sugarcane mosaic virus INTRODUCTION Maize is an important staple food crop in Kenya with per capita consumption estimated at 98 kg per annum (de Groote et al., 2011). It is both a food security crop as well as a commercial crop in various parts of the country (Olwande, 2012). According to the global information network report (GIEWS, 2013), maize production in 2013 declined by more than 100,000 metric tons from the previous year, due to various constraints key among them the Maize lethal necrosis disease (MLND) (Makubi and Wangai, 2012). In Kenya the disease is as a result of co infection by two viruses, Maize chlorotic mottle virus (MCMV) and sugarcane mosaic virus (SCMV). (Wangai et al., 2012). The effect of this disease has been devastating with high infection rates and significant damage, this seriously affects the yields leading to complete crop loss (Adams et al., 2012; Wangai et al., 2012). Since it was first reported in Kenya, the disease has spread in Uganda, Tanzania and Rwanda (Makumbi and Wangai, 2013, Adams et al., 2014). Control of MLND poses serious difficulties due to the involvement of the two viruses. The MCMV is vectored by Frankliniella williamsi Hood while SCMV is vectored by Rhopalosiphum maidis Fitch which are challenging to manage. Intercropping has been found effective against thrips (Frankliniella occidentalis) (Kasina et al., 2006, Nyasani et al., 2012). Therefore this study focused on two aspects of companion crops tested as border and intercrops and their potential to prevent transmission of related viruses to the maize crop. MATERIALS AND METHODS Studies were conducted at farmer fields in Bomet East Sub County in Bomet County, Kenya. The trial consisted of plots measuring 7.5 m x 5.25 m arranged in a randomized complete block design with four replications. This was done in two seasons (December, 2014 to April 2015 and April 2015 to September 2015). Two factors were tested: Intercropping and Border cropping by use of the following treatments: Elephant grass (Pennnisetum purpurerum), sorghum (Sorghum vulgare), Coriander (corriandium sativum) and Pearl millet (Pennisetum glaucum) . Olerai maize Zea mays, variety was used as the control and main crop. Border crops and the outer lines of the intercrop were planted 2 weeks earlier while the inner rows of intercrops were planted at same time with maize. Maize lethal necrosis disease severity and incidence was carried on two rows per plot. Severity of the MLND was taken based on the scale by (Kinyua et al., 2015). At harvest, 10 maize cobs were randomly picked in every plot on two inner rows. The cobs were weighed, shelled and seeds counted and weighed. The fresh and dry weight of the kernels was also taken. 173 DATA ANALYSES Data were analysed using Gen Stat 17th edition. Analysis of variance was performed at 95% level of confidence limit. Skewed data were transformed by square root or log10 (x +1). Post hoc analyses were carried out using the Fishers Protected Least Significance Difference Test (LSD). RESULTS MLN vectors infestation levels and MLND severity on maize grown under various cropping systems in Bomet County Pooled data from both seasons showed no variance in the number of vectors and MLND severity occurring on maize grown under all the polycultures cropping systems (Fig 1). However the MLND incidence levels was least in maize surrounded by Elephant grass border while the Olerai maize as the control registered a higher MLND incidence level. Fig. 1. Infestation levels of Corn thrips and Aphids; MLND severity and incidence on maize grown under various polycultures cropping in Bomet County Kenya n o .o f ve ct o rs M LN D in ci d e n ce & se ve ri ty cropping system Corn thrips Corn leaf Aphids MLND incidence MLND severity 174 Maize surrounded by sorghum had a significantly higher % fill (79%), cob weight (126g), number of good seeds (213) and fresh weight (59g) as compared to the other cropping system. However maize monocrop registered significantly similar weight and numbers among the same variables (Table 1). Maize from coriander border had a significantly lower percentage fill (44%) as those intercropped with millet and elephant grass (6.3g and 6.0g) respectively registered the highest weight of bad seeds. Fresh weight was significantly lower from maize intercropped with sorghum (34g) as well as those from coriander border (33g) Table 1. Table 1 Grain yield data of maize grown under various border and intercrops cropping system in Bomet County Kenya Croping system % cob fill cob weight No. good seeds No. bad seeds Weight of good seeds weight of bad sees Fresh weight Total dry weight Olerai Maize alone 81.56c 109.12c 239.9c 13.9 67.61d 2.561ab 70.17c 63b Maize in by E. grass border 56b 86.27b 152b 10 46.61b 1.913a 48.52b 42a Maize intercropped with E. grass 53.13ab 74.55a 114.9a 28.3 29.83a 6.007c 35.84a b 31a Maize in Sorghum border 79.06c 126.35c 213.2c 17.6 56.99c 2.97abc 59.96c 53ab Maize intercropped with sorghum 54.38ab 75.51a 110.6a 29 29.21a 5.259bc 34.47a 30a Maize in coriander border 44.06a 70.04a 112.1a 16.7 30.83a 2.798ab 33.63a 28a Maize intercropped with coriander 52.19ab 75.67a 125.1a 18.9 38.99ab 3.047abc 41.91a b 37a Maize in Millet border 59.69b 74.43a 145.6a 14.2 35.51ab 2.459ab 38.7ab 34a Maize intercropped with millet 57.5b 72.64a 116.2a 30.7 32.93a 6.261c 39.17a b 34a P value <.001 <.001 <.001 0.12 <.001 0.02 <.001 <.001 se 0.12 0.12 0.16 0.13 0.05 0.14 0.6 * Means within column followed by the same letter are not significantly different (P = 0.05). 175 Infestation of vectors of MLN viruses on companion crops grown under various polycultures cropping systems in Bomet County Corn thrips infestation significantly varied among all the companion crops. Infestation of thrips on Olerai maize treatments was significantly higher (46) than the entire companion crops (Table 2). This was followed by coriander border and coriander intercrop that had a mean infestation of 16.3 and 12.9. Corn leaf aphids’ infestation was significantly higher on coriander that was planted as an intercrop and a border crop of maize (Table 2). This did not vary significantly with the corn leaf aphid infestation on maize 15.4. Table 2 vectors of MLN viruses on companion crops grown under various Cropping systems in Bomet County Treatment Corn thrips Corn Aphids maize 46±0.5d 15.4±0.6 bc Elephant grass border 2.6±0.5a 5.1±0.6a Elephant grass intercrop 2.9±0.5a 5.5±0.6a Sorghum border 2.5±0.5a 6.3±0.6a Sorghum intercrop 6.2±0.5ab 8.2±0.6ab coriander border 16.3±0.5c 22.7±0.6c coriander intercrop 12.9±0.5bc 22.2±0.6c millet border 3.9±0.5a 8.7±0.6ab millet intercrop 4.0±0.5a 7.4±0.6a p value ≥ 0.001 0.04 * Means within column followed by the same letter are not significantly different (P = 0.05). DISCUSSION Both corn thrips and corn leaf aphids preferred maize as their main host when grown together with other companion crops. Therefore the different cropping system had no effect on infestation levels of thrips and aphids on maize. This could be attributed to the fact that corn leaf aphids are polyphagous in nature and this enables them to colonise maize in all treatments in every cropping system without discrimination (Potting et al. 2007). However, it is also clearly noted that a significant number of both corn thrips and corn leaf aphids found their habitat on coriander. Results show coriander harbouring the vectors more than the other companion crops (millet, sorghum and elephant grass). This could be advantageous to maize as the viral transmission process is interfered with. Since the companion crops were planted two weeks earlier than the target maize crop, the thrips and aphids are likely to inhabit the coriander, feed on it and in the process reduce the innoculum or lose it before they invade the maize. The transmitted virus subsequently does not affect the yield as in the case of the other yield from other cropping systems. The same trend is seen on maize surrounded 176 by elephant grass border. Elephant grass is likely to posses chemical cues that not only repel the vectors but also reduce the virulence nature of the virus. Khan et al., 2004 and Cook et al., 2007 had similar results when Elephant grass successfully controlled Chilo partellus through push and pull mechanism. The infestation levels of the corn leaf aphids on coriander were significantly similar to the one on maize. It is also key to note that lack of a significance variance in the infestation levels of aphids on both coriander and the maize varieties could be exploited in the control of the maize lethal necrosis disease. In the presence of both coriander and maize, aphids have a similar probability of feeding on either crop. This may greatly reduce the virulence nature of sugarcane mosaic virus and reduce its interaction chances with the MCMV virus. Eventually, MLND levels may greatly reduce in maize planted as a polycultures even though the number of vectors remains the same. Since aphids transmits SCMV non- persistently (Gwendolyne et al., 1996), it is possible to disrupt virus transmission to maize, if these aphids land on the companion crops and feed before infesting maize. This might have been the case in explaining the absence of SCMV in maize grown with coriander. CONCLUSION Coriander may turn out to be a good trap crop for the F. williamsi and Aphids. However more tests should be carried out to monitor the population build-up of the thrips throughout the growing season and relate it with the MLND manifestation in maize crops. Sorghum intercrop has more potential in as a trap crop for F. williamsi while millet, Napier and sorghum border has more potential as repel companion crops. RECOMMENDATIONS Studies are required to determine whether coriander is a host to Corn Thrips or a dead end trap crop that only offers oviposition sites but may not be able to support the development of the pest. There is need therefore to carry out more studies on other possible trap crops and stabilising the cropping systems so that a clear definite ratio can be used. ACKNOWLEDGEMENT This study was financially supported by KAPAP CGS project on Management and Eradication of Maize Lethal Necrosis Disease (MLND) in Kenya. We thank Kenyatta University for granting the opportunity for study to the first author, KARLO for facilitating the work and the Bomet farmers for partnering with us. REFERENCES Adams, I. P., Harju, V. A., Hodges, T., Hany, U., Skelton, A., Rai, S., Deka, M. K., Smith, J., Fox, A., Uzayisenga, B., Ngaboyisong,a C., Uwumukiza, B., Rutikanga, A., Rutherford, M., Ricthis, B., Phiri, N. and Boonham, N. 177 (2014). First report of maize lethal necrosis disease in Rwanda. New Disease Reports, 29:22.http://www.ndrs.org.uk/article.php?id=029022 GIEWS/ FAO, (2013). Food security Snap shot Gwendolyne Y, Fondofe, Irwin E, Harv B. and Gail E (1996). Aphids and disease in crops.Illinois Natural History Survey. Groote, H. D., Overholt, W. A. and Ouma J. (2011). Assesing the potential economic Impact of Bacillus thuringiensis (Bt) in Kenya. African Journal of Biotechnology 10: 23 pp 4741-4751 Kinyua Z.M., M. Kasina, A.W. Wangai, B. Langat and D.W. Gichobi (2015). Development of a robust severity assessment scale for maize lethal necrosis disease. Paper presented at the KAPAP at KALRO westlands 10-12th June 2015. Khan Z R and Pickett J A. (2004). The ‘push-pull’ strategy for stemborer management: a case study in exploiting biodiversity and chemical ecology. Ecological Engineering for Pest Management-Advances in Habitat Manipulation for Arthropods (ed. by GM Gurr, SD Wratten and MA Altieri), CABI Publishing, Wallingford, UK ; pp.155-164. Makumbi, D. and Wangai, A. (2013). Maize lethal necrosis (MLN) disease in Kenya and Tanzania: Facts and actions. CIMMYT- KARI. http://www.cimmyt.org/en/where-we-work/africa/item/maize-lethal- necrosis-mln-disease-in-kenya-and-tanzania-facts-and-actionsMMm Nyasani, J.O., Meyhofer, R., Subramanian, S. and Poehling, H. M. (2012). Effect of intercrops on thrips species composition and population abundance on French beans in Kenya. Entomologia experimentalis et applicata 142 (3) 236-246). Olwande, J. (2012). Smallholder maize production efficiency in Kenya. Presentation at the FAO/University of Nairobi Regional Workshop on an Integrated Policy Approach to Commercializing Smallholder Maize Production,6–7 June 2012, at: http://www.fao.org/fileadmin/templates/esa/Workshop_reports/Small holders_2012/Presentations_1/Olwande_Smallholder_Maize_Efficiency_ Kenya. Potting R, Perry J, and Powell JNW. (2005). Insect behavioral ecology and other factors affecting the control efficacy of agroecosystem diversification strategies. Ecological Modeling 182: 199-216 Wangai, A.W., Redinbaugh, M. G., Kinyua, Z. M., Miano, D. W., Leley, P. K., Kasina, M., Mahuku, G., Scheets, K. and Jeffers, D. (2012). First report of Maize chlorotic mottle virus and maize lethal necrosis in Kenya. Plant Disease 96(10):1582-1583. 178 5.Parallel Session 4 : Soil health management, input access and conservation issues 5.1 Time preference and resource use behavior by wetland users: The case of ewaso narok wetland, Kenya Magdalene Mutumi1*, Cecilia Ritho1, Daniel Kyalo2, and Paul Guthiga3 1Department of Agricultural Economics, Faculty of Agriculture, University of Nairobi, P.O Box 29053-00625,Kangemi,Nairobi, Kenya. 2Department of Agribusiness and Trade, Kenyatta University, P.O. Box 43844- 00100 Nairobi, Kenya. 3Regional Strategic Analysis and Knowledge Support System (ReSAKSS), International Livestock Research Institute (ILRI) P.O. Box 30709-00100, Nairobi, Kenya. *Correspondence; Email: mmutumi@gmail.com, Cell: +254 727 450 327 ABSTRACT This paper assessed impacts of individual rate of time preference (IRTP) on resource use behavior of wetland users in Ewaso Narok Wetland, Kenya. The impact of IRTP on the resource use behavior was tested using the seemingly unrelated regression (SUR) model. Analysis showed that long distance from a homestead to piped water resulted in a higher IRTP. Further results showed that individuals with a higher rate of time preference used the wetland resources intensively. Secure land tenure of wetland users in the upland parcels reduced the intensity of resource use. Generally results suggest that policies should be formulated to facilitate infrastructural development like installation of alternative sources of resources like water. Title deeds should be provided for users in the upland parcels of land. Land area under crop should be minimized and number of livestock allowed in the wetland controlled. This will promote sustainable management of the wetland. Keywords: Sustainable management; wetland resources; Land tenure; Ewaso Narok INTRODUCTION Wetlands have both ecological and socio-economic values such as livelihood promotion, habitats for flora and fauna, flood regulation and water filtration, (Mitchell, 2013). There are also activities which they provide like fish farming, forestry, wildlife conservancies, pastoralism and crop farming (MEMR, 2012; Thenya et al., 2011). Although the wetlands have a lot of merits there are major challenges facing wetlands in sub-Saharan Africa which include climate change, population pressure, siltation and pollution (Mitchell, 2013; Nonga et al., 2010; Kafle et al., 2008). Kenya is not an exception, for example the Ewaso Narok wetland supports many human activities (Thenya, 2001) but it is over grazed and 179 intensively cultivated with over 80% of the wetland having human activities in it leading to biodiversity loss(Thenya, et al., 2011; Behn, et al., 2013). Biodiversity loss is brought about by the change of human behavior in the use of resources (Milner-Gulland, 2012). Human behavior can be defined as practices or observable actions performed by an individual in achieving his/her objectives (Viet gung et al., 2007). There is need for a sustainable management of the wetland in order to reduce the impact of the human activities on the scarce resources. The sustainable management of natural resources is influenced by both internal and external factors such as socio-economic, individual rate of time preference and policy factors (Sullivan, 2011; Mulligan, 2007; Godoy, et al., 2001). Time preference is the tendency of a user to consider current consumption over future consumption. Rate of time preference (discount rate) induces a user to delay consumption and it refers to an amount of money which will compensate a user to forgo the current consumption (Frederick.et al., 2002; Gunatilake et al., 2009). A high rate of individual time preference means harvesting more of the resources in the present as opposed to the future (Gunatilake et al., 2009 and Di Falco, 2013). Resource use behavior refers to the economic activities undertaken by users in and around the wetland using the available resources to meet their objectives (Viet gung et al., 2007). The wetland users utilize the scarce resources in order to maximize current benefits with little known of their concern for the future generation. Understanding the impact of IRTP on resource use and its determinants will help in the sustainable management of the wetland. This will promote environmental sustainability and optimize social welfare for both the current and future generations. METHODOLOGY Research Design This paper was based on a cross sectional survey in which primary data was collected from wetland users of Ewaso Narok Wetland using semi structured interview schedules. A list of the total population of 6000 registered wetland users was obtained from the Water Resource Users Associations (WRUA) and village elders. Stratified random sampling was used to sample the villages of interest. The respondents per household were drawn from the villages proportional to size using the simple random sampling. A sample size of 300 respondents was distributed across the three categories of users as follows: small scale farmers, commercial farmers and pastoralist were 106, 95, and 99 respectively. Estimation of individual rate of time preference This involved a set of 10 offers which had two options for the respondent to choice as shown in Table 1. The first set of option had a constant value of KES 1000 which was assumed to be received after one month (present market value) and the second option had an increment of KES 150 on top of the present value 180 for each of the 10 offers and it was assumed to be received after six months (future value offered) (Teh, et al., 2014). The respondent were to choose what they preferred from each of the offers, for example, a respondent would prefer the first option of KES 1000 up to the 4th offer and switches to a value of KES 1750 at the 5th offer. In that case the IRTP of that respondent would be calculated using the 4th value as the present offer and the 5th value as the future offer. The IRTP was determined by the point at which they switched from choosing the immediate to future payment. This estimation is on experimental approach (choice task) known as the Multiple Price List m(Teh, et al., 2014; Andreoni and Sprenger, 2012; Harrison, et al., 2002). Another approach used in this paper was the matching task in which the respondent is given the initial offer and is required to indicate the value they would prefer given the time period (Frederick et al., 2002). This was in a case where the respondent persists on the smaller sooner reward up to the 10th offer. The two values (present and future) and the difference in the two times period of five months was used on the hyperbolic model in equation 1, to obtain the individual rate of time preference for each respondent (Hardisty et al., 2013; Andersen et al., 2008; Frederick et al., 2002). 1 Pre ..............................................................................................................................1 Fut r T      Where r is the IRTP; Fut is the future offer; Pre is the present offer and T is the difference in two time periods. The calculation of the individual rate of time preference was the first step in the determination of its impact on the resource use behavior. Table 1 presents the IRTP associated with different offers. The annual interest rate (AR) and the annual effective interest rate (AER) were calculated using the hyperbolic model formula and the effective annual interest rate formula on equations 1 and 2 respectively. The effective annual rate accounts for the effect of compounding. (1 ) 1..............................................................................................2ni n AER    Where AER is the annual effective interest rate; i is the stated annual interest rate and n is the number of compounding periods. Individual rate of time preference was used in the first regression as the dependent variable. Table 2 shows the variables used in the IRTP model. Equation 3 was used to in the estimation of individual rate of time preference. 1 n Y o iXi i       ……………………………………………………………………………3 Where, Y is the individual rate of time preference estimated in the previous equation, β0 is the intercept, βi are the coefficients to be estimated, lX represents explanatory variables and ε is error term. 181 To assess the determinants of resource use behavior The IRTP was estimated and used as an explanatory variable in the second estimation of determinants of resource use behavior. Land use intensity (LUI) index was used as an indicator of the resource use behavior. The land use intensity was a composite index calculated using six variables namely; capital inputs (Ksh/ha), labour input (man days/ha), land area under crop (ha), quantity of harvest (Kg/ha), rate of organic and inorganic fertilizer application (Kg/ha). These variables were subjected to dimension reduction through the Principle Component Analysis (PCA) approach The PCA was set such that the eigenvalue of one was used meaning each observed variable contributes one unit of variance to the total variance in the data set; the factors with eigenvalue of at least one are selected. Test of appropriateness of the factor analysis was evaluated using the Kaiser-Meyer- Olkin measure of sampling adequacy (KMO). The KMO statistic greater than or equal to 0.5 is adequate and acceptable (Field, 2009). The Bartlett’s test should be significant meaning the correlation between the variables is significantly different from zero. The factors extracted explain the total variance and form LUI index which was used in the second model as the dependent variable. The Breusch-Pagan test of independence was carried out to ascertain whether the two regression equations were to be estimated jointly or individually. The estimation of the equations would be done jointly using the SUR (Seemingly Unrelated Regression) estimator model, under the assumption that the error terms of the two equations are contemporaneously correlated. If this assumption does not hold then the regression equations are estimated individually using OLS. SURE model was used to estimate the determinants of individual rate of time preference and resource use behavior jointly using equation 3 and 4 respectively. 1 n W Zi i i         ……………………………………………………………….............4 Where W represented the LUI index,  0 is a constant  i are the coefficients estimated and Zi represents the explanatory variables and ε is the error term. The variables used the land use intensity model, their description, units of measurement and expected signs are presented in Table 3. RESULTS AND DISCUSSION Table 4 presents descriptive statistics of variables used in the IRTP and LUI index models. The average IRTP of users in Ewaso Narok Wetland of 20.05%.which is higher than the average market interest rate in Kenya for the year 2015 was 11.50% (CBK, 2015). 182 Determinants of IRTP Table 5 presents the factors which influenced the individual rate of time preference of wetland users in Ewaso Narok wetland. A one hectare increase in land area under crops in the wetland increases the IRTP by 33 percent. This means that a household which has a larger area of land on crops in the wetland utilizes more resources than a household with small land area occupancy. This contradicted the findings of (Yesuf and Bluffstone, 2008) which concluded that the land area had a negative effect to the IRTP. Being a commercial farmer increased the individual rate of time preference by 68% above the small scale farmers. The large commercial farms were mostly located at the upland area and pumped water for irrigation from the wetland. Most of the small scale farms were located inside the wetland and used mostly bucket irrigation which is more conservative. This means that commercial farmers are the largest contributors of degradation in the Ewaso Narok Wetland. Being a member of a group increases the IRTP by 61%. The group membership was associated with access to loans which were used for agricultural inputs and other farm activities (Mnimbo, 2013). A household member being in a group accelerates the degradation of the wetland through increased input use. One kilometer increase in the distance of a household to piped water increases the IRTP by 3 percent ceteris paribus. The piped water was used as an indicator of alternative source of water. This means that the use of resources like water would reduce if alternative sources of resources are provided and this would promote sustainability of the wetland. Determinants of land use intensity (LUI) index PCA results showed a KMO of 0.524 and the Bartlett’s test was significant at p<0.000 thus supporting the factorability of the correlation matrix. The factors extracted were explaining 72% of the variance. Table 6 presents the factors influencing land use intensity. IRTP had the largest magnitude in influencing the LUI; this was followed by socio-economic factors (group membership) and the policy factors (tenure security). One unit increase in the individual rate of time preference increases the land use intensity by 1.65. Individuals with a high rate of time preference use more of the wetland resources. This is consistent with findings by Gunatilake et al., (2009) and Di Falco, (2013) that high individual rate of time preference means harvesting more of the resources in the present and less conservation. This means that the IRTP should be a major consideration in the decisions of sustainable management of the Ewaso Narok Wetland. A unit increase in TLU increases land use intensity. This implies that, households which have large herd of livestock use more pasture and water resources from the wetland than those with fewer animals. This finding is consistent with the findings of Sourya et al., (2015) which found a positive effect of the number of 183 livestock owned and wetland use. This means the large numbers of livestock which graze within the Ewaso Narok Wetland contribute to its degradation. An increase in the family size of a wetland user increases the intensity of land use. Larger households have available labour that can be allocated to a wide range of activities including utilizing wetland resources. This was similar to the findings of Sourya et al., (2015) which also found a positive effect between household size and the use of the wetland resources. A one kilometer increase in distance from a homestead to piped water increases the land use intensity. This means provision of alternative sources of resources like piped water would promote sustainable management of the Ewaso Narok Wetland. A household member being in a group reduced the land use intensity by 1.24. This could be due to access of loans which facilitated diversification to other activities outside the wetland use. This was consistent with the findings of Mnimbo, (2013) which concluded that group membership improved the level of income through provision of loans and encouraged diversification into businesses. Group membership enhances diversification into activities which are not dependent on the wetland resources and therefore promote sustainability. A year increase in the level of education reduced the land use intensity. A high level of education of an individual could be associated with higher income which leads to diversification to alternative resources and hence the intensity of using the wetland resources decrease. This was consistent with the findings of Felix, (2012) which found a positive effect between education and resource use in the wetland. Having secure land tenure decreases the intensity of land use by 0.94. The wetland users with secure land tenure were those located in the upland area and had title deeds. This means that those users with secure land tenure used the resources conservatively as opposed to users who had insecure land tenure. This means that the securing land tenure in the upland area encourage conservation of resources in the wetland. 184 TABLES Table 1: Individual rate of time preference using choice and matching task methods Offer after 1 Month Offer after 6 Months AR AER Respondents on switch point (Constant value of KES 1000) 1150 0.36 0.39 32 1300 0.72 0.86 15 1450 1.08 1.4 9 1600 1.44 2.02 14 1750 1.8 2.71 13 1900 2.16 3.48 13 2050 2.52 4.33 42 2200 2.88 5.25 29 2350 3.24 6.26 24 2500 3.6 7.35 4 3000 4.8 11.5 8 4000 7.2 22.9 1 5000 9.6 38.1 12 6000 12.01 57.5 7 10000 21.6 178.7 37 12000 26.4 266.7 2 15000 33.6 434.7 3 17000 38.4 571.3 1 20000 45.6 814.4 8 50000 117.7 5970.9 12 60000 141.8 8869.1 3 100000 237.9 26857.1 9 Source: Author’s Survey, (2015). 185 Table 2: Variables used in individual rate of time preference model Variables Description Measurement Expected sign Dependent variable IRTP Individual rate of time preference Percentage Independent variables Group membership Wetland user being a member of a group Dummy(Yes=1) + Commercial farmer Wetland user being a commercial farmer Dummy(Yes=1) + Pastoralist Wetland user being a pastoralist Dummy(Yes=1) - Land area on crop Land area on crop in the wetland Hectares +/- Distance piped water Distance from a homestead to piped water Kilometers + Proportion sold Proportion of crop harvest sold Percentage + Total income Total income of the household KES - Distance to wetland Distance from a homestead to the wetland Kilometers - Conflicts of users Conflicts of wetland user on the use of resource Dummy(Yes=1) + Table 3: Variables used in land use intensity model of Ewaso Narok Wetland user Variables Description Measurement Expected sign Dependent variable LUI Land use intensity Index Independent variables Estimated IRTP Estimated individual rate of time preference Percentage + Land tenure land ownership type Dummy(secure=1) - Market access Distance from a homestead to the nearest shopping center Kilometers + Household size Household size Adult Equivalent + Education Level of education of the household head Years - Group membership Household member being in a group Dummy(Yes=1) -/+ Wildlife danger Danger of wildlife in wetland Dummy(Yes=1) - Distance piped water Distance from a homestead to piped water Kilometers + Total income Total household income KES - TLU Tropical livestock unit Number +/- Years wetland use Number of years of using the wetland Years + 186 Table 4: Descriptive statistics of the variables Variable Average Standard Minimum Maximum deviation LUI index 0 1.53 -4.97 7.87 Income (KES) 25162.42 37579.21 0 309934 Market access (Km) 7.25 7.91 0 36 Education (Years) 5.13 4.75 0 20 Landsize under crop in wetland (Ha) 0.82 0.84 0.04 7.48 Distance piped water (Km) 10.91 12.52 0 77 TLU (Number) 18.72 46.19 0 580 Years of using wetland 14.91 8.22 1 35 Household size (Adult equivalent) 4.43 1.89 0.75 10.05 IRTP (%) 20.02 47.29 0.36 239.98 Distance from homestead to piped water (Km) 10.91 12.52 0 77 Distance from homestead to wetland (Km) 1.95 1.88 0 10 Proportion of harvest sold (%) 0.44 0.23 0.04 1.11 Source: Author’s Survey, (2015) Table 5: Factors influencing individual rate of time preference of Ewaso Narok Wetland users Independent variables Coefficient Std error P value Distance of household to piped water (Km) 0.03*** 0.01 0.00 Distance of household to the wetland (Km) 0.14 0.10 0.13 Land area on crop in the wetland (Ha) 0.33** 0.15 0.03 Proportion of harvest sold (%) -0.66 0.62 0.28 Total household Income per annum (KES) -0.06 0.06 0.29 Group membership 0.61** 0.29 0.03 Commercial farmers 0.67** 0.32 0.03 Pastoralists 0.41 0.90 0.65 Conflicts among users 0.40 0.31 0.19 Intercept -0.05 0.49 0.91 Source: Author’s Survey, (2015) **, *** means significance at 5%, 1% respectively 187 Table 6: Determinants of land use intensity for Ewaso Narok Wetland users Independent variables Coefficient Std error P value Estimated IRTP 1.65*** 0.46 0.00 Household size (Adult equivalent) 0.60* 0.35 0.09 Education (Years) -0.11** 0.05 0.03 Total household income per annum(KES) -0.01 0.08 0.19 TLU (Number) 0.01*** 0.04 0.00 Group membership -1.24*** 0.16 0.00 Danger of wildlife 0.66 0.44 0.19 Tenure security -0.94** 0.39 0.02 Distance of household to piped water(Km) 0.05** 0.34 0.02 Years of using the wetland 0.02 0.02 0.35 Distance to the shopping centre(Km) -0.03 0.02 0.20 Intercept -1.69 1.51 0.27 Source: Author’s Survey, (2015) *, **, *** means significance at 10%, 5% and 1% respectively. CONCLUSION Sustainable management of the Ewaso Narok Wetland would entail a change of human behavior in the use of resources. The human behavior is influenced by both external (policy) and internal (IRTP and socio-economic) factors. The average IRTP of users in Ewaso Narok Wetland was higher (20.05%) than the mean market interest rate in the year 2015. The IRTP had the highest magnitude and a positive effect in influencing resource use behavior. This implied that the wetland users used more of the wetland resources in the present. Therefore, the sustainability of the wetland is at stake. In order to sustainably manage the wetland alternative sources of resources like piped water should be provided near the wetland area in order to minimize the over dependence on resources. Land area under crop should be minimized per household and the intensity of grazing livestock controlled. Land tenure of the wetland users in the upland parcels should be secured to enhance sustainable use of the wetland resources. Policy implications Following the above conclusion therefore, this study draws the following policy implications; there should be provision of title deeds in the upland fields. The land tenure of the wetland users in the upland parcels of land should be secured; this will encourage sustainable use of resources in the wetland. The government should enact policies that encourage infrastructural development such as installation of piped water in the upland area. This will reduce the rate of extraction of water from the wetland. Land area under crop within the wetland area should be minimized and the intensity of grazing livestock in the wetland should be controlled to facilitate sustainable management. 188 REFERENCES Andersen, S., Harrison, G. W., Lau, M. I., Rutstrom, E.E., (2008). Eliciting risk and time preferences. Journal of Econometrica. 76(3): 583-618. Andersen, S., Harrison, G. W., Lau, M. I. and Rutstrom, E. E., (2006). Elicitation using multiple price list formats. Journal of Experimental Economics. 9 (2006):383-405. Andreoni, J. and Sprenger, C., (2012). Estimating time preferences from convex budgets. Journal of American Economic Review.102 (7):3333-3356. Andy, F., (2009). Discovering statistics using SPSS.Sage publication Inc.Thousand Oaks,California 91320 .Third edition. Behn,K.,Alvarez,M.,Amler,E.,Beuel,S.,Gabiri,G.,Koko,I.,Macharia,J.,Umuhumuze ,G.,Wagner,K.,Willy,D.K.and Ziegler,S., (2013). Assessment of florists and structural variables of vegetation: the impact of agricultural use in East Africa wetland ecosystem. [Online http://www.botanik.unigreifswald.de/fileadmin/workshop/201 4/KaiBehnPosterKoblenz.pdf accessed on 12 September 2015]. CBK, (2015) Interest rates in Kenya [Online http://www.tradingeconomics.com/kenya/interest-rate accessed on 13 April 2016]. Di Falco, S., (2013). Explaining Investment in soil conservation; Weather Risk and Rate of time preference the Highlands of Ethiopia. In the 20th annual conference of European association for Environmental and Resource Economists,28/6/2013.Toulouse,France. FAO, (2014). Food security for sustainable development and urbanization. A paper presented to the United Nations Economic and Social Council (ECOSOC) Integration Segment.6pp. FAO, (2006). Agriculture, Trade negotiations and Gender. A contribution of the FAO Gender and Population Division to the United Nation interagency publication on Gender and Trade.49pp.www.fao.org/gender Felix, M. K., (2012). Socio-Economic factors influencing household wetland resources conservation and use in Ngaciuma sub-catchment, upper Tana, Kenya. Unpublished master’s thesis in integrated watershed management submitted to the school of pure and applied science in the Kenyatta University, Kenya. Frederick, S., Loewenstein, G., O’Donoghue, T., (2002), Time discounting and time preference: A critical review. Journal of Economic Literature .40(2):351-401. Georgiadis, N.J., (2010) Introduction: Conserving Wildlife in Kenya’s Ewaso landscape.Smithsinian contributions to Zoology.632:10pp. Gujarati, D.N. (2008). Basic Econometrics McGraw-Hill Education; ISBN-13: 978- 0073375779 fifth edition /lrwin, New York. 944pp. Gunatilake, H. M., Wickramasinghe, W. A. R., Abeygunawardena, P., (2009). Time preference and National resources use by Local Communities .The 189 case of Sin harasa forest in Sri Lanka. Sri Lanka journal of agricultural Economics .11:31-60. Hardisty, D. J., Thompson, K. F., Krantz, D. H and Weber, E.U., (2013). How to measure time preferences. An experimental comparison of three methods. Journal of judgment of decision making. 8(3): 236-249. Harrison, G.W., Lau, M.I and Williams, M.B., (2002). Estimating Individual Discount Rates in Denmark: A field Experiment. Journal of American Economic Review.92 (5):1606-1617. Herzog, F., Steiner, B., Bailey, D., Baudry, J., Billeter, R., Buk’acek, R., De Blust,G., De Cock, R.,Dirksen, J.Dormann, C.F,De Filippi, R.,Frossard, E., Liira, J., Schmidt, T., Stockli, R., Thenail, C. Van Winger den, W. and R. Bugter. (2006). Assessing the intensity of temperate European Agriculture at the landscape scale.European Journal of Agronomy.24 (2006):165-181. Kafle, G., Cotton, M., Adhikari, H., (2008). Status of and threats to water birds in Rupa Lake, Pokhara, Nepal. Journal of wetland Ecology. 2:9-12 Kothari, C.R., (2004). Research Methodology, Methods and techniques, Second revised edition. New age international publishers.401pp. Milner-Gulland, E.J., (2012). Interactions between human behavior and ecological systems. Journal of philosophical transactions of the Royal society B.367:270- 278. Mitchell, S.A., (2013). The status of wetlands, threats and the predicted effects of global climate change: the situation in sub Saharan Africa. Journal of aquatic science .75:95-112. Mulligan, R.F., (2007). Property rights and time preference. Quarterly journal of Austrian Economics. 10 (1): 23-49. Ministry of Environment and Mineral Resources (MEMR), (2012). Kenya Wetland Atlas. Government of Kenya, Nairobi. Progress Press Co. Ltd Malta. 140pp. Nonga, H. E., Mdegela, R. H., Lie, E., Sandvik, M., Skaare, J.U., (2010). Socio- economic values of wetland resources around Lake Manyara, Tanzania. Assessment of environmental threats and local community awareness on environmental degradation and their effects. Journal of wetlands ecology .4:83- 101. Seleti,Y.N and G. Tlhompho,(2014).Rural women subsistence farmers,Indigenous Knowledge systems and Agricultural research in South Africa.Journal of human ecology,48(1):33-41. Sourya, D., Behera, B. and A. Mishra., (2015). Determinants of household use of wetland resources in West Bengal, India. Journal of Wetlands Ecology and Management. 10.1007/s11273-015-9420-8. Sullivan, K. A., (2011). The Effect of Risk, Time preference and Poverty on the impacts of Forest Tenure Reforms in China. Open Access Dissertations. Paper 105. 190 Taruvinga, A. and Mushunje, A., (2010) Socio-Economic factors that influence households’ participation in wetland cultivation: binary Logistic regression of Wetland cultivators and non cultivators. A paper presented at the joint 3rd African Associaton of Agricultural Economics(AAAE) and 48th Agricultural Economists Association of South Africa(AEASA) Conference, Cape Town, South Africa, on September 19-23,2010.20pp. Teh, L. S. L., Teh, L. C. L.and Rashidsumaila, U., (2014). Time preference of small- scale fishers in open access and traditionally managed reef fisheries. Journal of Marine policy 44(2014):222-231. Thenya, T., (1998). Some Ecological aspects, Bio-Economic values and human impact on a tropical papyrus wetland, Ewaso Narok swamp, Laikipia District, Kenya. Unpublished Masters Dissertation at the Department of Geography, University of Nairobi, Kenya Thenya, T., (2001). Challenges of conservation of dry land shallow waters, Ewaso Narok swamp, Laikipia district, Kenya. Journal of hydrobiologia 458(1- 3):107- 119. Thenya,T.,Kitema,B.P.,Ouko,C.A.,Kahiu,N.,Njuguna,E.c.,Karanja,F.,Ojwang’D., Wambugu,G., (2011). Assessment of Ecological status and socio-economic dynamic of upper Ewaso Ngiro Basin wetlands-Summary of findings. CETRAD, Kenya. 40pp. Varian, H. R., (2010). Intermediate Microeconomics.A modern Approach. Eighth Edition. W.W. Norton &Company, New York. 740pp. Viet Dung, N., Le Nguyen, T., Xuan Thuy, H., Danh Tinh, N., (2007). Community behaviors towards nature conservation: A theoretical analysis for practical approaches. Journal of PanNature-People and Nature Conciliation 15pp. 191 5.2 The effect of invasive plant (Lantana camara) on soil chemistry at conservation area ol-donyo sabuk national park, Kenya *Marystella N. Wekhanya1, Prof. Paul Mbugua2, Prof. John Kiogora Mworia3. 1School of Pure and Applied Sciences, Kenyatta University P.O.BOX 43844- 00100 Nairobi-Kenya. *Corresponding author: nangoniswek@gmail.com ABSTRACT Invasive species are a major ecological and management concern in natural ecosystems and pose a threat to many of Kenya’s protected areas. Invasive plant species compete and hybridize with native species often to the disadvantage of the native species. Invasion of native plant habitats by invasive plants can drastically change soil chemical properties such as pH, mineral composition and mineral levels. The aim of this study was to establish whether the invasive plant Lantana camara L. alters the soil nutrients. Five study sites were selected by purposeful sampling. Soil samples were collected randomly from L. camara invaded areas and similarly from adjacent areas free from L. camara. The soil samples were analysed for the following: pH, potassium (K), calcium (Ca) magnesium (Mg), total nitrogen (N), phosphorous (P). All statistical analyses were done using Statistical Analytical Software (S.A.S) portable version 9.4 software to determine differences in nutrients composition in Lantana invaded and non-invaded areas. Results of the study indicated soils from L camara invaded areas had an increase in pH value, phosphorous, nitrogen compared to the non-invaded areas. Most nutrient levels were found to be higher in the invaded than the invaded due to high pH. The slight pH in this case also made phosphorous to be more available to the L. camara invaded areas. This study therefore reveals that Lantana camara has effect on the soil nutrient levels leading to changes in soil chemistry of invaded areas. This is in a bid to suit its survival to the detriment of the native plants. Key Words: Invasive species, native species, soil nutrients, habitats, Lantana camara. INTRODUCTION Invasive plant species are among the world's greatest threats to biodiversity and native species in protected areas (Lwando, 2009). Invasive species may change the community structure through competition which is categorised into two; exploitation competition whereby there is indirect interactions in the use of resource and interference competition (Callaway and Ridenour, 2004). Once invasive plants become established, they change the soil nutrients and shift nutrient cycling in an ecosystem. Soil is important to plants because it is a source of nutrients that are utilised by plants for various physiological processes. Other 192 than nutrients soil also has microorganisms, some have a relationship with plants. Both native and invasive plants form some relationships with bacteria and fungi in the soil that facilitate the extraction and conversion of elements to biologically usable forms (Batten et al., 2006). Some convert atmospheric nitrogen into usable forms that are then exchanged for carbon from the plant. Nitrogen in form of nitrates and nitrites are frequently limiting in soils, yet many invaded ecosystems have shown to have more carbon and nitrogen in plant tissues and soils compared with systems dominated by native plants (Mack et al., 2000). Invasive plant species pose a serious threat to biodiversity and are now considered second to habitat destruction in driving global biodiversity loss (Batten et al., 2006). Invasive plant species have many impacts on plant communities through their direct and indirect effects on soil chemistry (Ehrenfeld, 2003) as plant community composition changes, preferred wildlife habitats are destroyed hence destabilizing such protected areas. Lantana camara L. is one of the most invasive plants and has been ranked as the highest impacting invasive species (Batianoff and Butler, 2003). It is top among the world’s 100 worst invasive alien species (GISP, 2003), because it possess great potential to escape cultivation and have deleterious effect on species richness (Islam, 2001). Lantana camara L. is a major weed in over 60 countries and is one of the 10 worst weeds worldwide (Sharma et al., 2005). In Australia, Lantana is a weed which has evoked a national outcry, having invaded at least 4 million hectares of predominantly coastal and sub-coastal ranges of eastern Australia (Parsons and Cuthbert, 2001). Lantana camara has invaded disturbed forest land and neglected pasture in much of its naturalized range. Lantana also now seen invading native vegetation in woodlands and savannas notably in protected areas (Englberger, 2009). In some areas, competition by the shrub results in a reduction of biodiversity (Kumar and Rohatgi, 1999). Despite the establishment of a number of natural enemies of Lantana in exotic populations, control of its populations has been usually limited or failed (Day et al., 2003). In thick stands, the shrub increases costs in forest management by inhibiting access in stands for thinning and felling, competes through rapid reproduction and increases fire hazards (Graaff,1986). The impact of invasive plant species in protected areas, mostly National Parks and reserves in Kenya is currently poorly understood and the magnitude of the problem is not well appreciated (Simba et al., 2013). Studies show that Kenya has been invaded by 34 different species; eleven arthropods, ten micro-organisms, nine plant species and four vertebrates (Farrell et al., 1995; Lyons, 1999). Some of these invasions have yielded significant consequences on socio-economic status. Notable examples of invasive plants include the water hyacinth (Eichhornia crassipes Mart) (Hill et al., 1999) in Lake victoria, the water fern (Salvinia molesta D.S Mitch.) in Lake Naivasha, Mathenge (Prosopis juliflora S.W) in the Rift Valley (Baringo); Tickberry (Lantana camara L) in Nairobi National Park; Jimsonweed (Datura stramoniun L) grows more in disturbed habitats example 193 construction sites; Long spine cactus (Opuntia exaltata L); Sweet prickly pear (Opuntia ficus indica L); Drooping prickly pear (Opuntia vulgaris Miller) Opuntia species mostly in drier areas as hedges and boundaries (Naivasha); Wild garlic (Allium vineale L) Mt Elgon; Mexican marigold (Tagetes minuta L.) Morning glory (Ipomoea indica (L) Roth) in Kiambu County and Eucalyptus (Eucalyptus globules Labill) Kenyan riparian zones (Kedera, 2005). In the present study soil samples were analysed to determine soil nutrient composition in the areas invaded by Lantana camara and those that are not invaded. It is hypothesized that Lantana camara has no effect on soil nutrient composition. MATERIALS AND METHOD The study was carried out in Ol-Donyo Sabuk National Park. The Park is located about 85km North-East of Nairobi in Machakos County. It lies between latitudes 1005’ and 10010’S, longitudes 37010’ and 37020’E and altitude 2145 metres above sea level at the peak of the hill. The regime is bimodal with rains occurring mostly from March to May and October to December with peaks in April and November. The maximum annual mean temperature is between 220C-260C and minimum annual mean is 100C-140C. January-March is hot and dry, April-June is hot and wet, July-October is very warm and dry, November - December is warm and wet (KWS, 1999). The soils are varied due to geological formation; shallow and stony with rocky outcrops which have been subjected to geological and recent accelerated erosion losing their original characteristics (KWS, 1999). Communities around the Park utilise the Park resources for basic needs like firewood, grass, meat from wildlife, medicinal plants, honey and pollinator services; in some cases illegal charcoal burning and livestock grazing. Encroachment and edge effects resulting from increased population and associated disturbances are some of the activities are some of the factors that have contributed to the introduction of invasive plant species into the National Park. The most common invasive plant species found at Ol-Donyo National Park are: Lantana camara, Datura stramonium L, Solanum incanum L, Dovyalis caffra Warb, Ricinus communis L, Tagetes minuta L, Caesalpinia decapelata (Roth) Alston (Faden, 1974). The soil samples were taken in the months of November to December for the wet season sampling while the dry season sampling was carried out between January and March. By purposeful sampling, 5 blocks were selected; this was done to achieve a clear representation of the sample size of the whole Park. In each block, 2 quadrants (10mx10m) at interval of more than 100m were established randomly using tape measure and wooden pegs. One quadrant invaded by L. camara and the other with native plant species (non-invaded) and soil samples were then collected using a soil auger. The soil samples from the sampling points were composited to one sample and desired amount of the collected soil samples was scooped and packed in airtight polythene bags. They were then clearly labelled and taken to the laboratory for analysis. 194 Soil pH value was determined using a pH meter – Geo-technical Engineering Bureau method (State of New York Department, 2007). The soil samples were separated on the ¼ inch (6.3 mm) sieve. 30 g of soil was weighed and placed into the glass beaker, 30g of distilled water was added to the soil sample and stirred to obtain soil slurry and then covered with watch glass. The samples were left to stand for a minimum of one hour. This is to allow the pH of the soil slurry to stabilize. The temperature of the soil samples was measured and the temperature controller of the pH meter was adjusted to that of the soil sample temperature. The pH meter was standardized by means of the standard solutions. The soil samples were stirred well with a glass rod then the electrodes were placed into the soil slurry solution and the beaker turned gently to make good contact between the solution and the electrodes. The pH value was read and recorded. Phosphorous, potassium, calcium and magnesium nutrient elements (P, K, Ca, Mg) were analysed using the Mehlich Double Acid Method (Mehlich, 1953). The soil samples were oven dried to remove moisture. For each nutrient analysed, a soil sample was extracted in the ratio of 1:5 (weight/volume) and introduced in a mixture of 0.1M dilute hydrochloric acid and 0.025M dilute sulphuric acid. Ca, and K elements were determined using a flame photometric method (Toth and Prince, 1949) while P and Mg were determined using a Calorimeter. Data Analyses The soil samples were analysed in three replicates and averages obtained for the following nutrients and parameters: soil pH, potassium (K), calcium (Ca) magnesium (Mg), total nitrogen (N), phosphorous (P). All statistical analyses were done using Statistical Analytical Software (S.A.S) portable version 9.4 software. Differences among the treatment means were assessed by ANOVA. Significant differences among the means were found through Tukey’s Honestly Significant Difference Test at 95% means was used to separate the means .The data was presented in descriptive tables and graphs. RESULTS There was significant difference (P-value <0.0001) between the Lantana invaded and non-invaded areas (Figure i) in pH. The Lantana invaded areas had high pH value (6.68) compared to the non-invaded that had pH value (6.30). Total nitrogen concentration varied significantly (P-value 0.0271) between the Lantana invaded and non-invaded areas. The Lantana invaded areas had higher mean (0.26 %) in total nitrogen concentration compared to the non-invaded (0.18%). Phosphorous concentration varied significantly (P-value <0.0001) in the Lantana invaded and non-invaded areas. The Lantana invaded and non-invaded areas had the same mean (20.93me%) in phosphorous concentration. There was significant difference (P-value = 0.0037) between the Lantana invaded and non-invaded areas. The areas not invaded by L. camara had higher mean 195 (1.89me%) in potassium concentration compared to the invaded areas that had lower mean (1.79me%). The concentration of calcium did not vary significantly (P-value = 0.2054) between the L. camara invaded and non- invaded areas. There was significant difference (P-value <0.0001) between the Lantana invaded and non-invaded areas in magnesium concentration. The non-invaded areas had higher mean of (3.04me%) in magnesium concentration compared to the non-invaded areas that was lower mean (2.71me%). Comparison of nutrients levels in invaded and non-invaded areas Figure i: In summary comparison of nutrients between the Lantana invaded and non- invaded areas showed that Lantana invaded areas had higher levels of pH, total nitrogen, phosphorous, magnesium and potassium (Figure i). DISCUSSION AND CONCLUSION From the results obtained in this study invasive plant species, Lantana camara changes physical and chemical attributes of soils including pH and soil nutrients. The soil pH and most of nutrients varied significantly (P-value <0.05) among the study sites except for Calcium (P-value >0.05). Soil pH value was higher in the soils invaded by L. camara during this study. This is consistent with the results reported by Osunkonya et al. (2010), following L. camara invasion in South Eastern Queensland, Australia, Fan et al. (2010) due to Lantana invasion in China and Sharma and Raghubanshi (2009) following Lantana invasion in India. There was significant difference (P-value <0.05) in total nitrogen and the concentration was higher in the Lantana invaded areas than in non-invaded areas. This is in agreement with (Rout and Callaway, 2009) who have reported that many plant invasions, despite dramatic decreases in local diversity, appear to increase local soil nitrogen pools and total ecosystem nitrogen stocks. High nitrogen availability favours invasive plants, and low nitrogen availability favours native plants (Laungani and Knops, 2009). 196 These results are also in agreement with Simba et al. (2013) findings on L. camara at Nairobi National Park who reported that Lantana leaf litter on decomposition release cations into the soil within its root rhizosphere and subsequently increased the concentrations of magnesium, calcium and potassium. The accumulated organic matter also mulches the soil surface under the invasive plant species hence inhibiting leaching of nutrients from the soil surface (Simba et al., 2013). According to the findings of this study there was high concentrations of potassium were found in Lantana invaded areas, this concur with Basumatary and Bordolo (1992) who found out that a layer of organic matter increases the retention of potassium in the soil. In the current study high soil pH was found in the Lantana invaded areas and according to Simba et al. (2013) high soil pH accelerates litter decomposition. The results are also in agreement with Osunkonya et al. (2010) study on L. camara who reported that soil within Lantana patches had greater air dried water content, higher organic and total organic carbon, higher exchangeable calcium, and higher pH than soils from adjacent vegetation lacking the weed. More research has to be done to investigate if L. camara leaf litter causes changes in the composition of the soil in the Park. ACKNOWLEDGEMENT Special thanks go to Kenya Wildlife Service, Biodiversity Research and Monitoring Department for granting permission for the research project to be done at Ol-Donyo Sabuk National Park. Sincere gratitude to Prof. Paul Kamau Mbugua and Prof. John Kiogora Mworia for their valuable time, comments and guidance I greatly appreciate their technical input and intellectual support, members of the department Plant Sciences Kenyatta University for their support. REFERENCES Basumatary, A. and Bordolo, I. P. (1992). Forms of Potassium in Some Soils of Assam in Relation to Soil Properties. Journal Indian Society of Soil Science , 40(3): 443–446. Batianoff, G. and Butler, D. (2003). Impact Assessment and Analysis of Sixty-Six Priority Invasive weeds in South-East Queensland. Plant Protection Quartery , 18: 11-17. Batten, K. M., Scow, K. M., Davies, K. F. and Harrison, S. P. (2006). Two Invasive Plants Alter Soil Microbial Community Composition in Serpentine Grasslands. Biological Invasions , 8: 217–230. Callaway, R. and Ridenou, R. (2004). Novel Weapons; Invasive Success and the Evolution of Increased Competitive Ability. Frontiers in Ecology and the Enviroment , 2: 436-443. Day, M., Wiley, C., J., P. and Zalucki, M. (2003). Lantana: Current Management, Status and Future Prospects. . Australian Centre for International Agricultural Research,5: 1-20. 197 Ehrenfeld, J. (2003). Effects of Exotic Plant Invasions on Soil Nutrient Cycling Processes. Ecosystems , 6: 503-523. Englberger, K. (2009). A Guide for Identification and Public Awareness. Invasive weeds of Pohnpe: Conservation Society of Pohnpei , 29. Faden, R. B. (1974). Commelinaceae. Upland Kenya Wild Flowers. (A. D. Agnew, Ed.) New York: Oxford University Press. Fan, L., Chen, Y., Yuan, J., Yang, Z. (2010). The Effect of Lantana Invasion on the Soil Chemical, Microbial Properties and Plant Biomass accumulationin Southern China. Geoderma , 154: 370-378. GISP. (2003, May 21). The Invasive Alien Species Problem. Retrieved October 10, 2015, from The Global Invasive Species Programe: http//www.org/about/AIS.asp Graaff, J. L. (1986). Lantana camara, the Plant and Some Methods for its Control. South African Forestry Journal , 136: 26-30. Islam, K. A. (2001). Deforestation Effects on Vegetration and Soil Quality in Tropical Semi-evergreen Degraded and Protected Forests of Bangladesh. Land Degradation and Development , 12: 45-56. Kedera, C. K. (2005). Invasive Alien Species in Kenya: Status and Management. Retrieved October 6, 2015, from Food and Agricultural Organization of the United Nations: Http://www.fao.org/docrep/008/y5968e10.htm Kumar, S. and Rohatgi, N. (1999). The Role of Invasive Weeds in Changing Floristic Diversity. Annals of Forestry , 7(1): 147-150. KWS. (1999). Fourteen falls/Oldonyo Sabuk Integrated Management Plan. Nairobi: Kenya Wildlife Service. Laungani, R. and Knops, J. (2009). Species-Driven Changes in Nitrogen Cycling Can Provide a Mechanism for Plant Invasions. Proceedings of the Natural Academy of Sciences of the United States of America, (pp. 106: 12400-12405). Lwando, C. (2009). Effect of Lantana camara on Plant Diversity in Zambia. WWF- Education for Nature Program , 1-4. Mack, R., Simberloff, D., Lonsdale, W., Evans, H., Clout, M. and Bazzaz, F. (2000). Biotic Invasions: Causes, Epidemiology, Global Consequences, and Control. Ecological Applications, 10: 689-710. Mehlich, A. (1953). Determination of P, Ca, Mg, K, Na and NH4. North Carolina Soil Test Division , 2: 31-36. Osunkoya, O. and Perret, C. (2010). Lantana camara L. (Verbenaceae) Invasion Effects on Soil Physicochemical Properties. Biological Fertility Soils, 47: 349- 355. Parsons, W. and Cuthbertson, E. (2001). Noxious weeds of Australia. Collingwood: CSIRO Sharma, G., and Raghubanshi, A. (2009). Lantana Invasion Alters Soil Nitrogen Pool and Processes in the Tropical Dry Deciduous Forest of India. Applied Soil Ecology , 42: 134-140. 198 Sharma, G., Raghubanshi, A. and Singh, J. (2005). Lantana invasion: An Overview. Weed Biology Management , 5: 157-167. Simba, Y. R., Kamweya, A. M., Mwangi, P. N. and Ochora, J. M. (2013). Impact of the Invasive shrub, Lantana camara L. on Soil Properties in Nairobi National Park, Kenya. International Journal of Biodiversity and Conservation , 5 (12), 803-809. State of New York Department, T. D. (2007, April 4). Geotechnical Engineering Bureau. Retrieved August 13th, 2015, from GTM-24: http//www.dot.ny.gov/divisions/engineering/technical- services/technical-services-respository/GTM-24b.pdf Toth, S. and Prince, A. L. (1949). Estimation of Carbon Exchange Capacity and Exchangeable Ca, K and Na Contents of Soils by Flame Photometric Techniques. Soil Science , 67: 439 – 445. 5.3 Determinants of use of integrated soil fertility management strategies and phosphorus management for maize production intensification in Runyenjes sub-County, Kenya. Ada, M.A.,a,b, Mucheru-Muna, M.W., a , Mugwe, J.M.,a, Zingore, S.,b, Mugi-Ngenga, E., b, Mairura, F.a, Ayieko, D. a a Kenyatta University, P.O. Box 43844-00100, Nairobi, Kenya bInternational Plant Nutrition Institute, P.O. Box 30772-00100, Nairobi, Kenya ABSTRACT Poor soils have negatively impacted food production. Consequently, researchers have designed strategies that can reverse the poor soils trend and increase food production. However, the adoption breakthroughs of such solutions is wanting. Therefore, the objectives of this study were to (i) determine how household socio-economic and demographic factors influence use of integrated soil fertility management strategies for maize production, (ii) determine how management practices of farmers affect available phosphorus levels in the soil. The study considered three ISFM strategies; use of inorganic fertilizer, use of organic manure and combined use of inorganic and organic fertilizer with improved seeds. Data collected included demographic and socio-economic characteristics of small-holder farmers, type and amount of fertilizers used on maize, soil samples were also collected. Descriptive statistics, ordinary least squares regression and binary logistic regression models in SPSS version 20 were employed during analysis. Results showed household size (positive), size of land under maize (negative), off-farm income (positive) and maize yield (positive) as significant (p<0.05) predictors of the amounts of phosphorus fertilizer used for maize production. Household head occupation (negative), size of land under 199 maize (positive), off-farm income (positive), maize yield (positive) and training (positive) were the significant (p<0.05) factors influencing use of animal manure for maize production. Size of land under maize (negative), household labor (positive), off-farm income (negative) and tropical livestock unit (positive) were significant (p<0.05) predictors of combined use of inorganic and organic fertilizer with improved seeds. Soil analysis results showed that the study area had moderately acidic soils (mean=5.60) with moderate levels of available P (mean=39.75ppm). Amount of NPK (positive), intercropping of maize with a legume (positive) and liming of soil (positive) were found to be possible predictors of available P levels on maize farms at (p<0.05). Thus, these factors must be considered during the coining of yield increasing technologies. Key words: Soil fertility, inorganic fertilizer, animal manure, improved seeds, phosphorus management INTRODUCTION Soil capital is a major resource that smallholder farming households rely on for food and income resources (Marenya and Barrett, 2007). In Sub-Saharan Africa (SSA), smallholder farmers have been experiencing dwindling agricultural produce that is attributable to soil fertility depletion, which is resulting to food insecurity (Ngetich et al., 2012; Adekele et al., 2010). Maize crop being one of the most important cereal crop in sub-Saharan Africa (SSA) and an important staple food for more than 1.2 billion people in SSA and Latin America (Iken and Amusa, 2004); its production intensification must be emphasized to ensure food security. Maize yields on most smallholder farms in SSA has remained less than one tonne per hectare threatening households’ food security for over 70% of the rural populations as they primarily draw their livelihoods from agriculture (Muzari et al., 2012; Tauro et al., 2011). Integrated soil fertility management (ISFM) has been proven to reverse nutrient depleted soils to productive ones (Mucheru-Muna et al., 2014; Akinola et al., 2010; Vanlauwe et al., 2010), but the adoption of such technologies is still wanting. The overall goal of ISFM is to maximize the interactions that result from the potent combination of fertilizers, organic inputs, improved germplasm, and farmer knowledge (Vanlauwe et al., 2015). The ultimate outcome is improved productivity through wiser farm investments and field practices. It has also been proven that effective use of organic soil amendment methods in combination with inorganic fertilizer could help reverse the nutrient depletion trend (Mucheru-Muna et al., 2014; Akinola et al., 2010). The concern thus remains on why farmers are not adopting these strategies fully. Developing soil fertility management options for increasing productivity of staple food crops is a challenge in most parts of Sub-Saharan Africa, where soils are constrained by nitrogen (N) and phosphorus (P) deficiencies (Tauro et al., 2011). Phosphorus is one of the most important fertility components for maize as 200 it is the second largest consumed nutrient in global agriculture production, being surpassed only by nitrogen (Batten, 1992). Where sometimes nitrogen can be fixed into the soil by rotations of maize with beans having nitrogen fixing nodules (Ngetich et al., 2012), phosphorus deficiencies have to be supplemented from external inputs. Consequently, most of the N required for agricultural productivity in the continent comes largely from a judicious management of biological N fixation mostly through cereal-legume intercropping systems (Mucheru-Muna et al., 2010; Giller, 2001; Jerenyama et al., 2000; Giller et al., 1994) but P deficiencies have to be supplemented from external outputs. On another note, phosphorus has one of the lowest use efficiency among the plant macronutrients and a 50% yield reduction has been observed from P deficient conditions relative to the well fertilized production fields (Fageria and Baligar, 1997). Phosphorus availability is affected by farmer management practices and this must be explored. METHODOLOGY Description of the study site The study was carried out in Runyenjes sub-County in Embu County of Kenya. Runyenjes sub-County lies in the Upper Midland 2 and 3 (UM 2, UM3) at an altitude of approximately 1,200-2,070 meters above sea level. It covers 149 km2, out of which 96.26 km2 is arable land. The estimated population is 64,111 (GoK, 2010) with a population density of 561 persons per Km-2. The average farmholding is 0.4-0.8 ha. The annual mean temperature ranges from 12 °C to 27 °C. The average annual rainfall is 1495 mm and the range is between 1,000-2,000 mm. Rainfall distribution is bimodal with long rains from mid-March to May and short rains from October to December. The soils are well drained and mostly Nitisols. Sampling design and sample size The study design was a survey. Both quantitative and qualitative research frameworks were used. The study employed two sources of data. Primary sources; that included interviews, direct observation and photography. Secondary sources; included information from published work, relevant books, relevant seminars and conference meetings reports, and mostly the internet. Review of secondary data enabled the construction and adjustment of the questionnaire accordingly. Runyenjes sub-County was targeted for this study; Runyenjes was purposefully selected as it is characterized by smallholder farmers who grow both subsistence and cash crops that include maize. Stratified random sampling was used as the sub-County is divided into villages with households. A total of 100 households was obtained using sample size calculator software (CRS, 2007) at a confidence level of 95% and confidence interval of 9.8%. 201 S = Z2* (p) * (1-p)/ c2 (1) Where: S=Sample size, Z=Z value (e.g. 1.96 for 95% confidence level), p=percentage of picking a choice, expressed as decimal (0.5), c=confidence interval, expressed as decimal (0.098) Soil sampling and analysis Soil samples were collected from fields of each of the 100 farmers interviewed. The soil sample was collected at 20 cm depth. Three (3) soil cores were collected according to variability and size of land and then composited to produce one sample for analysis per farm. All samples were oven-dried at 400 C prior analysis. Available nutrient elements (P, K, Na, Ca, Mg and Mn) were analysed using Mehlich Double Acid method (Mehlich, 1984). Calorimetric method was used to determine organic carbon similar to Mairura et al. (2007). Total nitrogen was determined using Kjeldahl Digestion method (Anderson and Ingram, 1993). Soil pH was determined in a 1:1 (w/v) soil – water suspension with pH meter. Available trace elements (Fe, Zn and Cu) were extracted with 0.1 M HCl then determined with Atomic Absorption Spectrophotometer. Data Analysis Data was analysed using both descriptive and inferential statistics using SPSS version 20. Ordinary least squares regression was used to determine the factors affecting the use of inorganic and organic fertilizer for maize production. Ordinary least squares regression was also used to determine the management factors of farmers affecting amounts of available P on maize farms in Runyenjes sub-County. Binary logistic regression was used to determine the factors affecting combined used of inorganic and organic fertilizer with improved seeds for maize production in the sub-County. All these analysis was done using SPSS version. RESULTS AND DISCUSSIONS Use of inorganic fertilizer for maize production intensification by farmers in Runyenjes sub-County Results indicated that only 18% of the farmers interviewed in Runyenjes sub- County used sole fertilizer in maize production (Table I). Only two percent of the farmers interviewed applied sole manure to their maize crop, while 4% of the farmers had no fertilizer or manure input on their maize farms (Table I). Results showed 78% of the farmers as users of animal manure and out of these 76% combined fertilizer and animal manure in maize production (Table I). Results also indicated that there were 70% users of DAP fertilizer, 73% users of NPK and 38% users of CAN. The average rate of DAP fertilizer application in maize fields was 63.84 kg/ha (Table I). The average rate of NPK 23:23:0 application in the study area was 87.73 kg/ha, while that of CAN was 35.77 kg/ha (Table I). A 202 negative linear correlation was observed between size of land under maize and rate of DAP use, Pearson correlation = -0.534, p = 0.0001 (2-sided). This implies that as the size of land increases the rate of use of DAP decreases. Larger piece of land demands more fertilizer resources that a small holder farmer might not have access to. A negative linear correlation was also observed between size of land under maize and rate of NPK 23:23:0 use, Pearson correlation = -0.581, p = 0.0001 (2-sided). Larger holdings are likely to have lower rate of NPK use as larger farm holdings demand more fertilizer resources. This was also expounded by Zingore et al. (2007), who mentioned that farmers with little inputs usually apply them on farms where the main crop is grown; and these farms are usually near the homesteads. This means that farms further from the homes are usually under- fertilized. In recent years, there has been an increase in the use of high nutrient fertilizers, mainly for economic reasons (Ademba et al., 2015). This might be responsible for the increasing rates of fertilizer use in rural areas, although not up to recommended levels. A study by Rware et al. (2014), reported that fertilizer recommendations were still not followed in Kenya, although some increase in fertilizer use is notable. Socio-demographic factors affecting amounts of phosphorus fertilizer used by farmers in Runyenjes sub-County Three variables positively influenced use of phosphorus fertilizer by farmers in maize production (Table II), these were household size, off farm income and maize yield. Size of land under maize was negatively associated with phosphorus fertilizer use. This implies that as the household size and off-farm income of the household increased the higher the likelihood of using phosphorus fertilizer. Similarly as the maize yields obtained increased, the farmer is more likely to use phosphorus fertilizer. On the other hand, as size of land under maize increased the farmer was less likely to use phosphorus fertilizer. Results also indicated that a large household is more likely to use phosphorus fertilizer for maize production (Table II). Large households always demand more food resources and thus the need to increase maize production. Furthermore family labor associated with large household is cheap and readily available and thus increases tendency to use new technologies. As Marenya and Barret (2007) explain, family labor assumes great importance given that low incomes associated with rural small holder farmers constrain financial liquidity for hiring wage laborers. Mugwe et al. (2009a) and Mugwe et al. (2008) confirm the positive association of household labour and adoption of soil improving technologies. Maize yield was found to be positively associated with the use of phosphorus fertilizer in maize production (Table II). This reveals that households with increased yields attributable to inorganic fertilizer use were more likely to use phosphorus fertilizer in maize production. Mucheru-Muna et al. (2014) and Mucheru-Muna et al. (2007) concluded that inorganic fertilizer was a very 203 important input in the increase of maize yields in the Central Highlands of Kenya. With the poor fertility of soils in SSA, farmers are increasingly being aware of the fact that inorganic fertilizer is playing a major role in the increase of crop yields. Mapila et al. (2012) and Morris et al. (2007) explained that a household’s potential profitability from using fertilizer is determined by the responsiveness of the crop to which fertilizer is applied. The positive association of off-farm income with increased use of phosphorus fertilizers (Table II) in maize production is attributed to the fact that inorganic fertilizer purchase requires some monetary allocation. Having an off-farm income could enable a farmer to purchase fertilizer. Muzari et al. (2012) confirms this in that higher levels of income from other sources will lead to higher rates of adoption of yield-raising technology. Similarly, Marenya and Barret (2007) states that non-farm income from informal and formal non-agricultural employment proved quite important in fostering adoption of the integrated soil fertility management technologies in western Kenya. Negative association of size of land under maize and phosphorus fertilizer use (Table II) indicated that a farmer with a larger area under maize was less likely to use phosphorus fertilizer for maize production. Large farms demand more in terms of inputs and resources, and this might be a challenge to rural farmers who in most cases are poor. This agreed with Marenya and Barret (2007) who found that size of land was a significant factor affecting fertilizer use although estimated marginal effect of farm size was largest for inorganic fertilizer. Wiredu et al. (2014) measured the intensity of ISFM adoption by the proportion of land that was assigned to it, showing land as an important factor of adoption. Socio-demographic factors affecting use of animal manure among farmers in Runyenjes sub-County Regression results showed that use of animal manure for maize production was influenced by five variables; household head occupation (negative), size of maize land (positive), off farm earnings (positive), maize yield (positive) and training on animal manure (positive) (Table III). The positive association of off-farm income and animal manure use (Table III) in maize cultivation means that a household with an off farm income in the study area is more likely to use animal manure in maize cultivation. Limited manure availability is common in sub Saharan Africa and a household with extra income probably from off farm sources is able to purchase livestock that produce animal manure. Off-farm income enables a farmer to purchase inputs (Adolwa et al., 2012; Akinola et al., 2010) including livestock which are the main source of manure (Ajayi et al., 2007). For example, according to Zingore et al. (2007), small holder farmers with limited inputs resources usually apply them to farms near the homesteads only and neglect farms further from the homestead as opposed to farmers who are ‘resource richer’ who can afford to have resources enough for all the farms they have. 204 Maize yield positive association with the use of animal manure (Table III) in maize cultivation implies that a household with high maize yield was more likely to use animal manure in maize cultivation. Animal manure has been long known to improve soil fertility and increase yield (Mucheru-Muna et al., 2014; Ajayi et al., 2007; Mucheru-Muna et al., 2007). Good yields can allow a farmer to sell excess produce and purchase livestock associated with production of animal manure. Training had positive effect on the use of animal manure (Table III). This implies that training helps farmers to correctly use animal manure for crop production. In tandem with this observation, Misiko and Ramisch (2007) noted that integrated soil fertility management strategies have been met with major challenges because they are knowledge-intensive and their adaptations and applications are diverse (Adolwa et al., 2012), often requiring training for proper utilization. Macharia et al. (2012) also emphasizes that having knowledge on a technology puts a farmer at a position to use the technology. Socio-demographic factors affecting use of combined inorganic and organic manure with improved seeds among farmers in Runyenjes sub-County Regression results showed that size of land under maize (negative), household labour (positive), off farm income (negative) and tropical livestock unit (positive) had significantly statistical association with the use of combined inorganic fertilizer and animal manure with improved seeds (Table IV). Size of land under maize was negatively associated with the use of combined inorganic fertilizer and animal manure with improved seeds (Table 5). This shows that household with larger sizes of land under maize were less likely to use a combination of inorganic fertilizer and animal manure with improved seeds in maize cultivation. And although farmers owning large pieces of land are associated with use of technologies (Barret and Marenya, 2007; Mugwe et al., 2008), this is otherwise for farmers with large pieces of land under maize crop. Demand of resources and inputs associated with a large piece of land under maize can hinder a household’s ability to acquire them. Macharia et al. (2012) also found farm size to be a negative predictor as adoption increased with decrease in farm size. Enough household labour was a significant factor influencing use of combined inorganic fertilizer and animal manure with improved seeds for maize production (Table IV). Households with enough and available family labour have been known to use technologies. Marenya and Barret, (2007) agrees that enough family labour reduces a household dependence on hired labor that often also requires supervision. Mugwe et al. (2008) mentioned household labor as a factor immensely influencing adoption of soil management technologies. Having an off-farm earning was negatively associated with the combined use of inorganic fertilizer and animal manure with improved seeds (Table IV). Farmers with off-farm incomes were less likely to use a combination of inorganic fertilizer 205 and animal manure with improved seeds. Adolwa et al. (2012) explains that it is probable that individuals with higher off-farm incomes invested their time, energies and money in non-farm activities at the expense of ISFM, hence the disparity with other studies. It is thus not surprising that Reardon et al. (2000) identified having an off-farm income as one of the reasons for farmers’ failure to adopt hybrid maize in Botswana. Tropical livestock unit had positive significant association with the combination of inorganic fertilizer and animal manure with improved seeds for maize production (Table IV). This implies that a household with a higher livestock number is more likely to use a combination of inorganic fertilizer and animal manure with improved seeds in maize production. Livestock is very crucial in a farming system as it is the major source of manure and draft power (Adolwa et al., 2012). Livestock ownership is an indication of wealth status of small holder farmers (Adolwa et al., 2012; Zingore et al., 2007). Farmers can sell portions of their livestock to acquire maize inputs, as livestock and household assets increase the availability of capital which makes investment in land-enhancing technologies feasible (Akinola et al., 2010). Soil fertility status of maize farms in Runyenjes sub-County Soil analysis results indicated that all the farms surveyed had moderately acidic soils with pH ranging from 4.59 to 6.95 with a mean of 5.60 (Table V). The mean available phosphorus level in the study area as 39.75 ppm P indicating moderate levels of the nutrient. The available P observations were evenly distributed. The interquartile range for available P lied between 35 and 40 ppm while the median was about 35 ppm (data not shown). The soils had 0.21% total N, 2.21% total OC, 1.12% K, 5.9% Ca, 2.44% Mg, 1.04% Mn and 0.62% Na (Table 6). The soils also had Cu, Fe, Zn ions of 4.62, 26.07 and 28.01 respectively (Table V). Farmer management factors affecting phosphorus levels in farm soils of Runyenjes sub-County Regression results showed that amount of NPK 23:23:0, intercropping of maize, liming, and total land owned as positive significant management factors affecting amounts of available P in farmer soils (Table VI). The amount of NPK 23:23:0 a farmer applied on his maize farm had significant and positive associations with the amounts of available P in small holder farms in the study area. This means that a farmer who applied more NPK 23:23:0 on his maize farm was more likely to have a farm with high levels of available P. High phosphorus (P) fixation is a serious problems in Africa, and although phosphatic fertilizers are imported and costly for the average farmer (Yamoah et al., 1996), it is one of the most viable and practical way of increasing phosphorus levels in the soil. Intercropping of maize with a legume was found to be a positive predictor of whether levels of available phosphorus were high in small holder farms in Runyenjes sub-County (Table VI). This implied that farmers who intercropped 206 their maize with other legumes were more likely to have high amounts of available P in their maize farms. Synergies between N (probably added by legume intercrop) and P nutrients allow for more uptake of P than sole P application. Ademba et al. (2015) explained that there is higher P nutrient availability with combined N and P than the sole P application, and this could be attributed to the synergistic N enhancement of P uptake. Liming of soil by a small holder farmer was found to be statistically significant and positively associated with high available P amounts (Table VI). This shows the more a farmer applied lime to reduce soil acidity, the more the amount of available P. Sustainable agriculture is threatened by widespread acidity in many parts (Yamoah et al., 1996), however application of lime has been reported to significantly improve soil fertility (Nduwumuremyi et al., 2013). In regards to this, liming must be done considerably. For example, Yamoah et al. (1996) explains that experimental results showed that liming was most effective at low P rates and its effect on yield diminished with increasing P fertilization. Reduced acidity results in more nutrient availability, especially of nutrients prone to leaching and movement during rainfall. Mbakaya et al. (2011) explains that acidity has been found to retard plant growth through H+ and Al3+ ionic effects, mineral ion toxicity or by indirectly interfering with mineral availability. TABLES Table I: Fertilizer application among farmers in Runyenjes sub-County Type of application Percentage (N=100) Sole DAP application 3 Sole NPK application 1 Sole CAN application 0 Sole manure application 2 Combined DAP + NPK only 7 Combined DAP + CAN only 3 Combined NPK + CAN only 1 Combined DAP + NPK + CAN only 3 Combined DAP + manure only 7 Combined NPK + manure only 15 Combined CAN + manure only 1 Combined DAP + NPK + manure 22 Combined DAP + CAN + manure 7 Combined NPK + CAN + manure 6 Combined DAP + NPK + CAN + manure 18 No input 4 Fertilizer input Percentage Rate of use among users Rate of use among whole 207 sample DAP 70 87.76 63.84 NPK 23:23:0 73 109.67 87.73 CAN 38 94.13 35.77 Table II: Socio-demographic factors affecting amounts of phosphorus fertilizer used by farmers in Runyenjes sub-County Independent variables Β S.E. Beta t Sig. (Constant) 250.227 82.164 3.045 0.003 HHH gender -15.775 24.824 -0.058 -0.635 0.527 HHH age -0.650 1.221 -0.081 -0.532 0.596 HHH education -20.209 15.564 -0.133 -1.298 0.198 HHH occupation -22.865 17.316 -0.127 -1.320 0.190 HH size 8.979 4.837 0.185 1.856 0.067* Farming experience (yrs) -0.382 1.160 -0.053 -0.330 0.742 Total land owned (ha) 0.430 2.916 0.015 0.148 0.883 Size of land under maize (ha) -41.467 14.301 -0.328 -2.900 0.005*** Off-farm earnings (Ksh) 16.130 6.364 0.236 2.534 0.013** Tropical Livestock Unit 4.428 9.213 0.046 0.481 0.632 Yield (t/ha) 30.660 12.605 0.226 2.432 0.017** Dependent Variable: Amount of phosphorus fertilizer applied per season N =100, ***Significant at 1% probability level, **Significant at 5% probability level, *Significant at 10% probability Table III: Socio-demographic factors influencing use of animal manure among farmers in Runyenjes sub-County Independent variables Β S.E. Beta t Sig. (Constant) 0.201 1.134 0.178 0.859 HHH gender -0.386 0.270 -0.139 -1.428 0.157 HHH education 0.056 0.165 0.036 0.339 0.735 HHH occupation -0.586 0.306 -0.192 -1.918 0.058* HH size 0.060 0.051 0.123 1.184 0.240 Farming experience (yrs) 0.002 0.013 0.028 0.163 0.871 Size of land under maize (ha) 0.495 0.127 0.386 3.885 0.000*** Off farm earnings (Ksh) 0.138 0.069 0.199 2.016 0.047** Maize yield in (t/ha) 0.275 0.134 0.200 2.053 0.043** Training on animal manure 0.477 0.285 0.154 1.674 0.098* a. Dependent Variable: use of animal manure (TLU) N =100, ***Significant at 1% probability level, **Significant at 5% probability level, *Significant at 10% probability 208 Table IV: Socio-demographic factors affecting use of combined inorganic, animal manure and improved seeds among farmers in Runyenjes sub-County Independent variables Β S.E. Beta t Sig. (Constant) 0.317 0.363 0.874 0.384 HHH gender -0.140 0.097 -0.149 -1.445 0.152 HHH age -0.006 0.005 -0.209 -1.187 0.238 HHH education 0.062 0.058 0.118 1.064 0.290 HHH occupation 0.058 0.111 0.057 0.528 0.599 HH size 0.028 0.018 0.167 1.576 0.118 Farming experience 0.007 0.005 0.290 1.546 0.126 Size of land under maize -0.133 0.047 -0.307 -2.857 0.005** Enough HH labour 0.266 0.105 0.258 2.548 0.013** Off farm earnings -0.043 0.025 -0.182 -1.704 0.092* TLU 0.099 0.036 0.291 2.726 0.008** a. Dependent Variable: combined use of inorganic fertilizer, animal manure and improved seeds N =100, **Significant at 5% probability level, *Significant at 10% probability Table V: Soil properties of maize farms in Runyenjes sub-County Parameters (Soil chemical properties) Value Soil pH 5.60 Total Nitrogen % 0.21 Total Organic Carbon % 2.21 Phosphorus in mg/kg 39.75 Potassium me% 1.12 Calcium me% 5.9 Magnesium me% 2.44 Manganese me% 1.04 Copper ppm 4.62 Iron ppm 26.07 Zinc ppm 28.01 Sodium me% 0.62 209 Table VI: Farmers management factors affecting available phosphorus amounts in maize farms of Runyenjes sub-County N =100, **Significant at 5% probability level, *Significant at 10% probability level CONCLUSIONS AND RECOMMENDATIONS The implications of these results are that research should put into consideration the demographic and socioeconomic status of farmers during the coining of yield increasing technologies to ensure that the needs of all the small holder farmers are addressed. In regards to education level, researchers and extension workers can aim to disseminate integrated soil fertility management information in disintegrated forms that can be understood and incorporated by small holder farmers who have low literacy levels. The gender aspect should be considered and trainers should ensure that both female and male trainees are well represented during trainings and workshops. It is also important for governments to provide off-farm opportunities as off-farm income is important in the adoption of yield increasing strategies. To increase available P levels for maize production in small holder farms, farmers must be empowered to be able to access inorganic fertilizer, they should also be capacitated to access lime that reduces acidity on the farms and to use organic amendments. Farmers should also be encouraged to intercrop or rotate their maize crop with legumes as the positive interaction will not only increase yields but improve nutrients in the soil. Extensive grassroot support must be explored and encouraged to ensure that farmers adopt fully ISFM strategies. Farmers are willing to take up yield increasing strategies but frustrations during implementation due to lack of knowledge coupled with poor grassroot support can lead to abandonment of such strategies. Therefore, all these factors must be incorporated by policy Independent variable β S.E. Beta T Sig. (Constant) 66.254 36.888 1.796 0.076 Amount of DAP applied -0.026 0.099 -0.034 -0.259 0.796 Amount of NPK 23:23:0 applied 0.137 0.082 0.200 1.683 0.096* Amount TSP applied -0.122 0.343 -0.040 -0.357 0.722 Amount CAN applied 0.022 0.063 0.050 0.343 0.732 Amount of foliar applied 5.571 10.878 0.086 0.512 0.610 Intercropping of maize 11.301 5.474 0.227 2.064 0.042** Continuous P application 4.287 4.832 0.098 0.887 0.378 Animal manure use -3.663 7.676 -0.059 -0.477 0.634 Liming of soil 46.442 17.684 0.368 2.626 0.010*** Training on inorganic fertilizer 2.980 3.087 0.114 0.965 0.337 Training on animal manure -5.290 6.641 -0.098 -0.796 0.428 Size of land under maize -3.805 2.995 -0.171 -1.270 0.207 Dependent Variable: P in mg/kg 210 makers, researchers and extension workers to ensure that they are addressed appropriately during formulation of yield increasing technologies. ACKNOWLEDGEMENTS The authors would like to acknowledge International Maize and Wheat Improvement Center (CIMMYT) and International Plant Nutrition Institute (IPNI) that jointly funded this research. Special thanks to Ministry of Agriculture (MoA)-Embu and Kenya Agricultural Livestock Research Organization (KALRO)-Embu for their support during acquisition of data. Many thanks also go to the small holder farmers of Runyenjes sub-County who availed themselves during the interview sessions. REFERENCES Adeleke, S., Abdul B. K. and Zuzana, B. (2010). Smallholder Agriculture in East Africa: Trends, Constraints and Opportunities, Working Papers Series N° 105 African Development Bank, Tunis, Tunisia. Ademba, J. S., Kwach, J.K., Esilaba, A.O and Ngari, S.M (2015). The Effects of Phosphate Fertilizers and Manure on Maize Yields in South Western Kenya. East African Agricultural and Forestry JournalVolume 81, Issue 1, 2015, pages 1-11 Adolwa, I.S., Okoth, P.F., Mulwa, R.M., Esilaba, A.O., Mairura, F.S. and Nambiro,E.(2012). Analysis of communication and dissemination channels influencing the adoption of integrated soil fertility management in Western Kenya. The Journal of Agricultural Education and Extension, 18:1, 71-86, DOI: 10.1080/1389224X.2012.638782 Ajayi, O.C., Akinnifesi, F. K., Sileshi, G., and Chakeredza, S. (2007). Adoption of renewable soil fertility replenishment technologies in the southern African region: Lessons learnt and the way forward. Natural Resources Forum 31 (2007) 306–317. Blackwell Publishing Ltd Akinola, A.A., Arega, D.A., Adeyemo, R., Sanogo, D., Olanrewaju, A.S., Nwoke C., and Nziguheba, G. (2010). Determinants of adoption and intensity of use of balance nutrient management systems technologies in the Northern Guinea Savanna of Nigeria. Quarterly Journal of International Agriculture 49: 2010. Anderson, J.M., Ingram, J.S.I., 1993. Tropical soil biology and fertility. A Handbook of Methods. CAB International, Wallingford. Batten, G.D. (1992). A review of phosphorus efficiency in wheat. Plant Soil 146:163-168. Creative Research Systems (CRS), (2007). “The Survey System: Sample Size Calculator.” CRS. Accessed September 10, 2014. http://www.surveysystem.com/sscalc.html. 211 Fageria, N.K. and Baligar, V.C. (1997). Response of common bean, upland rice, corn, wheat, and soybean to soil fertility of an Oxisol. Journal of Plant Nutrition, 20, 1279-1289. http://dx.doi.org/10.1080/01904169709365335. Giller, K.E., (2001). Nitrogen Fixation in Tropical Cropping Systems, 2nd edition. CAB International, Wallingford. Giller, K.E., McDonagh, J.F., Cadisch, G., (1994). Can biological nitrogen fixation sustain agriculture in the tropics? In: Syers, J.K., Rimmer, D.L. (Eds.), Soil Science and Sustainable Land Management in the Tropics. CAB International, Wallingford. Government of Kenya, (GOK) (2010). The 2009 Kenya National Census Results. Ministry of Home Affairs, Nairobi, Kenya. Iken, J. E. and Amusa, N.A. (2004). Maize research and productivity in Nigeria. African Journal of Biotechnology 3(6):302-307. Integrated soil fertility management in Africa: From microbes to markets (ISFM Africa 2012). Conference Information, Program and Abstracts. Jerenyama, P., Hestermann, O.B., Waddington, S.R., Hardwood, R.R., (2000). Relay intercropping of sunnhemp and cowpea into a smallholder maize system in Zimbambwe. Agronomy Journal 92, 239–244. Macharia, J.M., Mugwe, J.N., Mucheru-Muna, M. and Mugendi, D.N. (2012). Factors influencing levels of knowledge in soil fertility management in the Central Highlands of Kenya. Journal of Agricultural Science and Technology, B 4 (2014) 701-711 Mairura, F.S., Mugendi, D.N., Mwanje, J.I., Ramisch, J.J., Mbugua, P.K., Chianu, J.N. (2008). Scientific evaluation of smallholder land use knowledge in central Kenya. Land Degradation & Development 19: 77–90. Mapila, M.A.T.J., Njuki, J., Delve, R. J., Zingore S. and Matibini. J. (2012). Determinants of fertiliser use by smallholder maize farmers in the Chinyanja Triangle in Malawi, Mozambique and Zambia. Agrekon Volume 51, Issue 1, March 2012, pages 21-41 Marenya, P.P., and Barrett, C.B. (2007). Household-level determinants of adoption of improved natural resources management practices among smallholder farmers in Western Kenya. Food Policy 32: 515–36 Mbakaya, D. (2007). Annual Report for KARI Kakamega showing mean maize yields for each soil fertility demonstrated technology in Kabras, long rains (2007). Mehlich, A. 1984. “Mehlich 3 Soil Test Extractant: A Modification of Mehlich 2 Extractant.” Comm. Soil Sci. Plant Anal. 15 (12): 1409–1416 Misiko, M. and Ramisch, J., (2007). Integrated soil fertility management technologies: Review for scaling up. In A. Bationo (eds.), Advances in Integrated Soil Fertility Management in Sub-Saharan Africa: Challenges and Opportunities, Soil Biology, 873-880. 212 Morris, M., Kelly, V., Kopicki, R. J., and Byerlee, D. (2007). Fertilizer use in African agriculture: Lessons learned and good practice guidelines. Washington DC: World Bank Mucheru-Muna, M., Mugendi, D., Kung'u, J., Mugwe, J., Bationo, A. (2007). Effects of organic and mineral fertilizer technologies on maize yield and soil chemical properties in a maize cropping system in Meru South District, Kenya. Agroforestry Systems, 69: 189-197. Mucheru-Muna, M., Pypers, P., Mugendi, D., Kung'u, J., Mugwe, J., Merckx, R., and Vanlauwe, B. (2010). A staggered maize–legume intercrop arrangement robustly increases crop yields and economic returns in the highlands of Central Kenya. Field Crops Research 115 (2010) 132–139 Mucheru-Muna, M.W., Mugendi, D.N., Pypers, P., Mugwe, J.N., Kung'u, J.B., Vanlauwe, B. and Merckx, R. (2014). Enhancing maize productivity and profitability using organic inputs and mineral fertilizer in central Kenya small-hold farms. Experimental Agriculture 50: 250-269 Mugwe, J.N., Mucheru-Muna, M., Mugendi, D.N., Kung’u, J.B., Bationo, A. and Mairura, F. (2009b) Adoption potential of selected organic inputs for improving soil fertility in the Central Highlands of Kenya. Agroforestry Systems, 76: 467-485 Mugwe, J.N., Mugendi D.N., Mucheru-Muna, M., Odee, D. and Mairura, F. (2009a). Effect of selected organic materials and inorganic fertilizer on the soil fertility of a Humic Nitisol in the central highlands of Kenya. Soil Use and Management, December 2009, 25, 434–440 doi: 10.1111/j.1475- 2743.2009.00244.x Mugwe, J.N., Mugendi, D.N., Mucheru-Muna, M., Merckx, R., Chianu, J. and Vanlauwe, B. (2008). Determinants of the decision to adopt integrated soil fertility management practices by smallholder farmers in the Central Highlands of Kenya. Cambridge University Press, United Kingdom. Muzari, W., Gatsi, W. and Muvhunzi, S. (2012). The Impacts of Technology Adoption on Smallholder Agricultural Productivity in Sub-Saharan Africa: A Review. Journal of Sustainable Development; Vol. 5, No. 8; 2012 ISSN 1913-9063 E-ISSN 1913-9071 Muzari, W., Gatsi, W. and Muvhunzi, S. (2012). The Impacts of Technology Adoption on Smallholder Agricultural Productivity in Sub-Saharan Africa: A Review. Journal of Sustainable Development; Vol. 5, No. 8; 2012 ISSN 1913- 9063 E-ISSN 1913-9071 Nduwumuremyi, A., Ruganzu, V., Mugwe, J.N., and Rusanganwa, A.C. (2013).Effects of unburned lime on soil pH and base cations in acidic soil. Hindawi Publishing Corporation, ISRN Soil Science. Volume 2013, Article ID 707569, http://dx.doi.org/10.1155/2013/707569 Ngetich, F.K., Shisanya, C.A., Mugwe, J.N., Mucheru-Muna, M. and Mugendi, D.N. (2012). The potential of organic and inorganic nutrient sources in Sub- Saharan African crop farming systems, soil fertility improvement 213 and integrated nutrient management - A global perspective. Dr. Joann Whalen (Ed.), ISBN: 978-953-307-945-5 Reardon, T., Taylor, J.E., Stamoulis, K., Lanjouw, P. & Balisacan, A. (2000). Effects of nonfarm employment on rural income inequality in developing Countries: An Investment Perspective. Journal of Agricultural Economics, 51(2), pp. 266_288. Rware, H., Wairegi, L., Oduor, G., Macharia, M., Romney, D., Tarfa, B.D., Maria, R., Ley G., Tetteh, F., Makumba, W., Dicko, M., Nouri, M., Leon, N.N., Gondwe, B., Kaizzi, K., Negash, D., Kibunja, C., Alawy, A., Wartmann C., Rutsimba, E. (2014). Assessing the Potential to Change Stakeholders Knowledge and Practices on Fertilizer Recommendations in Africa. Agricultural Sciences, 5, 1384-1391. http://dx.doi.org/10.4236/as.2014.514149 Tauro, T. P., Mapanda, F., Mtombeni, G., Shumba, A. and Dhliwayo, D. K. (2011). Soil acidity: Is it a problem in large scale commercial farms in Zimbabwe? Journal of Agricultural Science and Technology, ISSN 1939-1250, USA. February 2011, Volume 5, No.2 (Serial No.33) Vanlauwe, B., Bationo, A., Chianu, J., Giller, K.E., Merckx, R., Mokwunye, U., Ohiokpehai, O., Pypers, P., Tabo, R., Shepherd, K., Smaling, E.M.A., Woomer, P.L., (2010). Integrated soil fertility management: operational definition and consequences for implementation and dissemination. Outlook Agric. 39, 17–24. Vanlauwe, B., Descheemaeker, K., Giller, K. E., Huising, J., Merckx, R., Nziguheba, G., Wendt, J., Zingore, S. (2015). Integrated soil fertility management in sub- Saharan Africa: unraveling local adaptation. SOIL Discuss, 1, 491–508, 2015 Wiredu, A.N., Martey, E. and Fosu M. (2014). Describing adoption of integrated soil fertility management practices in Northern Ghana. Conference on International Research on Food Security, Natural Resource Management and Rural Development organised by the Czech University of Life Sciences Prague. Tropentag 2014, Prague, Czech Republic. September 17-19, 2014 Yamoah, C., N. Ngong, M. C., and Dias, D. K. W. Reduction of P fertilizer requirement using lime and mucuna on high P-sorption soil of NW Cameroon, African Crop Science Journal, vol. 4, pp. 441–451, 1996. Zingore, S., Murwira, H.K., Delve, R.J. and Giller, K.E. (2007). Influence of nutrient management strategies on variability of soil fertility, crop yields and nutrient balances on smallholder farms in Zimbabwe. Agriculture, Ecosystems and Environment 119 (2007) 112–126 214 5.4 Combining agroforestry with conservation agriculture for soil fertility enhancement in the dryland context: case of Machakos County, Kenya Vincent Rabacha,b*, Monicah Mucheru-Munaa, Jonathan Muriukib & James Koskea a Kenyatta University. P.O BOX 43844-00100, Nairobi b World Agroforestry Centre. P.O BOX 30677-00100, Nairobi *Corresponding author. Mobile: +254713876735.Email: rabachv@gmail.com ABSTRACT Kenyan drylands are faced with challenges of land and vegetation degradation, have low productivity, limited majorly by moisture, then soil nutrients. Conservation agriculture with trees is one opportunity to redress these challenges through minimum tillage, soil cover and crop rotation. The study examined the differences in soil properties between conservation agriculture (CA) and conventional agriculture (COA) both with and without trees at an onsite experiment station in Machakos, Kenya over three seasons (SR 2013, LR 2014 and SR 2014). A randomized complete block design experiment was laid with 2 blocks (CA and COA) and ten treatments replicated thrice. Calliandra calothyrsus Meissn. and Gliricidia sepium Jacq.) were planted in three different spatial patterns (1.5 m x 1 m, 3 m x 1 m, and 4.5 m x 1 m) for maize-legume intercrops. Soil samples were taken in October 2015 from top and sub soil and analyzed for selected physical and chemical characteristics using standard laboratory procedures. Data was subjected to analysis of variance for significant relations and differences, aided by Genstat software version 14. Fisher’s least significant difference test was used to separate treatment means at 5% alpha level. Results showed significant increase in soil moisture under conservation agriculture with trees. Sole CA retained more moisture than sole COA (31.56 and 26.54% vol. respectively, p<0.001). Nitrogen, organic carbon, sodium and potassium were higher under CA while cation exchange capacity (CEC) was significantly (p=0.004) higher (14.372 cmolc/kg) in COA than in CA (12.718 cmolc/kg), and strongly correlated with clay content (r=+0.869). Exchangeable calcium and magnesium were also significant under COA, which also had higher soil salt content (electrical conductivity) and phosphorus levels. Bulk density at the surface layer was higher under COA than CA but not significant (p=0.97).Conclusively; both conservation and conventional practice enhance soil fertility with an improvement in majority of the soil properties noted under CA, making CA a more preferred option for enhancing soil heath in the dry-lands. Key words: Conservation agriculture with trees; Conventional agriculture; spatial patterns; Calliandra calothyrsus; Gliricidia sepium; soil health 215 INTRODUCTION The soils of the arid and semi arid areas all over the World are known for historic degradation (Lal, 2011) .These zones are also prone to vegetation destruction as a result of the disproportionate human disturbance (Curtin et al., 2002) and soil erosion which are responsible for their reduced primary productivity (Reynolds et al., 2007) which will continue to decline (Parr et al., 1990; Enfors et al., 2008).There also occurs general land degradation as a result of poor land management and this has become a development and environmental issue of global concern (Zika and Erb, 2009).With these regard, strategies that are needed to cope with the challenges of the drylands must address the challenges of moisture limitations and the general degradation. Of primary importance are the strategies that offer better soil and water management in these areas if the productivity is to be sufficiently increased (Bryan et al., 2009; Dixon et al., 2013) Agroforestry systems suffice as an option for the drylands since they are sustainable land use alternatives ,tending to imitate natural system characteristics especially those that are beneficial to soil and enhance the well- being thereof (Tornquist et al., 1999). In fact, the potential of agroforestry systems as a means of achieving sustainable land use has been promoted since the 1980s (Kursten, 2000) and has since been widely and progressively integrated into good farming practice. These systems either with or without conservation agriculture (CA) therefore have a considerable influence on soil properties. Mitchell et al. (2007) (as cited in Garland et al., 2011) and Ketema and Yimer (2014) noted that no tillage practices increases soil quality, water quality and organic carbon thereby enhancing carbon sequestration to reduce emissions. Makumba et al. (2007) also realized increased soil organic carbon in Gliricidia based system in comparison to sole maize, emphasizing the importance of tree integration into farms. Further more, Jose (2009) show that the role of agroforestry in biological fixation of nitrogen is a common phenomenon in the tropical areas. Even in the case of non N-fixing trees, the soil properties in entirety are still enhanced as has been noted by other researchers (Schroth and Sinclair, 2003; Jose et al., 2004). Soil porosity (amount of pores) has been found to be greater to the tune of 5 times in tree based systems alongside increase in enzyme activities (Udawatta et al, 2008) which also have been shown to improve organic matter and microbial biomass of soils (Lee and Jose, 2003).The high organic matter in agroforestry and conservation tillage systems (Palm et al., 2014) is also known to improve soil porosity, aggregate and structure, contrary to conventional tillage that brings about discontinuity in pore space between cultivated layer and the sub layers of the soil (Yimer et al., 2008). Ketema and Yimer (2014) in their study additionally noted low bulk density in agroforestry based conservation tillage compared to maize based conventional tillage (1.09 gcm3 and 1.18gcm3, respectively). The high bulk density in conventional farming is contributed to by continuous tillage which causes structural deterioration; mineralization of soil organic matter and 216 compaction due to lower infiltration rates (Yimer et al., 2008).In accordance with the study of Harvey et al. (2006), the integration of trees into farming systems has also been known to lead to considerable conservation of soil biodiversity. Moreover, agroforestry based conservation agriculture improves soil moisture content due to less exposure of soil to the direct impacts of sunlight thus a reduced evapotranspiration rate, meaning more of the moisture is able to be maintained within the soil (Ketema and Yimer, 2014; Seobi et al., 2005 ). Minimum tillage and maintenance of permanent soil cover, which are among the two principles of conservation agriculture have also been known to be moderators of the surface conditions of land, leading to improved yields, surface runoff control, increased net benefits as a result of the reduction in the costs of production, moderating soil temperatures as well as enhancing the rooting of crops (Gill et al., 1996; Govaerts et al., 2009). The retention of residue on the soil also lead to greater soil organic carbon in the soil surface as compared to removal of such residues which deteriorates the organic matter dynamics thus perpetuating soil carbon loss and impacting low fertility as was realized in a study by Guto et al. (2011). MATERIALS AND METHODS The study was carried out at the Machakos Agricultural Training Centre (coordinates E037o14.303ˈ and S 01o32.738ˈ), and utilized a randomized complete block design experiment with two blocks (conservation agriculture (CA) and conventional agriculture (COA)) with ten treatments replicated thrice, summing to 60 plots measuring 12m by 12 m. The treatments included Calliandra calothyrsus, Gliricidia sepium and pigeon peas (Cajanas cajan) at three spatial patterns i.e. 1.5m × 1m, 3 m ×1m and 4.5m × 1m; and control. The trees were integrated within maize-legume intercrops. Delta T capacitance probe access tubes were installed in every plot, and moisture readings taken fortnightly. The experiment ran from short rains (SR) 2013 to short rains 2014. In October 2015, composite soil samples were taken from 5 spots within all the plots at depths of 10-30cm. Two Soil profiles were also dug each at a control plot within both blocks and sampled for bulk density at depth intervals of 20cm, up to 120cm depth. The samples were analyzed for organic carbon and nitrogen, total carbon and nitrogen, soil texture, cation exchange capacity, electrical conductivity, pH (using standard methods described by Anderson and Ingram (1993)), extractable P, Ca, Mg, K & Na and exchangeable bases (using Mehlich III extraction method ( Mehlich, 1984)).The analyses followed the standard operation procedures of the World Agroforestry Centre Plant-Soil Spectral analysis laboratories. The data was subjected to analysis of variance using Genstat software version 14 and the parameters were considered significant at p<=0.005. Different in means of soil parameters between blocks and treatments were separated using Fisher’s least significant difference test. 217 RESULTS AND DISCUSSION Soil moisture Soil moisture retention was found to be generally higher in conservation agriculture (CA) trials compared to conventional agriculture (COA) in all the three seasons (Figure 1), and statistically significant (p<0.001) (Table I). CA recorded grand mean moisture retention of 31.56 % vol. against COA’s 26.24 % vol. Reducing tree spacing increased moisture retention, with Calliandra at 1.5 m showing high moisture rates under CA (40.25% vol.) compared to the same treatment under COA (29.67% vol.) (Table I). Moisture consistently increased after each season (Table I) and was significant among seasons (p=0.03) and treatments (p<0.001). Cation Exchange Capacity (CEC) CEC was not significant among treatments (p=0.641) but was significant between blocks (COA=14.39 cmolc/kg and CA=12.75 cmolc/kg, p=0.004) (Table II).The lowest CEC was in Calliandra at 1.5m (9.11 cmolc/kg) under CA and the highest in Pigeon peas at 4.5m (15.35 cmolc/kg) under COA. There was no CEC below 5 cmolc/kg (which’s normally considered very low (Landon, 1991). Bulk Density (BD) Bulk density was also not significant among treatments and between blocks. Mean BD for CA was 0.971g/cm^3 while that for COA was 0.92g/cm^3 (p=0.18).Calliandra spaced at 4.5m had the highest BD of 1.37g/cm^3 under CA and the lowest being 0.877g/cm^3 for pigeon peas at 1.5m under COA (Table III).Comparison of BD down the soil profile for both blocks did not reveal any statistical difference though the grand mean BD for COA was slightly higher than CA (0.945 and 0.88 gcm^3 respectively) as shown in Table III. Electrical Conductivity (Ecd) The study revealed higher (and not significant) amounts of soil in COA system than in CA (COA=0.0627 dS m^-1 and 0.5406 dS m^-1 respectively; p=0.33) (Table II). Intervention with trees at different spacings moreover did not differently affect salinity of the soils both under CA and COA (p =0.503), although the electrical conductivity in Calliandra at 4.5m under CA was a high of 4.715 dS m^-1. Soil pH (water) The highest pH was recorded at Calliandra at 4.5m under CA (13.6) while the lowest was at Calliandra at 1.5m under COA (6.93).CA had higher pH than COA (7.6 and 7.1 respectively), though the difference was not significant (p=0.391) as is in Table II. Tillage did not affect pH. 218 Exchangeable bases Mehlich 3 phosphorus (m3.P), sodium (ExNa) and potassium (ExK) were more available under CA compared to COA. On the other hand, exchangeable calcium (ExCa) and magnesium (ExMg) were significantly higher in COA than in CA (p<0.001) (Table IV). Intervention with trees both in CA and COA also led to more nutrient availability as compared to control treatments. Soil Texture The percentage of sand, silt and clay were not statistically different. However, the clay content was found to be strongly positively correlated to the Cation Exchange Capacity (CEC) of the soil (correlation coefficient =0.869), an implication that clay content increases CEC. This phenomenon of CEC increasing with the amount of clay was evident in the study where the treatment with the lowest clay content of 57% (Calliandra at 4.5m under CA) recorded the lowest CEC of 9.11Cmolc/kg (Table V).Individual treatment means were not significant between farming systems CA and COA. Carbon and Nitrogen CA had more Nitrogen both total (TN) and Organic (ON) compared to COA (TN=0.17% for CA and 0.12 % for COA; ON =0.16% for CA and 0.11% for COA).Total Carbon (TC) was however higher under COA (Table VI) but was not significant (p=0.07). The high moisture content experienced under CA with trees is due to the adequate vegetative soil cover provided by the tree component and plant residue and therefore evapotranspiration was reduced in these treatments, thus more moisture retained. Ketema and Yimer (2014) argued that agroforestry practices certainly improve soil moisture. High moisture under CA tallies with the studies of other researchers (Thierfelder and Wall, 2009). CA significantly increases soil water availability and thus higher moisture content (Corbeels et al., 2014), and also as result of mulch and minimum soil disturbance (Thierfelder et al., 2013; Rockstrom et al., 2009).The incorporation of trees into CA systems is therefore further proven in this study as an apt option for moisture retention and soil improvement. The of results on the soil physical and chemical properties point to a general acceptance that CA has led to much improvement of soil quality compared to COA though majority are not significantly high. The lack of statistical significance in the properties does not however infer lack of practical importance in the greater values recorded, which is in line with other studies. Mloza-Banda et al. (2016) for instance in a study to compare tillage effects of CA and ridge tillage did not find any significance though CA had lower bulk density than ridge tillage (1.49-1.58 Mg m_3 and 1.53- 1.59 Mg m_3 respectively).Comparatively to the study of Mloza-Banda et al .(2016), the study found lower BD at the top most soil layer (0-20cm) in CA than in COA ( 0.937 and 0.985g/cm3 respectively 219 p=0.97) .This can be attributed to cultivation induced compaction in the COA practice. Soil pH not being affected by tillage in the study is in tandem with the findings of other researchers (Rasmussen, 1999; Busari et al., 2015).However, pH was very crucial in effectuation of CEC, the correlation of which is negatively strong (coefficient= -0.8258), where higher pH translated to very low CEC and vise versa. This is also reinforced by the study of Schwab et al. (2015) in which they ascribed low CEC in their samples to low pH. Different researchers have found conflicting results on the effects of tillage on pH. Rahman et al. (2008) found it to be lower in no-till systems (a form of CA) than in Conventional tillage and attributed this to accumulation of organic matter and elevation in electrolyte concentration while Lal (1997) found it to be conversely high in no-till systems compared to conventional systems. The high carbon and nitrogen in CA over COA can be attributed to minimum soil disturbance which lessens the destruction of soil structure and aggregate exposure. This in turn means that less of the inherent organic matter is decomposed and therefore the ultimate result is more organic carbon and total nitrogen maintained in the soil compared to COA where the opposite is true (Xue et al., 2015; Małecka et al., 2012).Control treatments exhibited among the lowest values of carbon and nitrogen, proving that interventions with trees enhanced carbon and nitrogen availability, further stressing the importance of incorporation of trees into croplands in improvement of soil carbon and nitrogen among other soil physical and chemical properties (Nair et al., 2009).Owing to the fact that nitrogen is one of the major nutrient inhibiting plant growth in dryland ecosystems (Sainju et al., 2009), the high values obtained under CA in this study is sufficient proof of the capability of CA to enhance nitrogen in the dryland context. The less carbon obtained under COA can also be ascribed to soil disturbance through tillage which has been documented in literature to be the sole cause of persistent soil carbon losses (Baker et al., 2007; Reicosky, 2003) Sodium and potassium were also found to be high under CA compared COA and this can be attributed go the fact that the tillage layer of soil is less destroyed and also because there was residue (maize, legume and tree biomass) which was returned back to the soil during the experimentation period in CA and totally removed in COA (Tan et al. 2015). Reverting back biomass to the soil therefore could have led to the enrichment of the nutrients and thus explains the higher levels of N, K and Na under CA than COA. This is well agreed since CA yielded much more biomass in leaf and wood than COA in the study. The importance of agroforestry and maintenance of crop cover (thus CAWT) in building healthy fertile soils is therefore once again stressed in this study as it was in Garrity et al. (2010). 220 Figure 1: Comparison of soil moisture content between CA and COA in Machakos 0 5 10 15 20 25 30 35 40 SR 13 LR14 SR14 M o is tu re % v o l Season CA COA 221 TABLES Table I: Moisture retention during the SR13, LR14 and SR14 seasons in Machakos Treatment Mean treatment moisture retention % vol. Retention by season (% vol.) Retention by farming system (% vol.) SR 2013 LR 2014 SR 2014 CA COA calliandra at 1.5m 34.96 a 32.73abcd 35.73ab 36.42a 40.25a 29.67cd calliandra at 3m 28.52bcd 26.96cdef 28.96abcdef 29.64abcdef 31.21c 25.83ef calliandra at 4.5m 31.99 ab 28.6abcdef 34.6abc 32.78abcd 31.66c 32.63c Gliricidia at 1.5m 29.24bcd 27.68cdef 29.68abcdef 30.37abcdef 32.23c 26.65ef Gliricidia at 3m 32.67 ab 30.17abcdef 33.57abc 34.26abc 37.64ab 27.7def Gliricidia at 4.5m 29.43bc 27.8bcdef 29.9abcdef 30.58abcde 36.96b 21.9gh pigeon pea at 1.5m 26.6cd 24.38ef 27.38cdef 28.06bcdef 27.91de 27.87ef pigeon pea at 3m 26.54cd 25.48def 26.72cdef 27.41cdef 29.45ef 27.62def pigeon pea at 4.5m 24.17 d 22.61f 24.61ef 25.29def 27.87def 20.47h control 24.87 cd 22.97ef 24.97def 26.66cdef 24.75fg 24.99efg p <0.001 0.031 0.031 0.031 <0.001 <0.001 LSD 5.188 2.513 2.513 2.513 0.995 0.995 Means with same superscript letters are not different from each other,*significant at p<0.005. CA and COA denote conservation agriculture and conventional agriculture respectively. 222 Table II: Soil Chemical properties between conservation agriculture and conventional agriculture in Machakos for the period SR 2013 to SR 2014 Farming system Conservation Agriculture Conventional Agriculture t-test p values Treatment pH Ecd (dSm-1) CEC (cmolc/kg) BD (g/cm^3) pH Ecd (dSm-1) CEC (cmolc/kg) BD (g/cm^3) PH Ecd CEC BD Calliandra at 1.5m 6.957 0.057 12.854 0.956 6.934 0.062 14.031 0.914 0.99 0.99 0.82 0.98 Calliandra at 3.0m 7.080 0.054 13.970 0.902 7.188 0.047 14.841 0.943 0.98 0.98 0.87 0.98 Calliandra at 4.5m 13.576 4.715 9.111 1.307 7.130 0.056 14.382 0.917 0.15 *0.03 0.28 0.79 control 6.910 0.059 12.321 0.926 7.147 0.056 13.874 0.943 0.95 0.99 0.76 0.99 Gliricidia at 1.5m 6.936 0.062 13.585 0.927 7.111 0.114 14.617 0.903 0.96 0.9 0.84 0.99 Gliricidia at 3.0m 6.957 0.076 13.278 0.895 7.015 0.056 14.349 0.944 0.99 0.96 0.84 0.97 Gliricidia at 4.5m 7.006 0.052 13.978 0.940 7.005 0.060 13.605 0.961 0.99 0.98 0.94 0.99 Pigeon pea at 1.5m 7.051 0.049 14.007 0.969 7.124 0.058 13.785 0.877 0.98 0.98 0.97 0.95 Pigeon pea at 3m 6.913 0.057 12.225 0.944 6.994 0.054 15.040 0.921 0.82 0.99 0.59 0.99 Pigeon pea at 4.5m 6.968 0.061 12.862 0.927 6.973 0.061 15.348 0.882 0.99 ND 0.64 0.97 LSD 1.402 3.154 2.446 0.243 1.402 3.154 2.446 0.243 p 0.391 0.33 *0.004 0.18 0.391 0.33 *0.004 0.18 * Significant at p<=0.05,SD = Least significant difference, ND denotes no difference between means 223 Table III: Bulk density down the soil profile between conservation agriculture and conventional agriculture in Machakos over study period SR 2013 to SR 2014 Profile Depth Farming system t-test; p values CA COA 0-20cm 0.937 0.985 0.97 20-40cm 0.880 0.953 0.96 40-60cm 0.836 0.955 0.93 60-80cm 0.870 0.947 0.96 80-100cm 0.857 0.982 0.93 100-120cm 0.902 0.849 0.97 p 0.603NS 0.603NS SED 0.458 0.458 NS: Not significant at p<=0.05 224 Table IV: Exchangeable bases availability among treatments and between farming systems in Machakos over the study period SR 2013 to SR 2014 Farming system Conservation agriculture Conventional Agriculture Treatment m3.P (mg /kg) ExNa (cmolc/kg) ExCa (cmolc/kg) ExMg (cmolc/kg) ExK (cmolc/kg) m3.P (mg /kg) ExNa (cmolc/kg) ExCa (cmolc/kg) ExMg (cmolc/kg) ExK (cmolc/kg) Calliandra at 1.5m 26.811 0.115 7.094 2.501 1.904 25.828 0.113 7.925 2.624 1.942 Calliandra at 3m 22.842 0.150 7.817 2.633 1.977 20.840 0.171 9.093 2.912 1.772 Calliandra at 4.5m 13.417 2.752 6.205 2.440 5.147 26.653 0.142 8.310 2.564 2.088 control 27.494 0.116 6.661 2.398 1.881 23.307 0.130 8.473 2.957 1.720 Gliricidia at 1.5m 25.315 0.121 7.441 2.550 2.019 26.157 0.154 8.466 2.731 2.129 Gliricidia at3m 28.816 0.144 6.858 2.356 2.247 23.228 0.134 8.201 2.766 1.897 Gliricidia at 4.5m 25.636 0.141 7.946 2.532 1.897 27.017 0.118 7.869 2.685 1.894 Pigeon pea at 1.5m 25.137 0.151 8.393 2.697 1.874 22.810 0.147 7.931 2.541 1.957 Pigeon pea at 3m 26.103 0.120 6.702 2.283 1.822 24.333 0.142 8.639 2.793 1.992 Pigeon pea at 4.5m 22.237 0.109 7.036 2.631 1.965 25.062 0.138 8.803 2.983 2.028 Mean 24.467 0.399 *7.136 *2.483 2.282 24.651 0.138 *8.346 *2.750 1.948 p 0.86 0.341 <0.001 <0.001 0.338 0.86 0.341 <0.001 <0.001 0.338 LSD 2.728 0.555 0.584 0.146 0.715 2.728 0.555 0.584 0.146 0.715 * Significant at p <0.005 225 Table V: Composition of sand, silt and clay between conservation and conventional agriculture in Machakos for the period SR 2013 to SR 2014 Farming system CA COA t-test, p values Treatment Clay (% vol.) Silt (% vol.) Sand (% vol.) Clay (% vol.) Silt (% vol.) Sand (% vol.) Clay silt sand Calliandra 1.5m 75 19 17 75 20 17 ND 0.87 ND Calliandra 3.0m 77 19 15 78 18 15 0.93 0.87 ND Calliandra 4.5m 57 18 11 76 20 16 0.09 0.75 0.34 control 74 20 18 75 19 18 0.93 0.87 ND Gliricidia 1.5m 75 19 17 76 19 16 0.93 ND 0.86 Gliricidia 3.0m 75 20 16 76 19 17 0.93 0.87 0.86 Gliricidia 4.5m 75 20 17 75 20 18 ND ND 0.87 Pigeon peas 1.5m 76 19 17 77 19 16 0.94 ND 0.86 Pigeon peas 3.0m 74 20 18 76 19 16 0.87 0.87 0.73 Pigeon peas 4.5m 78 18 16 76 19 16 0.87 0.87 ND SED 0.281 0.633 2.251 0.281 0.633 2.251 ND denotes no difference between individual treatment means between farming systems 226 Table VI: Carbon and Nitrogen differences between conservation agriculture and conventional practice in Machakos for the period SR 2013 –SR 2014 Farming System Conservation Agriculture Conventional Agriculture Treatment TN (%weight) TC (%weight) ON (%weight) OC (%weight) TN (%weight) TC (%weight) ON (%weight) OC (%weight) Calliandra(1.5m) 0.12 1.64 0.11 1.49 0.12 1.77 0.11 1.62 Calliandra( 3m) 0.12 1.60 0.11 1.48 0.12 1.53 0.11 1.42 Calliandra(4.5m) 0.64 1.07 0.58 25.81 0.13 1.71 0.12 1.56 control 0.12 1.65 0.11 1.50 0.12 1.59 0.11 1.42 Gliricidia(1.5m) 0.12 1.74 0.11 1.58 0.13 1.77 0.12 1.87 Gliricidia(3m) 0.12 1.64 0.11 1.49 0.12 1.69 0.11 1.53 Gliricidia(4.5m) 0.12 1.66 0.11 1.51 0.11 1.65 0.11 1.48 Pigeon pea(1.5m) 0.11 1.59 0.11 1.46 0.11 1.51 0.11 1.36 Pigeon pea(3m) 0.11 1.60 0.11 1.45 0.13 1.75 0.11 1.59 Pigeon pea(4.5m) 0.11 1.48 0.10 1.31 0.13 1.85 0.12 1.70 Mean 0.17 1.57 0.16 3.99 0.12 1.69 0.11 1.56 p 0.377 0.07 0.376 0.342 0.377 0.07 0.376 0.342 SED 0.545 0.054 0.065 0.049 0.545 0.054 0.065 0.049 TN=Total Nitrogen, TC=Total Carbon, OC=Organic carbon, ON=Organic Nitrogen CONCLUSION Conclusively, sufficient evidence oozes to back the claim that CA both with and without trees significantly enhances soil moisture compared to conventional farming. CA also enhances amounts of available Nitrogen, Phosphorus and Potassium; and organic carbon. Conventional agriculture on the other hand significantly enhances cation exchange capacity and leads to increased amounts of salts in the soil alongside increase in availability of magnesium, calcium and exchangeable bases. Soil pH, texture and bulk density do not differ between CA and COA although conventional practice increases bulk density at the surface (0-20cm).CA thus surpasses COA in soil fertility enhancement and is thus a preferred option. ACKNOWLEDGMENTS The writing of this paper was enabled through the Evergreen Agriculture Project of World Agroforestry Centre, funded by the European Commission (EC) and International IFAD, the support of which is highly acknowledged. We thank also the management of Machakos Agricultural Training Centre for providing the land on which the experiment was hosted. REFERENCES Anderson, J. M. & Ingram, J.S.I. (Eds). (1993).Tropical soil biology and fertility: A handbook of Methods, Second Edition. Wallingford: C.A.B International. Baker, J.M., Ochsner, E.T., Venterea, R. T. & Griffis, T.J. (2007). Tillage and soil carbon sequestration-What do we really know? Agriculture, Ecosystems and Environment 118 (2007) 1–5. 227 Booker Tropical Soil Manual: A Handbook for Soil Survey and Agricultural Land Evaluation in the Tropics and Subtropics. In: Landon, J.R. (Ed.). (1991). Longman Scientific & Technical, Essex. Bryan, E., Deressa, T.T., Gbetibouo, G.A. & Ringler, C. (2009). Adaptation to climate change in Ethiopia and South Africa: options and constraints. Environ. Sci.Policy 12, 413–426. Busari, M.A., Kukal, S.S., Kaur, A., Bhatt, R. &Dulazi, A.A. (2015). Conservation tillage impacts on soil, crop and the environment. International Soil and Water Conservation Research 3 (2015)119–129 Corbeels, M., Jan de Graaff, Ndah,T.H., Penot, E., Baudron, F., Naudin, K., Andrieu, N., Chirat,G., Schuler, J., Nyagumbo, I., Rusinamhodzi, L., Traore, K., Mzoba, D.H., Adolwa,I.S.(2014). Understanding the impact and adoption of conservation agriculture inAfrica: A multi-scale analysis. Agriculture, Ecosystems and Environment 187 (2014) 155–170 Curtin, C. G., Sayre, N. F., & Lane, B. D. (2002). Transformations of the Chihuahuan Borderlands: grazing, fragmentation, and biodiversity conservation in desert grasslands. Environmental Science & Policy, 5(1), 55-68. Enfors, E. I. & Gordon, L. J. (2008). Dealing with drought: The challenge of using water system technologies to break dryland poverty traps. Global Environmental Change, 18(4), 607-616. Garland, G.M., Suddick, E., Burger, M., Horwath, W.R. & Six, J. (2011). Direct N2O emissions following transition from conventional till to no-till in a cover cropped Mediterranean vineyard (Vitis vinifera). Agriculture, Ecosystems and Environment 141 (2011) 234–239 Garrity D.P., Akinnifesi F. K., Ajayi O. C., Sileshi G. W., Mowo J. G., Kalinganire A., Larwanou M., Bayala J.et al.(2010). Evergreen Agriculture: a robust approach to sustainable food security in Africa. Springer: Food security 2:197– 214. Gill, K.S., P.R. Gajri, M.R. Chaudhary, and B. Singh. (1996). Tillage, mulch and irrigation effects on corn (Zea mays L.) in relation to evaporative demand. Soil and Tillage Research. 39:213–227 Govaerts, B., N. Verhulst, A. Castellanos-Navarrete, K.D. Sayre, J. Dixon, and L. Dendooven.(2009). Conservation agriculture and soil carbon sequestration: Between myth and farmer reality. Crit. Rev. Plant Science. 28:97–122. Guto, S.N., Pypers, P., Vanlauwe, B., de Ridder, N., and Giller, K.E. (2011). Socio- ecological Niches for Minimum Tillage and Crop-residue Retention in Continuous Maize Cropping Systems in Smallholder Farms of Central Kenya. Agronomy Journal 103:1–11 Jose, S. (2009). Agroforestry for ecosystem services and environmental benefits: an overview. Agroforestry systems, 76(1), 1-10. Jose, S. (2009).Agroforestry for ecosystem services and environmental benefits: an overview. Agroforest Systems (2009) 76:1–10 .DOI 10.1007/s10457-009-9229- 7 228 Jose, S., Gillespie, A.R., Pallardy, S.G. (2004). Interspecific interactions in temperate agroforestry. Agroforestry Systems 61:237–255 Ketema, H. & Yimer, F. (2014). Soil property variation under agroforestry based conservation tillage and maize based conventional tillage in Southern Ethiopia. Soil & Tillage Research 141 (2014) 25–31. Kursten, E. (2000). Fuelwood production in agroforestry systems for sustainable land use and CO2-mitigation. Ecological Engineering 16 (2000) S69 – S72 La1, R. (1997). Long-term tillage and maize monoculture effects on a tropical Alfisol in western Nigeria. II. Soil chemical properties. Soil and Tillage Research, 42, 161–174. Lee, K.H., Jose, S. (2003). Soil respiration and microbial biomass in a pecan–cotton alley cropping system in southern USA. Agroforestry Systems 58:45–54 Makumba, W., Akinnifesi F.K., Janssen, B., Oenema, O. (2007). Long-term impact of a Gliricidia-maize intercropping system on carbon sequestration in southern Malawi. Agriculture, Ecosystems & Environment 118:237–243. Małecka, I. , Blecharczyk, A., Sawinska, Z.& Dobrzeniecki, T.(2012). The effect of various long-term tillage systems on soil properties and spring barley yield. Turk J. Agric. For. 36 (2012) 217-226. Doi: 10.3906/tar-1104-20. Mehlich, A. (1984). Mehlich 3 soil test extractant: A modification of Mehlich 2 extractant. Commun. Soil Sci. Plant Anal.15: 1409-1416. Mloza-Banda, H.R, Makwiza, C.N. & Mloza-Banda, M.L. (2016). Soil properties after conversion to conservation agriculture from ridge tillage in Southern Malawi. Journal of Arid Environments 127 (2016) 7-16. Nair ,R.P.K.,Vimala D. N., Kumar , B. M. & Haile , S.G. (2009). Soil carbon sequestration in tropical agroforestry systems: a feasibility appraisal. Environmental Science & Policy 12 (2009) 1099 – 1111 Palm, C., Blanco-Canqui, H., DeClerk, F., Gatere, L. & Grace, P. (2014). Conservation agriculture and ecosystem services: an overview. Agriculture, Ecosystems and Environment 187(2014)87–105. Parr, J. F., Stewart, B. A., Hornick, S. B., & Singh, R. P. (1990). Improving the sustainability of dryland farming systems: a global perspective. In Advances in soil science (pp. 1-8). Springer New York. Rahman, M.H., Okubo, A., Sugiyama, S.,& Mayland, H.F.(2008).Physical, chemical and microbiological properties of an Andisol as related to land use and tillage practice. Soil and Tillage Research, 101, 10–19. Rasmussen,K.J.(1999).Impact of ploughless soil tillage on yield and soil quality: A Scandinavian review. Soil and Tillage Research, 53, 3–14. Reicosky, D.C. (2003). Tillage-induced CO2 emissions and carbon sequestration: effect of secondary tillage and compaction. In: Garcia-Torres, L., Benites, J., Martinez-Vilela, A., Holgado-Cabrera, A. (Eds.), Conservation Agriculture. Kluwer Acad. Pub., Dordrecht, The Netherlands, pp. 291–300. 229 Reynolds, J. F., Smith, D. M. S., Lambin, E. F., Turner, B. L., Mortimore, M., Batterbury, S. P. ... & Huber-Sannwald, E. (2007). Global desertification: building a science for dryland development. science, 316(5826), 847-851. Rockstrom, J., P. Kaumbutho, J. Mwalley, A.W., Nzabi, M. Temesgen, L. Mawenya, J. Barron, J. Mutua, and S. Damgaard-Larsen. (2009). Conservation farming strategies in East and Southern Africa: Yields and rain water productivity from on-farm action research. Soil and Tillage Research. 103:23–32. Schroth, G. & Sinclair, F. (2003) Trees crops and soil fertility: concepts and research methods. CABI, Wallingford, UK, p 464 Schwab, N., Schickhoff, U. & Fischer, E. (2015). Transition to agroforestry significantly improves soil quality: A case study in the central mid-hills of Nepal. Agriculture, Ecosystems and Environment 205 (2015) 57–69. Seobi, T., Anderson, S. H., Udawatta, R. P., & Gantzer, C. J. (2005). Influence of grass and agroforestry buffer strips on soil hydraulic properties for an Albaqualf. Soil Science Society of America Journal, 69(3), 893-901. Tan, C., Xue Cao, Shuai Yuan, Weiyu Wang, Yongzhong Feng & Bo Qiao.(2015). Effects of Long-term Conservation Tillage on Soil Nutrients in Sloping Fields in Regions Characterized by Water and Wind Erosion. Sci. Rep. 5, 17592; doi: 10.1038/srep17592 (2015). Thierfelder, C. & Wall, P. C. (2009). Effects of conservation agriculture techniques on infiltration and soil water content in Zambia and Zimbabwe. Soil and tillage research, 105(2), 217-227 Thierfelder, C., Mwila, M. and Rusinamhodzi, L. (2013).Conservation agriculture in eastern and southern provinces of Zambia: Long term effects on soil quality and maize productivity.Soil and Tillage Research 126(2013)246-258 Tornquist, G. C., Hons, F.M., Feagley, E.S. & Haggar, J. (1999). Agroforestry system effects on soil characteristics of the SarapiquõÂ region of Costa Rica. Agriculture, Ecosystems and Environment 73 (1999) 19-28 Udawatta, R.P, Gantzer, C.J., Anderson, S.H., & Garrett, H.E. (2008) Agroforestry and grass buffer effects on pore characteristics measured by high-resolution X-ray computed tomography. Soil Sci Soc Am J 72:295–304 Xue, Jian-Fu, Chao Pu, Sheng-Li Liu, Zhong-Du Chen, Fu Chen, Xiao-Ping Xiao et al. (2015). Effects of tillage systems on soil organic carbon and total nitrogen in a double paddy cropping system in Southern China. Soil & Tillage Research 153 (2015) 161–168 Yimer, F., Messing, I., Ledin, S. & Abdu, A. (2008). Effects of different land use types on infiltration capacity in a catchment in the highlands of Ethiopia. Soil Use Management 24, 344–349. Zika, M., & Erb, K. H. (2009). The global loss of net primary production resulting from human-induced soil degradation in drylands. Ecological Economics, 69(2), 310-318. 230 5.5 Effects of selected plant leaf extracts on soil ph and betacarotene levels in amaranthus species grown under varying conditions in Kiambu county Njeru, N1,*, Mukiri wa Githendu1 and J. Murungi2 1Department of Agriculture, Science and Technology, Kenyatta University 2 Department of Chemistry, Kenyatta University P. O. Box 43844 - 00100, Nairobi, Kenya *Corresponding author E-mail: charitynnjeru@yahoo.com ABSTRACT Amaranthus (Amaranthus spp.) leaves are a good source of vitamins and source of dietary minerals including calcium, iron, and potassium. Soil pH is a very important chemical property of the soil, as it dictates the availability of plant nutrients. Acidic soils reduce availability of plant nutrients and hinder growth. This is corrected by adding lime which is expensive, does not add nutrients and requires re-application. The objective of this study was to evaluate the effects of leaf extracts from selected tree leaves on soil acidity and beta-carotene levels of amaranthus. Plant leaves that were tested were Turril (Vitex keniensis) Mexican sunflower (Tithonia diversifolia) and Indian nettle (Plectranthus barbatus).The experiment was setup in 2 x 7 factorial arrangement in a randomized complete block design (RCBD) replicated three times. Leaves were dried and soaked in water and extract was added to soil for amanthus grown with plant extracts. Other amaranthus were gown in soils treated with either fertilizer or lime.or manure. Nothing was added. to control soil Beta carotene was determined using UV/Visible instrument. Data was subjected to ANOVA at 5% level of significance using SAS while least significant difference (LSD)used for means separation. Leaf extracts were found to increase betacarotene level of amaranthus to 52.36mg under greenhouse conditions , while those grown with inorganic fertilizer were 21.23mg under irrigation. Soil pH increased from 5.9 to 6.7 for leaf extract treatment and with inorganic fertilizer decreased from 5.9 to 5.3.The outcome, of the research will be used in recommendation of economical organic fertilizer to improve soil pH and increase beta-carotene levels of amaranthus which will benefit small scale farmers who can not afford inorganic fertilizers. Keywords: Key words: Plant extract, Beta-carotene , Soil pH INTRODUCTION There has been a rising demands of African Leafy Vegetables (ALVs) in the recent past in Kenya. The priority species marketed include leafy amaranth (Amaranthus spp.), African Black nightshade species (Solanum spp.), cowpeas (Vigna unguiculata), Ethiopian kale (Brassica carinata), kahuhura (Cucurbita ficifolia), jute plant (Corchorus olitorius) and pumpkin leaves (Cucurbita maxima and C. moschata), (Irungu,et al, 2011) African Leafy Vegetables (ALVs) have gained commercial importance over the past 15 years as a result of the enormous growth in market (Irungu et al, 2007). ALVs 231 production has its advantages because of the uniqueness of such as short production cycles thrives in poor soil, are resistant to pests and diseases, require a few purchased inputs, and are quite acceptable to local tastes (Ekesa,et al, 2009) Continuous cultivation of diminishing farms to feed the growing population has resulted in soil degradation and consequently rise in use of inorganic fertilizers to increase crop yield. Increasing and continuous use of inorganic fertilizers lead to lowered soil pH, poor soil texture (Glinsiki and Witold, 1985), poor quality crop yields and finally pollution of the rivers and lakes resulting in eutrophication. The inorganic fertilizers are expensive and hence out of rich to most rural farmers, majority of whom are women. In addition they are not always available, especially the subsidized ones which is being blamed for late planting and thus poor yields. Simulation studies of plant leaves using leaf extracts showed increase in soil pH and crop biomass possibly due to availability of nutrients (Mwangi, 2012; Lawrence, 1990) Also certain plants have litter that can decompose easily releasing plant nutrient and increasing buffering capacity of soil (Murungi, 1990; Njagi 2008; Njagi et al., 2012; Kikuchi, 2004, Mwangi, 2012). However, the study did not look at soil texture or actual crop yield and was under controlled environment. In addition, only a few non-leguminous plants were studied using only a mixture of leaves and leaf extracts. Therefore, there is need to optimize productivity using natural environment and individual as well as combined leaves and leaf extracts to optimize plant nutrients and soil pH. The present study used leaves extract of plants shown to have high buffering capacity and high rate of mineralisation individually for raising pH of the soil as well as adding nutrients to the soil and improving soil texture. Various leaves extracts were tested for their effect on soil pH and nutrients levels. They were used to improve the acidic soils where the amaranthus was grown on improved soils and their growth rate, biomass and yield monitored. Soil pH was also monitored at intervals. Carotenoids, have been credited with other beneficial effects on human health enhancement of the immune response and reduction of the risk of degerarative diseases such as cancer, cardiovascular diseases,cataract, and macular degeration (Astrog 1997, Bendich 1994, Burri 1997, Gaziano and Hennekens 1993, Krinsky 1993, Mayne 1996, Olson 1999a, Olson and Krinsky 1995). The action of carotenoids against diseases has been attributed to antioxidant property, specifically ,their ability to quench singlet oxygen and interact with free radicals(Palozza and Krinsky 1992). Therefore it was selected for analysis It is hoped that the results of this study would be used to improve soil, increase vegetable production, with high levels of betacarotene and alleviate poverty amongst small-scale farmers,as well as encouraging planting of more plants and trees thus increase plant cover especially in Kenya. General Objective To determine the effect of selected leaves extracts on soil pH, and betacarotene levels of amaranthus species. 232 Specific Objectives (i) To determine the effect of Vitex keniensis ,tithonia diversifolia and plectranthus barbatus leaves extracts on pH of acidic soil. (ii).To determine levels of betacarotene in amaranthus grown after treatment with leaves extracts of Vitex keniensis ,tithonia diversifolia and plectranthus barbatus ,inorganic fertilizer, lime and with no treatment. MATERIALS AND METHODS Study Area and Experiment Procedure The leaves of the target species were collected from Juja Ward in Kiambu County, where also the experiment was carried out in the field. This was done under three regimes; under irrigation, on rainy season and in a greenhouse at Kenyatta University field demonstration site. Analysis for soil pH, were done at laboratories in Chemistry Department at Kenyatta University, while for betacarotene analysis was done at JKUAT, Laboratories Food Science Department. Plant leaves tested were Turrill (Vitex keniensis), Mexican sunflower (Tithonia diversifolia) and Indian nettle (Plectranthus barbatus).The concentration of the leaves extract was 500g diluted with two litres of distilled water. The leaves were washed, sun dried subjected to milling/hand crushing, weighing packing and labeling. For leave extracts, both milled and hand crushed leaves was soaked in distilled water for 30 days and pH and buffering capacity monitored for individual plant leaves. Optimization of the leaf extracts for soil remediation and growth of the crops was done for open fields. For pH and buffering capacity controls, distilled water was kept under the same conditions and monitored for pH and buffering capacity (Miller and Kissel, 2010). Study Design and Treatment Application The experiment setup was in a randomized complete block design (RCBD), in a two by seven factorial. The treatments were the three leaves extracts, lime, Farm Yard Manure inorganic fertilizer and the control. All the seven treatments were replicated three times.Individual plots measured 2 m by 1.2 m. Individual blocks were spaced1 m apart while the plots within the blocks separated by 0.5m. The seeds were planted directly and plots prepared to a fine tilth at a spacing of 30cm x 10cm. Experimental plots were 42 each having 45 plants. The plants were subjected to treatments after two weeks of germination in the plots. This was applied through drenching in the soil at the rate of 20 ml per plant, Farm yard manure applied once during planting at the rate of 200g per hole, Diammonium phosphate and Lime were applied once at planting at a recommended rate of 5g per plant and 130g respectively. 233 Management of the Experiment Land preparation was done one month before sowing by clearing the weeds followed by ploughing. The field was harrowed to create a suitable tilth. Seeds were planted at a depth of 2 mm in the raised plots. Routine field maintenance practices such as weeding, watering were done and also spraying against pests and fungal diseases when necessary using recommended pesticides and fungicides respectively. Data Collection The data were collected per individual plot for the 8 sampled plants per plot and were dried. This was done after 10 weeks from planting. Sample preparation and analysis Effect of Leaves Extract on Soil pH Soil samples for pH were taken from the experiment site and using 1:2 ratio, Soil to Water, from a depth of 0-20 and 20-40cm. The samples were air-dried and ground to pass through a 2mm sieve and analyzed for pH according to (Miller and Kissel, 2010). Effects of Leaves Extracts on Betacorotene Levels The amaranthus leaves were harvested at intervals of one week and betacarotene was extracted.Determination of Beta-carotene was by carotene equivalent using acetone as solvent, by AOAC method (AOAC, 1995). The concentration of carotene was read directly from UV visible spectrophotometer at 440nm after proper calibration of the instrument with standard solutions of pure beta-carotene (Sigma chemical Co., St. Louis, and Mo). Statistical Analysis All the plant biochemical data were subjected to analysis of variance (ANOVA) using the General Linear Model. Proc GLM code of SAS- computer software (SAS 2002; Version 16.0) where significant, mean separation were done with LSD. Results and Discussion Effects on the soil pH The pH changes with various treatment are given in Table 1. the pH under green house regiment ranged from--5.604 to 6.233 with inorganic soil having the lowest pH , while under irrigation regime the pH ranged from 5.53 to 6.172with inorganic soil having the lowest and Plectranthuus barbatus the highest. Rainy regime the pH ranged from 5.5 to 6.1 with lime having the highest and inorganic soil the least. Significant differences (P≤0.05) were observed between the treatments under all the three regimes(Table i) There was no significant difference in the two varieties. Analysis of variance for soil pH indicated significant (p<0.05) differences among the treatments with leaf extract plectranthus barbatus having highest effect (5.9 to 6.7) for the two varieties. However the inorganic fertilizer had the lowest effect. (5.9 to 5.3) as shown in (Figure 1). It is important to note the pH of control and inorganic remained 234 below 6 in the three regimes while that of the the leaf extract remained above six in the three regime and Plectrathus barbatus and Tithonia increased betacorotene levels substacially in amaranthus compared to inorganic fertilizers. Effect on betacarotene From Table ii and figure 2 Under rainy season, FYM had the highest effect (44.87) while control (21.77) had the lowest. On irrigation, leaf extract 3 had the highest effect (36.87) and control had the lowest effect (18.77). On greenhouse, FYM and leaf extract 3 had the highest effect (52.36 and 47.32) and control (27.32) had the lowest. (Figure 2) It is important to note that under various regimes , inorganic fertilizer had the lowest beta carotene compared to leaf extract and lime. Its levels were only higher than that of the control. Tables And Figures Table i; Effect of Treatments on the Soil pH under the Three Regimes Treatment Greenhouse Irrigation Rainy CONTROL 5.962e 5.947c 5.956d FARM YARD MANURE 6.031d 5.964c 5.973cd INORGANIC FERTILIZER 5.604f 5.53d 5.555e LEAF EXTRACT 1- TITHONIA 6.068cd 6.023b 6.018bc LEAF EXTRACT 2-VITEX KENIESIS 6.096c 6.028b 6.046b LEAF EXTRACT 3- PLECTRAS BARBATUS 6.233a 6.172a 6.148a LIME 6.183b 6.153a 6.153a LSD 0.03894 0.02864 0.04953 CV(%) 0.80 0.60 1.00 Treatments with different letters in the same column are significantly different at 5% probability Table ii:ANOVA on the Effect of Treatment on Betacarotene on Variety Under Combined Regime 235 Table ii Source of variation d.f. s.s. m.s. v.r. F pr. Rep stratum 2 3.13 1.56 1.77 Regime 2 1413.4 1 706.71 799.5 <.001signf . Variety 1 2.26 2.26 2.56 0.114 Treatment 6 7315.9 9 1219.33 1379. 4 <.001signf . Regime.Variety 2 0.44 0.22 0.25 0.778 Regime.Treatment 12 162.56 13.55 15.33 <.001signf . Variety.Treatment 6 6.64 1.11 1.25 0.289 Regime.Variety.Treatment 12 12.05 1.00 1.14 0.343 Residual 82 72.48 0.88 Total 125 8988.9 7 Table: iii Effect of Treatment on Betacarotene on the Regimes. Treatment Greenhouse Irrigation Rainy CONTROL 26.67l 19.22o 21.89n FARM YARD MANURE 51.51a 38.36e 45.02o INORGANIC FERTILIZER 30.47j 23.64m 24.47m LEAF EXTRACT 1-TITHONIA 41.82d 33.49h 37.82ef LEAF EXTRACT 2-VITEX 37.17f 31.67i 34.5h LEAF EXTRACT 3-PLECTRAS BARBATUS 46.14b 35.98g 41.98d LIME 32i 26.13l 28.13k LSD 1.08 CV(%) 2.8 Treatments with different letters in the same column are significantly different at 5% probability 236 Figure 1: Effect of the Treatments on the Soil pH Figure 2: Comparison of Betacarotene Levels among Regimes and Treatments CONCLUSIONS AND RECOMMENDATIONS Significant differences (P≤0.05) were observed between the treatments under all the three regimes. This was important as buffering of the soil, in making the soil pH rise and in maintaining it to enable the growth of the amaranthus. This was observed in the greenhouse unlike in the rainy season as some of the soils were leached. Leaf extract3, had the highest effect in the three regimes and control the lowest effect. (Table i) There was no significant difference in the two varieties. 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 So il p H Treatment After Before 0 10 20 30 40 50 60 B et ac ar o te n e Treatment Greenhouse Irrigation Rainy 237 The soil pH indicated significant (p<0.05) differences among the treatments with leaf extract 3, Plectranthus barbatus highest effect, under irrigation. The inorganic fertilizer decreased the soil pH.FYM and Plectranthus barbatus had the highest effect on betacarotene levels.(Table ii & iii) More research is needed to evaluate the effect of the leaf extracts over a longer period of time on the soil pH, and also growing other different vegetables especially those with betacarotene to check their levels. ACKNOWLEDGEMENT I present my fervent thanks and appreciation to Forum for Enhancing Sustainable Agricultural Production and Market Systems through Science Technology and Innovation for a chance to present. I thank the Vice Chancellor of Kenyatta University for providing the funding for this project.I am grateful to Dr. Mukiri Wa Githendu and Professor Jane Murungi who are my supervisors and Professor Ruth Wanjau, Chemistry Department and Dr. Mugo who are also in this project for financial support in publishing this work. I acknowledge Dr. Catherine Muui and Dr. Wilson Thagana, AST Departments for their scientific guidance and invaluable support during the research. REFERENCES Astrog 1997, Bendich 1994, Burri 1997, Gaziano and Hennekens 1993, Krinsky 1993, Mayne 1996, Olson 1999a, Olson and Krinsky 1995). Agboola, A.A. and Sobulo, R.A. (1981). A review of soil fertility in southwestern zone, Nigeria. FDALR. Kaduna, Report 6: 1-15. Akinfasoye (2009). Effect of organic Fertilizer and cutting height on growth, shoot yield and nutrient uptakes of amaranth (Amaranthus cruentus). Journal of Life and Physical Sciences, Acta SATECH., Nigeria 3 (1): 714-721. Andrasofszky, E., Szöcz, Z., Fekete, S., Jelenits, K. (1998). Evaluation of the nutritional value of the amaranth plant. I. Raw and heat-treated grain tested in experiments on growing rats. Acta Vet. Hung. 46: 47-59. ASHS Press, Alexandria, VA. Babatola, L.A and Olaniyi, J.O. ( 1999). Performance of Amaranthus caudatus at different levels of NPK fertilization and under different storage methods. Proceedings of the 16 th Annual Conference of Horticultural Society of Nigeria, UNAAB, Abeokuta, 7th -10th 41 Bationo, A., Hartermink, A., Lungu, O., Naimi, M., Okoth, Pl, Samling, E. M. A. and Thiombiano, L. (2006).African Soils: Their productivity and profitability of fertilizer use. Background paper presented, for the African Fertilizer Summit, 9th – 13th June, 2006, Abuja, Nigeria. Bulaev, V. E. (ed) (1984). Methods of Fertilizer Application, Oxonian Press PVT. Ltd. New Delhi. Pp 52. Carver, B. E. and Ownby, J. D. (1995). Acid soil tolerance in wheat. Advanced Agronomy. 54, 117-173. 238 Crawford, T. W. Jr., Singh and Breman, H. (2008). Solving Agricultural Problems Related to Soil Acidity in Central Africa’s Great Lakes Region Determination of Beta-carotene was by carotene equivalent using acetone as solvent, by AOAC method (AOAC, 1995) Gachengo, G.N., Palm, C.A., Jama, B. and Otieno, C. (1999). Tithonia and Sena Green Manure and Inorganic Fertilizers as Phosphorous Source for maize in Western Kenya. Agrof Systs 44: 21-26 Glinsiki, J.J. and Witold, S. (1985). Soil Aeration and Its Role for Plants, CPC, Press Inc., Florida; Pp 110-112. Influence of Biofertilizer-Fortified Organic and Inorganic Faculty of Agricultural Sciences, LAUTECH, Ogbomoso, Nigeria, 2012 Kikuchi, R. (2004). De-acidification effect of the litter layer on forest soil during snowmelt runoff: laboratory experiment and its basic formularization for simulation modeling. Chemosphere 54:1163–1169 Kimetu, J.M., Mugendi, D.N., Palm, C.A., Mutuo, P.K., Gachengo, C.N., Bationo, A., Nandwa, S. and Kungu, J.B. (2004).Nitrogen fertilizer equivalencies of Njagi, S.E., Murungi, J.I., and Wanjau, R.N. (2012).The pH of leaf water extracts and amount of acid required for lowering the pH of leaf water extracts to 5.0.American International Journal of contemporary research; 2 (11).Pp 72-78 Miller, J. C. and Miller, J. N. (1988). Statistics for Analytical Chemistry (2nd ed). Ellis Harwood Ltd., Chichester. Pp 33-77; 109-115. Miller, R.O. and D.E. Kissel. 2010. Comparison of soil pH methods on soils of North America. Soil Science Society of America Journal 74 Murungi, J. I. (1990). Effects of Acid Rain and Humic Substances on Aluminium Toxicity. A Comparative Study of Chemical Composition of Old and Newly Deposited Plague from Heart Patients, Louisiana State University. PhD Thesis. Pp 180-206. Mwangi J. N. (2012). Mineral levels of some plants and their extracts and the effect on soil pH and growth rate of selecte non-legunious plants. M.Sc Thesis, Department of Chemistry, Kenyatta University, Nairobi, and Kenya. Njagi, S. E. (2008). Buffering Capacity and Levels of Macronutrients Found in Leaves of Selected Plants from High lands East of Mount Kenya, M.Sc Thesis, Department of Chemistry, Kenyatta University, Nairobi, and Kenya. Pp 64. Njagi, S. E., Murungi, J. I., and Wanjau, R. N. (2012).The pH of leaf water extracts and amount of acid required for lowering the pH of leaf water extracts to 5.0. American International Journal of contemporary research; 2 (11).Pp 72-78 239 6. Parallel Session 5: Agricultural Innovations, Intensifications and Risk Management 6.1 Farmer knowledge and utilization of rainwater harvesting and conservation technologies in a semi-arid agro-ecozone of kenya F.W. Ng’ang’a 1, M. Mucheru-Muna 2 F.Waswa2, F.S Mairura1 1 Department of Environmental Science and Technology, South Eastern Kenya University P.O Box 70-90100, Kitui, Kenya 2 Department of Environmental Science, Department of Agricultural Resource Management, Kenyatta University, P.O. Box 43844- 00100,Nairobi, Kenya Corresponding author:fmuriu@seku.ac.ke ABSTRACT Food insecurity is a major problem facing farming communities in the semi-arid zones of Tharaka Nithi County in the Eastern parts of Kenya. Recurrent drought, inter- seasonal dry spells and soil fertility challenges are key causes of food insecurity, particularly within the smallholder farming sector in Kenyas’ semi-arid lands. Rain water harvesting and technologies are important strategies to enhance water use and food productivity, however factors affecting utilisation of these technologies by farmers have not been well understood. Therefore this study was conducted in Tharaka South sub-county to establish the relationship between knowledge levels of rain water harvesting and conservation technologies by farmers and technology utilization. Farm and social data was collected using interviews from a total of 353 household heads. Farmers knowledge scores for less knowledge intensive technologies was higher compared to more knowledge intensive technologies, indicating the need for more focused training to enhance knowledge of all rain water harvesting and conservation technologies. The study concluded that farmer socio-economic characteristics were important factors which should be considered in future approaches to develop and disseminate rain water harvesting and conservation technologies in Tharaka sub- county. This study fills an important gap by providing recommendations on increasing productivity and sustainable management of soil and water resources for improved food security by extension officers, government policy makers and smallholder farmers in the semi-arid lands. Key words: Soil and water conservation, climate change, agriculture, green water stress, local knowledge INTRODUCTION Rain water harvesting and conservation technologies are important technologies necessary to enhance agricultural production in the semi-arid agro-ecosystems of Kenya which depend on rain-fed agriculture. Several bio-physical studies in Africa have shown the potential of rain water harvesting and conservation technologies to enhance 240 agricultural productivity. The potential of RWH in providing supplementary water, increase crop yield and reduce the risk of crop failure is high (Critchley and Siegert, 1991). In Kenya, studies and development work around rain water harvesting and conservation technologies have been undertaken for more than 50 years in semi-arid agroecological zones (Tiffen et al., 1994), however these technologies have not been widely adopted by farmers. Rain water harvesting techniques can be classified into two major types depending on source of water collected; namely, the in situ and the ex situ types of rainwater harvesting, respectively. In essence, in-situ rainwater harvesting technologies are soil management strategies that enhance rainfall infiltration and reduce surface runoff. Different rain-water harvesting and conservation technologies can also be classified depending on their intensive levels of knowledge or resource requirements during their installation and maintenance. Lack of knowledge and awareness coupled with several socio-economic characteristics is a key factor which can explain low adoption rates of rain water harvesting and conservation technologies. It is important to establish causes of low adoption in light of challenges such as food insecurity, water scarcity and limited fertile land (Wildemeersch et al., 2013). Critchley (2009) discussed the importance of agricultural extension and the role of farmer knowledge and farmer-to-farmer extension in rain water harvesting adoption, underlining the important role of capacity building in uptake. Hassan and Nhemachena (2008) show that in Africa, 50% of farmers believe that temperatures are increasing and precipitation is becoming less, but only 20% of the respondents believe that soil and water conservation could help them adapt, showing the existence of knowledge gaps regarding rain water harvesting and conservation technologies among farmers. Various water harvesting and conservation technologies have been successfully tested and popularized by governmental and non-governmental agencies (Pachpute, 2010). Few of these efforts have succeeded in combining technical efficiency with social acceptability to local farmers (Pachpute, 2010). Consequently, rainwater harvesting has not been practiced to the recommended levels hence the continued crop failure leading to food insecurity (Kahinda & Taigbenu, 2011). It is of paramount importance to learn about the available rainwater harvesting and conservation technologies, and the socio- economic characteristics influencing their use and non-utilisation. Several studies on agricultural innovations including rain water harvesting technologies have shown that farmers in Sub-Saharan Africa are knowledgeable about their ecological agro-ecological settings, but the knowledge on specific soil water conservation technologies is variable which influences adoption and utilization of rain water harvesting and conservation technologies. There is lack of scientific knowledge on the adoption, utilisation behaviour, and farmer knowledge and perception of recently introduced rain water harvesting and conservation technologies in Kenya hence the gist of the current study. Most studies on rain water harvesting and conservation technologies have been based 241 on the biophysical aspects of the technologies with a minimal focus on farmer perspectives, knowledge and experience with these technologies. Helmreich and Horn (2009) indicated that local skills and resources are key inputs for rain water harvesting and conservation technology development. The aim of the study was to examine farmer knowledge and experience with various rain-water harvesting and conservation technologies using scientific scaling criteria developed to verify farmer knowledge systems. The study also aims to investigate the relationships between knowledge of different types of rain-water harvesting and conservation technologies to determine patterns in rain-water harvesting knowledge and farmer typologies. The study proposes that farmer knowledge and experience with the technologies is an important entry-point for the development and sustainable adoption and utilisation of rain-water harvesting and conservation technologies in low-input, low-output water limiting agricultural systems in semi-arid zones of Eastern Kenya. MATERIALS AND METHODS Study Area The study was carried out in Tharaka South Sub-County in Tharaka Constituency, Tharaka Nithi County. The constituency covers an area of 1,569.5 km2 with 175,905 people as per the 2009 population census of Kenya (Republic of Kenya, 2010). Tunyai and Nkarini locations were sampled for the survey. Tharaka South Sub-county has a wide representation of agro-ecological zones (AEZs). These are Lower Midland 4 (LM 4), Lower Midland 5 (LM 5), Intermediate Lowland Zone 5( IL 5) and Intermediate Lowland Zone 6 (IL 6) (Smucker & Wisner, 2008). The study area was mainly in the Lower Midland 4 (LM 4) and the Lower Midland 5 (LM 5) (Icheria, 2015) zones. Figure 1 represents the study area. 242 Figure 1: Map of the study area Tharaka South Sub County is located on the Eastern side of Mount Kenya, and experiences a bi-modal rainfall pattern: the long rains are experienced in the months of March and May each year, while the short rains occur between October and December. A large segment of the population in Eastern Kenya depends on the short rain season which is considered to be more reliable and can be predicted with a reasonable degree of accuracy (Recha et al., 2012). Rainfall is unevenly distributed within these rainy seasons and shows significant variability from year-to-year and season-to-season (Cooper et al., 2008). The total bimodal rainfall averages between 200mm and 800mm annually with its temperature ranging from 110 C to 25.90 C during cold and hot seasons respectively (Jaetzold et al., 2006). The soils in Tharaka South Sub-county are predominantly sandy loam and shallow hence the March- May rainfall which is little may not buffer crops from agricultural drought (Recha et al., 2012). Crops grown in the area include Sorghum (Sorghum bicolor), pearl millet (Pennisetum glaucum), cowpea (Vigna unguiculata) and green grams (Vigna radiata). These four crops are adapted to drought and form the basis of Tharaka cropping systems. Maize (Zea mays) and pigeon pea (Cajanus cajan), pumpkins (Cucurbita sp.), cotton (Gossypium sp.), beans (Phaseolus sp.), cassava (Manihot esculenta), and dolichos (Dolichos sp.) are also grown in the study area (Labeyrie et al., 2014). Survey sampling procedure The study adapted a survey design, where surveys and interviews were used to gather data from farmers and informants. Using information from the local administration and Ministry of Agriculture extension staff and local leaders, a list of household heads 243 was generated. Sample size was calculated using formulae 1 (CRS, 2007) resulting in a sample size of 351 farmers. 2 2 )1(*)(* c ppZ S   Where: S=Sample size, Z=Z value (e.g. 1.96 for 95% confidence level), p=percentage of picking a choice, expressed as decimal (0.5), c=confidence interval, expressed as a decimal (0.5±5%). Two locations, Nkarini and Tunyai were sampled. The households were clustered into sub-locations. The number of households to be interviewed per sub location in each location was identified through proportionate sampling method after the total sample size was determined. The households were arrived at through systematic random sampling where every 10th household head was interviewed. Data collection and analysis Semi-structured questionnaires were used to collect data from sampled farm house- holds. Frequencies and descriptive statistics were used to summarise the data, while cross-tabulations were used to display relationships in the data. Scores for each rainwater harvesting and conservation technology were added to calculate knowledge scores for each farmer interviewed after which they were used for data analysis (Nampanya et al., 2010). The scoring was implemented using a marking scheme on a 3-point scale, with 3 marks for correctly answered questions, 2 (partly correct), and 1 mark (wrongly answered). Different key questions were asked for the different technologies, meant to gauge the basic knowledge for the respective technologies in relation to physical structures of the technologies and their application to agriculture and soil/water conservation. Comparison of mean score was carried out for each rainwater harvesting and conservation technology between farmers utilizing the technology and those who were not utilizing the technology using analysis of variance (ANOVA). Correlation co-efficients were used to relate household characteristics and knowledge mean scores. RESULTS AND DISCUSSIONS Farmer socio-demographic characteristics Table 1 describes the socio-economic characteristics of farmers in Tharaka South Sub- county. Male-headed households consisted of 80% of the sample, while female-headed households comprised 20% of the total sample size of 353 households. The age distribution showed that household heads in the 21-36 year age-cohort were 27%, while those in the 37-52 groups were 36% of the total sample. 244 Farmers who were over 69 years consisted of only 7.4% (Table 1). Table 1: Socio-economic characteristics of the household heads in Tharaka South Sub-county Variable Categories Frequency Percent Gender of household head Male 283 80.2 Female 70 19.8 Age household head 21-36 91 27.1 37-52 120 35.7 53-68 100 29.8 Over 69 25 7.4 Marital status of household head Single 57 16.2 Married 295 83.8 Education level of household head None 99 28.0 Primary 161 45.6 Secondary 69 19.5 College 18 5.1 University 6 1.7 Main occupation of household head Full time farmer 261 74.1 Casual laborer 15 4.3 Business 52 14.8 Formal employment 24 6.8 Group membership status Non-group members 253 71.7 Group members 100 28.3 Labour source Family labour 233 66.0 Hired labour 120 34.0 Title deed ownership Holds no title deed 161 45.6 Holds title deed 192 54.4 Credit access With no access to credit 242 68.6 With access to credit 111 31.4 Married households’ heads consisted 84%, while up to 72% of the household heads had achieved education beyond primary school (Table 1). Most of the household heads were full time farmers (74%), while 72% of the farmers did not belong to a farmers group. Most of the farm labour was family based (66%), while hired labour was used by 34% of the farms (Table 4.1). Up to 54% of the farmers held a title deed, while only 31% of the farmers had access to credit (Table 1). The study indicates that most of the farmers were younger than 52 years. This age set is important in manpower and experience provision for labour and knowledge- intensive technologies (physical soil structures) such as Fanya juu, Negarim pits and Zai pits which may demand substantial installation and maintenance labour input especially when this labour is household derived. Most of the farmers were educated to the secondary school level, which is beneficial for rain-water harvesting technology 245 development and utilisation especially for knowledge intensive technologies, including the recently introduced technologies which may not be well understood in the farming system. Ample farm households were members in groups (28%) indicating the potential to disseminate and utilise technologies because these technologies can be more efficiently disseminated in group frameworks with potential synergistic social benefits especially if group dynamics are effectively managed. Given that most of the respondents were full-time farmers, with title deeds with formal education, and with sufficient family labour, the social setting is generally viable to introduce, develop, and disseminate rainwater harvesting technologies in a more cost-effective and socially acceptable manner among the farmers in Tharaka sub County. However, the low level of access to credit by most of the farmers (less than 31%) could be an impediment to dissemination of technologies, especially those that are resource intensive in terms of their installation and maintenance. In resource-limited households, less resource-intensive technologies especially biological technologies such as grass strips, may be suitable technology options especially when soil, terrain, and climatic characteristics are favourable for the installation of these technologies. Household size averaged 5 members, while household members involved in day to day activities consisted of 3 members (Table 2). Years of farming experience ranged from 1- 57 years, with an average of 18 years. Farm size ranged from 0.1 ha to 18 ha (mean =2.3 ha) while land under crops was 1.5 ha. Land under pasture averaged 0.8 ha while fallow land averaged 0.4 ha. 246 Land areas under dry-season irrigation and permanent irrigation ranged between 0-8.1 ha averaging 0.2 ha (Table 2). Table 1: Household and farm characteristics of small-holder farmers in Tharaka-Sub- County Parameter N Minimum Maximum Mean Std.Dev Household size 353 1.0 11.0 4.9 1.94 Household members involved in day to day farming 338 1.0 10.0 2.9 1.78 Years of farming 340 1.0 57.0 17.6 13.62 Farm size (ha) 353 0.1 18.2 2.3 2.50 Land under crops (ha) 350 0.0 8.9 1.5 1.10 Land under pasture (ha) 323 0.0 14.2 0.8 1.70 Land under fallow (ha) 301 0.0 6.1 0.4 0.80 Land under continuous irrigation (ha) 301 0.0 8.1 0.2 0.60 Land under dry spell irrigation (ha) 297 0.0 8.1 0.2 0.60 The number of household members involved in active running of the farm would have an influence on the rainwater harvesting and conservation technology utilized. This is because the technologies vary in their labour requirement. Households with more individuals available to contribute farm labour are likely to install and utilize labour intensive technologies for instance zai pits and fanya juu terraces. Some of the technologies are highly dependent on the farm size. Technologies that lay a demand on the cultivated land for instance the use of grass strip may not be favourable among farmers with smaller sizes of cultivated land. Additionally larger total land size is an important factor in utilization of technologies such as stone terraces which require expansive piece of land for stones collection. Linkages between socio-economic characteristics and knowledge Knowledge questions connected to rain water harvesting and conservation technologies were focused on technical design and their application in agriculture and soil/water conservation. The knowledge mean scores were highest for trashlines (1.7), followed by stone terraces (1.4), grass strips (0.9), Zai pits (0.9) and Negarim pits (0.7) (Table 3). These scores indicate that farmers had higher knowledge levels for less knowledge intensive and traditional soil technologies compared to more advanced soil structures. 247 Table 3: Mean knowledge scores among farmers in Tharaka sub-county Technology Mean N Std. Deviation Trash lines 1.71 344 0.4 Stone terraces 1.40 305 0.56 Grass strips 0.90 343 0.85 Zai pits 0.88 344 0.82 Negarim pits 0.67 240 0.73 Knowledge mean scores were significantly higher among technology users in all technologies than among farmers who did not utilize the rainwater harvesting and conservation technologies (Table 4). The mean difference tended to be highest in Negarim pits and Zai pit scores compared to other technologies that were utilized by farmers (Table 4). There were no significant differences in mean scoring by education and marital status. Gender influenced significantly Zai pits and stone terrace technology scoring. In addition, the age category significantly affected the knowledge score for Trashlines and Zai pits (Table 4). Table 4: Knowledge mean scores conditioned by socio-economic characteristics of household heads in Tharaka South Sub-county Parameter Trashlines Zai pits Negarim micro- catchments Stone terraces Strip catchment tillage Utilization of technology No 1.6(91) 0.7(250) 0.6(199) 1.2(152) 0.8(273) Yes 1.8(248) 1.4(85) 1.4(34) 1.5(149) 1.4(62) Sig. 0.029 0.000 0.000 0.000 0.000 Education level None 1.8 (94) 0.9 (94) 0.7 (70) 1.5 (85) 0.9 (94) Primary 1.7 (160) 0.8 (160) 0.6 (103) 1.3 (140) 0.8 (160) Secondary 1.7 (67) 0.9 (67) 0.7 (50) 1.5 (61) 1.1 (67) College 1.7 (18) 0.9 (18) 0.7 (13) 1.2 (14) 0.9 (17) University 1.8 (5) 0.9 (5) 0.8 (4) 1.1 (5) 0.8 (5) Sig. 0.311 0.83 0.862 0.129 0.208 Marital status Single 1.7 (56) 0.9 (56) 0.7 (41) 1.4 (49) 1 (56) Married 1.7 (287) 0.9 (287) 0.7 (198) 1.4 (255) 0.9 (286) Sig. 0.608 0.562 0.639 0.846 0.391 Gender Male 1.7 (275) 0.8 (275) 0.7 (189) 1.4 (244) 0.9 (274) Female 1.7 (69) 1.1 (69) 0.7 (51) 1.2 (61) 1 (69) 248 Sig. 0.508 0.033 0.657 0.017 0.637 Age category 21-36 1.6 (88) 0.6 (88) 0.6 (62) 1.3 (77) 0.8 (88) 37-52 1.7 (117) 1 (117) 0.7 (77) 1.5 (101) 1 (117) 53-68 1.8 (98) 1 (98) 0.7 (74) 1.4 (88) 0.9 (97) Over 69 1.5 (24) 1 (24) 0.6 (18) 1.3 (24) 1.1 (24) Sig. 0.008 0.014 0.688 0.115 0.473 Mean score 1.7 (344) 0.9 (344) 0.7 (240) 1.4 (305) 0.9 (343) Values in parentheses are number of farmers Knowledge mean scores were significantly associated with utilization categories, whereby farmers with knowledge recorded significantly higher scores in all technologies. This finding collaborates well with the influence of training in most of the technologies in earlier results elsewhere. Rainwater harvesting and conservation technologies comprise a major contributor to the measures that the water sector in Africa needs to undertake to cope with future climate change (Kahinda et al., 2010). Farmers armed with new knowledge and skills are better placed to make informed decisions that pertain to improved technologies in managing their resources for improved livelihoods (Wegulo et al., 2009). Education and land size were positively and significantly correlated with knowledge levels while age was negatively correlated (Singha & Devi, 2013). This study agrees with the current where farm size is positively correlated to the knowledge on all the rainwater harvesting and conservation technologies with significance found in relation to Zai pits technology. In reference to age, knowledge level on Zai pits recorded a significant positive correlation with age contrary to the findings in (Singha & Devi, 2013). This study found out that knowledge mean scores were significantly higher among farmers utilizing the rainwater harvesting and conservation technologies as compared to the farmers who did not utilize the rainwater harvesting and conservation technologies. According to Amsalu & Graaff (2006) in a study carried out in Ethiopia; farmers’ decisions pertaining to soil water conservation were largely determined by their knowledge of the problem and the perceived benefits from conservation. The soil and water conservation measures included, stone terraces/bunds, soil bunds, waterways, grass strips, contour terraces and tree planting herein referred to as rainwater harvesting and conservation technologies. Gender influenced the knowledge level on specific rain water harvesting and conservation technologies. This was exemplified in the score for Zai pits where females were more knowledgeable than the male gender and Stone terraces where the male gender scores were significantly higher than the female gender (p<0.05). In determining 249 the traditional ecological knowledge among the Turkana of Kenya, the variation in ethnobotanical knowledge among the respondents was not explained by their age and gender (Stave et al., 2006). This is partly in agreement with the current study where age category did not explain the knowledge mean score among the respondents while gender significantly influenced the knowledge score. All technology scores were positively and significantly associated, implying that knowledge of one technology was associated with another (Table 5). Years of farming experience, Age and farm size was linked significantly to scoring on Zai technologies. Years of farming experience was also positively associated with the knowledge scored in regard to negarim micro-catchments, stone terraces and strip catchments. 250 Table 5:Correlation co-efficient between technology knowledge scores and farm characteristics in Tharaka South sub- county Parameters 1 2 3 4 5 6 7 8 9 10 11 12 Trashline score (1) 1 Zai pits score (2) 0.26** 1 Negarim score (3) 0.13* 0.53** 1 Stone terrace score (4) 0.46** 0.30** 0.34** 1 Strip catchment score (5) 0.30** 0.52** 0.47** 0.32** 1 Years of farming (6) 0.08 0.18** 0.15* 0.14* 0.11* 1 Age (7) 0.05 0.14* 0.01 0.06 0.02 0.57** 1 Farm size (8) 0.05 0.11* 0.10 0.03 0.10 0.31** 0.36** 1 Crop area (9) 0.09 0.04 -0.03 0.01 -0.04 0.20** 0.26** 0.59** 1 Household size (10) 0.04 0.07 0.06 0.01 -0.08 0.16** 0.20** 0.14* 0.12* 1 Irrigated land(continuous) (11) 0.01 -0.05 0.01 0.04 -0.02 0.09 0.08 0.23** 0.53** 0.01 1 Irrigated land(dry) (12) 0.08 -0.00 0.03 0.11 0.06 0.08 0.05 0.30** 0.54** -0.01 0.91** 1 **. Correlation is significant at the 0.01 level (2-tailed), *. Correlation is significant at the 0.05 level (2-tailed). The knowledge of one type of technology was associated with the knowledge score of other technologies which is expected, because the basic principles behind all of the technologies is almost similar and based on water and soil conservation principles. Negarim pit and Zai pit scores were highly associated possibly because these technologies are both structurally similar in some aspects-based on basins, and being that they are newly introduced in the area. From the findings of this study, it can be concluded that farmers gain experience over time and adapt better to the risks involved in their production activities hence over time they are likely to embrace rainwater harvesting and conservation technologies. 251 This is deduced from the positive association between years of farming experience and the knowledge mean score on most of the rainwater harvesting and conservation technologies investigated. Mulumba et al. (2012) reported diversity of knowledge on farm management among smallholder farmers in Uganda. This position is supported by various researchers including Mango (2002); Joshi et al. (2005) and Kiptot (2007) who asserted that knowledge evolve over time as the technology adapts to the unique circumstances of a location. Interdependent relationships in farmer knowledge A Factor analysis of scores of 6 rainwater harvesting technologies revealed 3 factors describing farmer rating patterns of the technologies, explaining 67% of the total variance in farmer perception. The first factor of types of technologies that were closely associated in their scoring matrices due to the high loading (correlation) values with component 1 included Negarim pits (0.783) and grass strips (0.815). There was a second distinct set of farmers who rated stone terraces and trash lines similarly (component 2). This group of farmers tended to rate technologies in component 1 and component 3 differently. Fanya juu and Zai scores formed a third distinct set of farmer rating pattern. All scored parameters were relevant and contributed sufficiently to the model as represented by the magnitude of their communalities (Table 6). Table 6: Rotated 3-factor model of farmer scoring of rain water harvesting technologies Technology Component Communalities 1 2 3 Grass strips 0.815 0.715 Negarim pits 0.783 0.691 Stone terrace 0.784 0.644 Trash lines 0.731 0.545 Fanya juu terraces 0.862 0.787 Zai pits 0.613 0.680 Eigen values 2.0 1.1 1.0 Variance % 33.2 18.9 15.6 Cumulative variance 33.2 52.1 67.7 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Cut point for loading extraction= 0.5 The multivariate analysis indicated underlying contrasts in the utilization and perception of the technologies by farmers. Farmer scoring of technologies was based on their specific experiences with the technologies, and as a result the scoring matrix could not have arisen by chance but rather indicates farmer typologies and underlying processes in relation to various technologies that were utilized and rated. The 3 components indicated different types or clusters of farmers in relation to the technologies due to correlation scoring matrices associated with the technologies which 252 could have resulted from different socio-economic and biophysical circumstances of the farmers. For example, stone-line and trashline component scorers indicated that these farmers predominantly used these technologies in association more than with other technologies, hence tended to assign similar rating patterns to these technologies. This finding was expected because these technologies are likely to be used in similar climatic, soil, and socio-economic settings than with other technology sets or combinations of technologies. CONCLUSIONS AND RECOMMENDATIONS Farmer scores of different technologies indicated that farmers scored highly in less knowledge intensive technologies such as grass strips compared to more knowledge- intensive technologies such as Zai pits. Overall scores indicated that farmers had a below average to average understanding of rain water harvesting and conservation technologies in terms of their technical aspects, design and application in soil water conservation. The knowledge of one type of technology was associated with the knowledge score of other technologies which was expected. Multivariate analysis showed underlying contrasts in the utilization and perception of the technologies by farmers. The factor analysis revealed 3 types of farmers in regard to technology utilization which could have resulted from different socio-economic and biophysical circumstances. These types of farmers could be strategically targeted in interventions to develop technologies in the study area, including group formation, resource allocation, training and dissemination. ACKNOWLEDGEMENTS I wish to express my sincere appreciation to my supervisors for their guidance, advice, patience, supervision, professional and moral support in the course of my studies. I wish to thank the Deuscher Akademischer Austausch Dienst (DAAD) and National Commission for Science, Technology and Innovation (NACOSTI, Kenya) for providing financial support for paying my tuition and supporting my research, respectively. The cooperation and contributions of the Tharaka Sub-county Ministry of Agriculture, livestock and Fisheries and the community members (farmers and administrations officers) is highly acknowledged. 253 REFERENCES Amsalu A, De Graaff J. 2006. Farmers’ views of soil erosion problems and their conservation knowledge at Beressa watershed, central highlands of Ethiopia. Agriculture and Human Values 23: 99–108. Cooper PJM, Dimes J, Rao K P C, Shapiro B, Shiferaw B, Twomlow S. 2008. Coping better with current climatic variability in the rain-fed farming systems of sub- Saharan Africa: An essential first step in adapting to future climate change? Agriculture, Ecosystems & Environment 126: 24–35. Creative Research Systems (CRS). 2007. The Survey System: Sample Size Calculator: Retrieved from http://www.surveysystem.com/sscalc.htm. Accessed on 3rd February, 2013 Critchley W, Siegert K.1991. Water harvesting. Food and Agricultural Organisation of the United Nations. FAO .2009. Rapid assessment of aid flows for agricultural development in Sub- Saharan Africa. Investment centre division discussion paper, September 2009. http://www.fao.org/docrep/012/al144e/al144e.pdf. Accessed 11 Nov 2014 Helmreich B, Horn H.2009. Opportunities in rainwater harvesting. Desalination 24: 118–124. Icheria B K. 2015. A Social Narrative on Tharaka People in Kenya, Africa. International Journal of Humanities Social Sciences and Education 2(2): 49–55. Jaetzold R, Schmidt H, Hornetz, B, Shisanya, C.2006. Farm Management Handbook of Kenya Subpart C1, II/C1.Nairobi: Ministry of Agriculture. Joshi PK, Jha AK, Wani SP, Joshi L, Shiyani RL. 2005. Meta-analysis to Assess Impact of Watershed Program and People’s Participation. Comprehensive Assessment of Water Management in Agriculture Research Report 8. Colombo: International Water Management Institute. Kahinda J M, Taigbenu A, Sejamoholo B, Lillie E, Boroto R. 2009. A GIS-based decision support system for rainwater harvesting (RHADESS). Physics and Chemistry of the Earth, Parts A/B/C 34: 767–775. Kahinda JM, Taigbenu A E, Boroto R J. 2010. Domestic rainwater harvesting as an adaptation measure to climate change in South Africa. Physics and Chemistry of the Earth, 35: 742–751. Kiptot, E. 2007. Eliciting indigenous knowledge on tree fodder among Maasai pastoralists via a multi-method sequencing approach. Agriculture and Human Values 24: 231–243. Labeyrie V, Rono B, Leclerc C. 2014. How social organization shapes crop diversity : an ecological anthropology approach among Tharaka farmers of Mount Kenya, Agriculture and Human Values, 31: 97–107. Mango NAR. 2002. Husbanding the land: agricultural development and socio- technical change in Luoland, Kenya. Published PhD Thesis. Netherlands: Wageningen University and Research Centre. Mulumba J W, Nankya R, Adokorach J, Kiwuka C, Fadda C, De Santis P, Jarvis D I.2012. A risk-minimizing argument for traditional crop varietal diversity 254 use to reduce pest and disease damage in agricultural ecosystems of Uganda. Agriculture, Ecosystems & Environment, 157: 70–86. Nampanya S, Rast L, Khounsy S, Windsor P A.2010. Assessment of Farmer Knowledge of Large Ruminant Health and Production in Developing Village- Level Biosecurity in Northern Lao PDR. Transboundary and Emerging Diseases, 57: 420-429. Pachpute JS, Tumbo SD, Sally H, Mul ML. 2009. Sustainability of Rainwater Harvesting Systems in Rural Catchment of Sub-Saharan Africa. Water Resources Management, 23: 2815–2839. Peacock T, Ward C, Gambarelli G. 2007. Investment in Agricultural Water for Poverty Reduction and Economic Growth in Sub-Saharan Africa: Synthesis Report. Collaborative Program of African Development Bank, FAO, IFAD, IWMI and World Bank. IWMI, Colombo. Recha CW, Mukopi MN, Otieno J O. 2014. Socio-Economic Determinants of Adoption of Rainwater Harvesting and Conservation Techniques in Semi- Arid Tharaka. Land Degradation & Development. Article in press Singha, AK, Devi, S. 2013. Analysis of Factors Influencing Farmers ’ Knowledge on Resource Conservation Technologies ( RCTs ) in Rice ( Oryza sativa L .) Cultivation. Journal of Agricultural Science, 4(1): 13–19. Smucker TA, Wisner B.2008. Changing household responses to drought in Tharaka, Kenya: vulnerability, persistence and challenge. In: Overseas Development Institute (doi:10.IIII/ j.0361-3666.2007.01035.x). New York: Blackwell. Stave J, Oba G, Nordal I, Stenseth N C. 2006. Traditional Ecological Knowledge of a Riverine Forest in Turkana, Kenya: Implications for Research and Management. Biodiversity and Conservation, 16(5): 1471–1489. Tiffen M, Mortimore M, Gichuki F. 1994. More people, less erosion: Environmental recovery in Kenya. Nairobi:Acts Press. Wegulo FN, Wandahwa P, Shivoga W, Tabu I, Muhia N, Inoti S. 2009. Egerton University, Kenya Sustainable Development in Africa. The Role of Higher Education: Engaging Communities in Soil Fertility Management for Sustainable Agricultural Production: Case Studies from Kakamega and Nakuru Districts. Wildemeersch JCJ, Timmerman E, Mazijn B, Sabiou M, Ibro G, Garba M, Cornelis W. 2013. Assessing the constraints to adopt water and soil conservation techniques Tillaberi, Niger. Land Degradation & Development DOI: 10.1002/ldr.2252 255 6.2 Efficacy of Selected Crude Plant Extracts on Post-Harvest Pathogens in Tomato Fruits in Kirinyaga County, Kenya L. G. Mugao1,G. M. Kariuki2 J. K. Birgen1and A. Machocho3 1Department of Plant Sciences, Kenyatta University, P.O. Box 43844-00100, Nairobi, Kenya. 2Department of Agricultural Science and Technology, Kenyatta University, P.O. Box 43844-00100, Nairobi, Kenya. 3Department of Chemistry, Kenyatta University, P.O. Box 43844-00100, Nairobi, Kenya. Correspondence : mugaolydia@gmail.com ABSTRACT Post-harvest diseases of tomato fruits pose a very big challenge in the food chain worldwide. Several types of synthetic chemicals have previously been used to control post-harvest decay of fruits but there is increasing consumer concern over pesticide residues on foods. Pesticides are known to be toxic, carcinogenic and cause environmental pollution. In this study, three different crude plant extracts (ginger, neem and garlic) were evaluated in vitro and in vivo at different concentrations for the control of pathogens causing post-harvest rot of tomato fruits. The concentrations used were: 1mg/ml, 2mg/ml, and 3mg/ml. The pathogen species isolate from tomato and used in this study were Fusarium, Rhizopus, Geotrichum and Erwinia. Pathogen growth media was amended with the crude extracts and the pathogens introduced into the media. Radial growth of the fungal pathogens was measured at an interval of twenty four hours after the second day for five days and was compared to the control. Colony forming units were counted for the bacterial pathogen and also compared to the control. To evaluate the efficacy of the extracts, fresh healthy tomato fruits were treated by dipping them in the crude extracts (3mg/ml of each of the extract). Disease development was evaluated and compared to the control. Results obtained showed that all extracts’ concentrations inhibited the test pathogens but the inhibitory effect was proportional to the extract concentration. Garlic extract at all concentrations was more effective against all the test pathogens as compared to other extracts. The study also demonstrated that Fusarium spp. was more susceptible to the crude extracts as compared to other test pathogens. For in vivo evaluation, there was no rot development on the treated tomato fruits but rots appeared in the controls. From the study it is evident that plant extracts can be used as safe alternatives for management of post- harvest rot causing pathogens in tomatoes thus safeguarding the safety of plant, animal and human health and the environment. Key words: Post-harvest, Plant extracts, Pathogens, Tomato fruit, Inhibition 256 Significance of the study This study provides important information on the management of tomato post-harvest losses using crude plant extracts. The outputs of this study will be important in reducing tomato post-harvest losses by improving the efficiency of post-harvest handling due to better post-harvest management strategies. INTRODUCTION Tomato is an important commercial vegetable crop in Kenya which is grown both in green houses and open fields. However it is a very perishable fruit crop with a short shelf life and very prone to post-harvest pathogens which cause significant losses on harvested fruits. Different kinds of synthetic fungicides have been successfully used to control the post-harvest decay of fruits and vegetables (Adaskaveg et al., 2004, Kanetis et al., 2007). However use of the fungicides has been coupled with major concerns which include: increasing consumer concerns over pesticide residues on foods which may be toxic and carcinogenic; predominance of fungi strains resistant to fungicides due to excessive use of fungicides; and environmental pollution. There is therefore the need to explore safer and effective means of post-harvest disease management that pose less risk to human and animal health and the environment. Natural products originating from plants and their analogues have been found as important sources of agricultural bio-pesticides which serve as anti-microbial properties of the plant extracts (Cardelina, 1995; Okigbo, 2009). Pesticides from plant origin “referred to as botanicals” are bio-degradable, environmentally friendly, cheap and easy to obtain. According to Arokiyaraj et al., (2008), Shanmugavalli et al., (2009), Swarnalatha and Reddy (2009), plants are sources of natural pesticides that lead to new pesticide development. Ramazani etal., (2002) reported that chemicals of plant origin are economical, broad spectrum and bio-efficacious. They degrade faster than chemical pesticides, are less likely to destroy beneficial organisms and as such they are considered to be environmental friendly. According to Ilondu et al. (2001) some plants contain phenolic compounds and essential oils, which have inhibitory effects on micro- organisms. This study aimed at evaluating the efficacy of selected crude plant extracts (neem plant leaf, ginger rhizome, and garlic bulb) against post-harvest rot causing pathogens (Fusarium spp., Geotrichum spp., Rhizopus spp. and Erwinia spp.) in tomato fruits. MATERIALS AND METHODS Experimental Design The experiment was conducted at Kenyatta University, Department of Agricultural Science and Technology Laboratory. The test crude plant extracts were obtained from neem leaves, garlic cloves and ginger rhizomes. The neem leaves were collected from Kenya Agricultural and Livestock Research Organization (KARLO) station in Embu 257 while the ginger rhizomes and garlic bulbs were obtained from Mwea market. The materials were then air dried at Kenyatta University, Department of Agricultural Science and Technology Laboratory. The experimental was laid out as a completely randomized design consisting of four treatments replicated four times. The treatments included: three concentrations of the crude plant extracts (1mg/ml, 2mg/ml and 3mg/ml) dispensed into four petri-dishes and a control. Preparation of plant crude extracts The extraction process followed the procedure described by Handa et al., (2008). The leaves, peeled garlic cloves and rhizomes were washed under tap water, rinsed three times in sterile distilled water and blotted dry using sterile blotting paper. They were then placed in the oven and dried at a temperature of 40°C for three days. The plant extracts were then ground into powder by use of a sterile mortar and pestle and placed in sterile specimen bottles. This was done to maximize the surface area which in turn enables the mass transfer of active ingredients from the plant material to the solvent. Fifty (50 gms) of each of the powder were put into separate sterile conical flasks and 150 ml of methanol added ensuring that the powder was completely immersed into the solvent. The mixture was shaken vigorously and allowed to stand at room temperature while shaking it at different intervals for two days. A sterile funnel was placed into a 500 ml conical flask and then a whit man’s (No.2) filter paper was folded and placed into the funnel. Some amount of the extract was filtered through a filter paper and the filtrate poured into sterile universal bottles. The crude extract was concentrated by evaporating the solvent using a vacuum evaporator for 60 minutes at a temperature of 50°C. After evaporation, a spongy like material was obtained which was later dried in an oven at a temperature of 40°C for two days. This generated a powder like substance which was stored in the refrigerator at temperature of 4°C.Different concentrations of the crude extracts were prepared by weighing separately 1mg, 2mg and 3mg of ginger, garlic and neem powder respectively. Each powder was dissolved in 1ml of sterile distilled water to form solutions of different concentrations. One hundrend (100) ml of PDA and NA were amended with 3 ml of each of the different extract concentrations respectively and dispensed into four Petri-dishes and replicated four times. Three (3) ml of water was mixed with the media for the negative controls. The media was allowed to cool and solidify. The bacteria were streaked in each of the amended media and the replicates. Evaluation of the effectiveness of crude plant extracts on growth of fungal mycelia and bacterial colonies The technique of Amadioha and Obi (1999) was used to determine the effectiveness of the crude extracts on the test pathogens. The colony forming units (CFU) were counted after 48 hr. For the fungal treatment, 5mm fungal culture discs from one week old 258 cultures of Geotrichum spp., Fusariumspp. and Rhizopusspp. were cultured at the centre of each Petri-dish per replicate and incubated at room temperature. Measurement of radial growth from each of the treatment was done after the second day and at an interval of 24 hr up to the seventh day. The mean growth of the fungi on the amended media was compared with the control. For the bacteria, colony forming units in different extract treatment were counted. DATA ANALYSIS The bio-cidal activity of the plant extracts to the pathogens was analysed using SAS one way ANOVA and Students-Newman-Keuls Test (SNKT p<0.05) was used for means separation. RESULTS Efficacy of different crude extracts concentrations on radial growth of Fusarium spp. The study revealed that the plant extracts caused radial growth inhibition of Fusarium spp. as compared to control but the efficacy varied with concentration (< 0.001). At 1 mg/ml neem, Fusarium spp. was completely inhibited (Plate 1). There was slight growth in garlic (5.40 mm) and a mean diameter of 6.4 mm was noted on treatment with ginger (Table1). At 2 mg/ml neem and garlic concentration growth of Fusarium spp. was completely inhibited (Plate 1 and plate 3) but there was slight growth in the PDA amended with ginger (Plate 2). At 3gm/ml concentration there was no Fusariumgrowth in all the extracts. There was a significant difference in the effect of different concentrations of ginger on Fusarium spp. (p<0.001). The effects of different concentrations of garlic also differed significantly (p=0.012) (Table 1). Table 1 Efficacy of crude extracts on radial growth of Fusarium spp. Concentration (mg/ml) Neem meanRG±SD Ginger meanRG±SD Garlic meanRG±SD p- value Control 42.85±2.68B 42.85±2.68D 42.85±2.68C 1 5.00±0.00A 6.40±0.50Cb 5.40±0.82Bb <0.001 2 5.00±0.00A 5.20±0.41Ba 5.00±0.00Aa 0.012 3 5.00±0.00A 5.00±0.00Aa 5.00±0.00Aa - <0.001 0.012 Mean values followed by the same lower case within the same column are not significantly different (One way ANOVA, Students-Newman-Keuls test, α=0.05) while mean values followed by the same upper case within the same row are not significantly different (One way ANOVA, Students-Newman-Keuls test, α=0.05) RG refers to radial growth and it includes inoculum disc which was 5mm. 259 Figure 1: Plates showing growth of Fusarium spp. on PDA amended with varying concentrations of garlic plant extracts Figure 2: Plates showing growth of Fusarium spp. on PDA amended with varying concentrations of ginger plant extracts Figure 3: Plates showing growth of Fusarium spp. on PDA amended with varying concentrations of neem plant extracts C B A D D C A B 260 Efficacy of different crude extracts concentrations on radial growth of Geotrichum spp. Results obtained indicated that the three plant extracts inhibited radial growth on Geotrichum spp. as compared to the control but the rate of inhibition varied significantly with different extracts’ concentrations (p<0.001) (Table 2). At 1 mg/ml concentration, the three plant extracts differed significantly in their effects on Geotrichum spp. (p<0.001) as indicated in table 2. At the concentration of 2 mg/ml the effects of neem and ginger on Geotrichum spp. also differed significantly (p<0.001). The most effective concentration in reducing Geotrichum spp. radial growth in all extracts was the 3 mg/ml. The effectiveness of neem differed significantly (p<0.001) at different concentrations (Plate 5). The effect of ginger also differed significantly (p<0.001) at different concentrations and 3 mg/ml concentration was the most effective in reducing the radial growth of Geotrichum spp. (Plate 6). Garlic was the most effective in inhibiting growth of Geotrichum spp. among the tested extracts (Plate 4). Table 2 Efficacy of crude extracts on radial growth of Geotrichum spp. Concentration (mg/ml) Neem meanRG±SD Ginger meanRG±SD Garlic meanRG±SD p- value Control 34.80±2.42d 34.80±2.42c 34.80±2.42B 1 7.70±2.08Cc 6.10±1.07Bb 5.00±0.00A <0.001 2 6.20±1.20Cb 5.60±0.50Ba 5.00±0.00A <0.001 3 5.00±0.00a 5.00±0.00a 5.00±0.00 - p-value <0.001 <0.001 - Mean values followed by the same lower case within the same column are not significantly different (One way ANOVA, Students-Newman-Keuls test, α=0.05) while mean values followed by the same upper case within the same row are not significantly different (One way ANOVA, Students-Newman-Keuls test, α=0.05) RG refers to radial growth and it includes inoculum disc which was 5mm. C B D A 261 Figure 4: Growth of Geotrichum spp. on PDA amended with varying concentrations of garlic crude plant extracts Figure 5: Growth of Geotrichum spp. on PDA amended with varying concentrations of neem crude plant extracts Figure 6: Growth of Geotrichum spp. on PDA amended with varying concentrations ginger crude plant extracts C B D A B C A D 262 Efficacy of different crude plant extracts concentrations on radial growth of Rhizopus spp. The study revealed that the three plants extracts inhibited radial growth of Rhizopus spp. as compared to the control although the rate of inhibition differed significantly with the type of each extract and level of concentration. At 1 mg/ml and 2 mg/ml concentrations, the tested extracts differed significantly (p=0.010) in inhibiting radial growth of Rhizopus spp. with garlic proofing to be the most effective extract (Table 3). At 3 mg/ml concentration radial growth of Rhizopus was completely inhibited in all the extracts as shown in plate 7, 8 and 9 for neem, garlic and ginger respectively. The 1 mg/ml and 2 mg/ml concentration did not differ significantly in their effectiveness on Rhizopus spp. Table 3 Efficacy of crude extracts on radial growth of Rhizopus spp. Concentration (mg/ml) Neem meanRG±SD Ginger meanRG±SD Garlic meanRG±SD p- value Control 80.15±0.89d 80.15±0.89c 80.15±0.89b 1 7.20±1.77Bc 6.70±3.57Bb 5.00±0.00Aa 0.010 2 5.85±0.88Bb 5.55±1.15Bb 5.00±0.00Aa 0.007 3 5.00±0.00 A a 5.10±0.11Aa 5.00±0.00 Aa 0.012 p-value <0.001 0.028 - Mean values followed by the same lower case within the same column are not significantly different (One way ANOVA, Students-Newman-Keuls test, α=0.05) while mean values followed by the same upper case within the same row are not significantly different (One way ANOVA, Students-Newman-Keuls test, α=0.05) RG refers to radial growth and t includes inoculum disc which was 5mm. D A C B 263 Figure 7: Growth of Rhizopus spp. on PDA amended with varying concentrations of ginger crude plant extracts Figure 8: Growth of Rhizopus spp. on PDA amended with varying concentrations of neem crude plant extracts C D A B B A C D A S 264 Figure 9: Growth of Rhizopus spp. on PDA amended with varying concentrations of garlic crude plant extracts Efficacy of crude plant extracts on Erwinia (Pectobacterium) The study revealed that the number of colony forming units of Erwinia spp. on the amended Nutrient Agar (NA) differed significantly (p<0.001) in the different concentrations of the tested extracts as compared to the control. At both 1 mg/ml and 2mg/ml, garlic was the most effective extract with least colony forming units followed by neem and ginger respectively (Table 4). The efficacy of ginger crude extract on growth of Erwinia spp. differed significantly (p<0.001) at different concentration compared to the control. Three mg/ml was the most effective while the 1 mg/ml was the least effective. The effectiveness of garlic also differed significantly (p<0.001) compared to the control. Three mg/ml and 2 mg/ml concentrations had the same effect such that no growth occurred on the amended NA. A few colony forming units were on the NA amended with 1 mg/ml. On evaluation of the three extracts, garlic was the most effective but ginger and neem crude extracts did not differ significantly. Table 4 Efficacy of crude extracts on the C.F.U of Erwinia spp. Concentration (mg/ml) Neem Mean±SE Ginger Mean±SE Garlic Mean±SE 0 276.25±6.88d 276.25±6.88d 276.25±6.88c 1 89.50±8.77c 105.00±6.45c 28.25±1.65b 2 37.50±4.79b 11.25±1.49b 0.00±0.00a 3 0.00±0.00a 0.00±0.00a 0.00±0.00a p-value <0.001 <0.001 <0.001 Mean values followed by the same lower case within the same column are not significantly different (One way ANOVA, Students-Newman-Keuls test, α=0.05) D A C B 265 DISCUSSION The study revealed that the tested crude plant extracts were effective in inhibiting growth of the tested post-harvest loss causing pathogens although in most cases their efficacy varied with the concentration. The results obtained proofed that garlic, ginger and neem extracts contain anti-fungal and anti-bacterial properties by their ability to inhibit growth of the test fungi and bacteria. The findings show that garlic was the most effective crude plant extract in inhibiting the growth of all the test pathogens at the lowest concentration as compared to ginger and neem. In addition, all evaluated concentrations of garlic had significant reduction on both mycelial growth of fungal pathogens and the colony forming units of Erwinia hence could be considered as the best alternative for the management of post-harvest rot causing pathogens in tomato. The findings are in agreement with those of other scientists such as Dutta et al., (2004), who reported that 10% concentration of crude garlic showed total inhibition of sclerotial production and 20% concentration showed excellent mycelial inhibition of R.solani causing sheath blight of rice. Reports of Anjorin et al., (2008) showed that garlic effectively inhibited Fusarium spp. Bhuiyan et al., (2008), found garlic extracts to be effective in controlling growth of Colletotrichum dematium at 20 % concentration. Sowjanya and Manohara (2000) also reported that amongst five plant extracts tested, garlic was the most effective completely checking the mycelial growth at 10% concentration. Furthermore, reports by Paradza et al. (2011) showed that when garlic and neem extracts were used to control bacterial soft rot, garlic was the most effective in reducing bacterial maceration of the potato tissue. The study further established that the tested crude plant extracts demonstrated different inhibition ability against the tested pathogens which could be attributed to differences in degree of sensitivity of each of the test pathogen to the different doses of the extracts. Fusarium spp. proofed to be the most sensitive pathogen in all the plant extracts tested with Neem leaf extract showing to be the most effective extract on the Fusarium spp. These results agree with the report of Hassanein et al. (2008), where four concentrations of neem extracts were evaluated and the lowest concentration (20 %) effectively suppressed mycelial growth of F. oxysporum (100 %).According to Amadi and Olusanmi (2009), extracts from garlic and neem have anti-microbial properties against a wide range of pathogens but garlic was the most effective. Garlic contains a compound known as allicin which is readily membrane-permeable and undergoes thiol-disulphide exchange reactions with free thiol groups in proteins and the compound is anti-bacterial, anti-fungal and anti-viral (Miron et al., 2000 and Daniela et al., 2008). When garlic bulbs are damaged a substrate alliin mixes with the enzyme alliin-lyase and forms a volatile compound which is fungicidal and disrupts fungal cell metabolism due to oxidation of proteins (Slusarenko et al., 2008). Such properties may be the basis for anti-microbial action. According to Alan et al. (2008) reduction in disease was due to direct action against the pathogen since no accumulation of salicylic acid was observed after treatment with garlic crude extract to control downy mildew of 266 Arabidopsis. Reports by Udo et al. (2001) showed inhibition of growth and sporulation of fungal pathogens in Ipomea batatas by garlic extracts. Stangarlin et al. (2011) revealed that aqueous extract of ginger at different concentrations had effect on the mycelial growth and sclerotial production of Sclerotina sclerotiumin vitro. The anti-microbial property of ginger in reducing the mycelial growth of fungal pathogens are in agreement with the results of this study. The inhibitive effect was proportional to the concentration of the crude extract: the higher the concentration the higher the inhibitory effect. According to Ijato (2011), extracts of Z. officinale and Ocimum gratissimum were mycotoxic to F. oxysporum, A. flavus and A. niger that causes post-harvest rot of yam tubers and that the effectiveness of the extracts increased with increase in concentration as was observed in this study. Chiejina and Ukeh (2012) reported that the efficiency of ginger extracts may be due to high contents of alkaloids contained in it. Reports of Okwu (2004) ranked alkaloids as the most significant, efficient and therapeutic plant substances. Results of Chuku etal. (2010) showed that ginger extract at a concentration of 3 gm/20ml of extract completely inhibited fungal growth. Reports from Nahed (2007) showed that cold extracts of A. indica inhibited growth of F. oxysporum, which causes rot of cucumber. Cassava anthracnose caused by C. gloesporides was controlled using neem extracts (Fokunang etal., 2000). Hoque et al. (2007) reported that neem contains a compound known as mahmoodin which is active against gram- positive and gram-negative bacteria. This supports the finding of this study which showed a significant slowed growth of Erwinia spp. in petri dishes treated with neem extracts. However the results from this study did not agree with those of Chuku et al. (2010) who reported that garlic was not effective in controlling fruit rot causing pathogens. CONCLUSION Ginger, garlic and neem crude plant extracts were shown to be potential inhibitors to tomato fruit rot causing pathogens at different concentrations. They were found to have anti-microbial compounds that act against the pathogens. Each extract was found effective against the test pathogens but efficacy varied with different extract concentrations. The findings of this study are in line with others that have shown that crude plant extracts could provide an alternative means for the management of post- harvest rot causing pathogens in tomato fruits. Results of this study thus provide an important step in developing plant based bio-pesticides for the management of fruit rots because the tested plants extracts are readily available and cheap. Adoption of the plant extracts could offer a viable solution to the management of post-harvest pathogens in tomato hence safeguarding plant, human and animal health as well as the environment. 267 RECOMMENDATIONS There is need for further research to determine residual levels of the plant extracts on fruits and their implication on the health of the consumers. REFERENCES Adaskaveg, J. E., Forster, H., and Sommer, N. F. 2002. Principles of post-harvest pathology and management of decays of edible horticultural crops. In: Post- harvest Technology of Horticultural Crops, 332: 200-250. Alan, J. S., Anant, P. and Daniela, P. 2008. Control of plant diseases by natural products: Allicin from garlic as a case study. European Journal of Plant Pathology, 121: 313-322. Amadioha, A. C. and Obi, I. V. 1999. Control of Anthracnose diseaseof cowpea by Cymbopogon citrates and Ocimumgratissimum. Journal of Medicinal Plants, 34 (1-2):85-89. Anjorin, S. T., Makun, H. A., Adesina, T. and Kudu, I. 2008. Effects of Fusarium verticillioides, its metabolites and neem leaf extract on germination and vigour indices of maize ( Zea mays L.). African Journal of Biotechnology, 7:2402-2406. Arokiyaraj, S., Martin, S., Perinbam, K., Arockianathan, P. M. and Beatrice, V. (2008). Free radical scavenging activity and HPTLC finger print of Pterocarpus santalinus L. – an in vitro study. Indian Journal of Scientific Technology, 1 (7): 1-7. Bhuiyan, M. K., Shovan, L. R., Begum, J. A. and Pervez, Z. A. 2008. In vitro control of Colletotrichum dematium causing anthracnose of soybean by fungicides, plant extracts and Trichoderma harzianum. International Journal of Sustainable Crop Protection,3(3):10-17. Cardelina, J. H. 1995. Natural products in the search for new agro-chemicals. In H. G. Gulter, (Ed). Biologically active natural products and potential use in agriculture. Biologically Active New York Science Journal, 200:305-311. Chiejina, N. V. and Ukeh, J. A. 2012. Anti-microbial properties and phyto-chemical analysis of methanolic extracts of Aframomum melegueta and Zingiber officinale on fungal diseases of tomato fruit. Journal of Natural Sciences Research, 2(6):10-15. Chuku, E. C., Osakwe, J. A. and Daddy-West, C. 2010. Fungal spoilage of tomato (Lycopersicon esculentum Mill) control using garlic and ginger. Scientific Africana, 9(2): 42-50. Daniela, P., Eckhard, K. and Alan, J. S. 2008. Effects of garlic (Allium sativum) juice containing allicin on Phytophthorainfestans and downy mildew of cucumber caused by Pseudoperonospora cubensis. European Journal of Plant Pathology, 122:197-206. Dutta, S., Chaudhury, A. K. and Laha, S. K. 2004. In vitro fungi-toxicity of plant extracts against Pycularia oryzae, Rhizoctoniasolani incidence of blast sheath blight of rice. Indian Phytopathology Journal, 57:344-352. Fokunang, C. N., Ikotun, T., Dixon, A. G. O., Akem, C. N., Tembe, E. A. and Nukenine E. N. 2000. Efficacy of anti-microbial plant crude extracts on the 268 growth of Colletotrichum gloesporoides f.sp manihotis in Pakistan. Journal of Biological Sciences, 3 (6):928 – 932. Handa, S. S., Khanuja, S. P. S., Longo, G., and Rakesh, D. D. 2008. Extraction technologies for medicinal and aromatic plants. International centre for science and high technology, 1:21-25. Hassanein, N. M., Aou Zeid, M. A., Youssef, K. A. and Mahmoud, D. A. 2008. Efficacy of leaf extracts of neem (Azadirachta indica) and chinaberry (Melia azedrach) against early blight and wilt diseases of tomato. Australian Journal of Basic and Applied Sciences, 2(3): 763-772. Hoque, M. D., Bari, M. L., Inatsu, Y., Juneja, V. K. and Kawamoto, S. 2007. Anti- bacterial activity of guava (Psidium guajava L.) and neem (Azadirachta indica, A.juss.) extracts against foodborne pathogens and spoilage bacteria. Foodborne Pathogens and Diseases,4(4): 481-488. Ijato, J. Y. 2011. Inhibitory effect of two indigenous plant extracts of Zingiber officinale and Ocimum gratissimum on post-harvest yam (Dioscorea rotundata) rot in vitro. Journal of American Science, 7(1):43-47. Ilondu, E. M., Ejechi B. O. and Souzey J. A. 2001. Microbial stability of jam prepared from velvet tamarind and preserved by combined processes. Nigerian Journal of Microbiology, 5: 93-96. Kanetis, L., Förster, H., and Adaskaveg, J. E. 2007. Comparative efficacy of the new post-harvest fungicides azoxystrobin, fludioxonil, and pyrimethanil for managing citrus green mold. Plant Diseases, 91:1502-1511. Miron, T., Rabinikov, A., Mirelman, D., Wilchek, and M., Weiner, L. 2000. The mode of action of allicin: Its ready permeability through phospholipid membranes may contribute to its biological activity. Indian Phytopathology Journal, 1463: 20-30. Nahed, Z. H. 2007. Improving biological control of fusarium rot incucumber(Cucumis sativus L) by allelopathic plant extracts. International Journal of Agriculture and Biology, 1560 (3): 459 – 461. Okigbo, R. N. 2009.Variation in phytochemical properties of selected fungicidal aqueous extract of some plant leaves in Kogi State, Nigeria. American European Journal of Sustainable Agriculture, 3 (3): 407-409. Okwu, D. E. 2004.Phytochemicals and vitamin content of Indigenous spices of South Eastern Nigeria. Journal of Sustainable Agricultural Environment, 6:30-34. Paradza, V. M., Icishahayo, D., and Ngadze, E. 2011. Efficacy of botanical extracts from garlic and neem on controlling potato soft rot pathogens. Crop Protection, 20:20-25. Ramazani, H. P., Batish, D. R. and Dragan, J. S. 2002. Fungicidal effects of volatile oils from Eucalyptus citridera and its major constituents. Citronellal. A paper at the 55th conference proceedings, New Zealand and Plant Protection Society Incorporated, 327-330. Shanmugavalli, N., Umashankar, V. and Raheem. 2009. Anti-microbial activity of Vanilla planifolia. Indian Journal of Science and Technology, 2 (3), 37-40. 269 Slusarenko, A. J., Patel, A. and Portz, D. 2008. Control of plant diseases by natural products: Allicin from garlic as a case study. European Journal of Plant Pathology, 121:313-322. Stangarlin, J. A., Kuhn, O. J., Assi, L., and Schwan-Estrada, K. R. F. 2011. Control of plant diseases using extracts from medicinal plants and fungi, In: Communicating Current Research and Technological Advances, 25:1033- 1042. Swarnalatha, L. and Reddy, P. N. 2009. Anti-microbial, anti-diarrhoeal and analysis of phytochemical constituents of Sphaeranthus amaranthoides. Indian Journal of Science and Technology, 2 (3): 45-48. Udo, S. E., Madunagu, B. E. and Isemin, C. D. 2001. Inhibition of growth and sporulation of fungal pathogens on sweet potato and yam by garlic extract. Nigerian Journal of Botany, 14:35-39. 6.3 Evaluation of Belt 480 SC (Flubendiamide 480 g/L) against tomato leaf miner, Tuta absoluta at Mwea in Kirinyaga County 1Kambo CM, 1Kasina M, 2Kurendi A, 1Kinyua ZM and 1Njenga GK 1Kenya Agricultural and Livestock Research Organization 2Bayer East Africa Ltd Correspondence: Email: cmkambo@gmail.com , 0722836911 ABSTRACT Tomato, Lycopersicon esculentum Mill is one of the major vegetable crops grown in Kenya, mainly for domestic consumption. The vegetable has many health benefits including excellent source of vitamin A, B6, C, E and K. The main tomato growing areas include, Kirinyaga, Kajiado and Taita Taveta counties. However, tomato production is faced with many production challenges, including, soil types, water requirements, weeds, disease and arthropod pests. The tomato leaf miner, Tuta absoluta, a major invasive pest of tomato was reported for the first time in Kenya in April, 2014. The pest was reported as attacking tomato in both green house and field leading to great losses in both yields and quality. Following the outcry by farmers, Bayer EA Ltd in collaboration with KALRO conducted field experiments to evaluate the efficacy of Belt 480 SC (Flubendiamide 480 g/L) against the leaf miner. The evaluation was carried out in two sites, namely Redsoil and Kiorugari in Mwea, Kirinyaga County between July and December 2014. Four treatments consisting of Belt 480 SC at (2, 4 and 8 ml per 20 Litres of water) and untreated control were applied on tomato, variety Kilele F1. The experiment was set up in a randomized complete block design with 4 replicates. Data on Tuta larvae, mines per leaf, total fruit yield and percentage marketable yield were processed by use of SAS statistical package. In conclusion, Belt 480 SC at 2 ml in 20 litres of water which recorded 96-100% as compared to the untreated control with 62-73% as 270 marketable yield was recommended for the management of tomato leaf miner, Tuta absoluta. Key words: Tomato, Belt SC 480, tomato leaf miner, Tuta absoluta INTRODUCTION Vegetable sub-sector is an important and integral part of the Kenya’s agricultural development. It is important in attaining food security and improving livelihood for smallholder farmers. The major vegetables produced in the country include cabbages, tomatoes and kales. Tomato is the third most important vegetable in the country in terms of tonnage produced. In 2013, the area under tomatoes was 22,865 hectares planted mainly under irrigation. The total production was 494,036.5 tons with a value of KES 14.1 Billion HCDA (2013). Tomato growing is a countrywide activity and is done mainly by small scale farmers including the youth. Pests and diseases are a major challenge in production of tomatoes. Diseases of economic importance include bacterial wilt and late blight. The common pests are African bollworm, red spider mites and whiteflies. In 2014, the Ministry of Agriculture, Livestock and Fisheries reported that a new invasive pest, identified as tomato leaf miner, Tuta absoluta was present in most of tomato growing areas of the country (Karanja T. (2014), Koskei F. (2014). The pest was found to attack tomatoes both in the field and also in screen-houses specifically in Isiolo, Meru, Kirinyaga, Nyeri, Kajiado and many other counties in the country, (Kambo et. al., (2014), IPPC, (2014), Kasina and Lusike (2014). The pest is a native of South America, particularly Peru, Bolivia and Chile where it was described in 1917. The pest has been reported in Europe particularly, Spain in the year 2006. It was later reported in North America and other European Countries including Germany, Netherlands, U.K., Italy and Turkey in 2010 (Nicholas et. al. (2010). The pest established in Africa as follows:- Morocco 2007, Tunisia 2008, Canary Islands 2008, Algeria 2008, Senegal 2012, Niger 2012 Sudan 2012 and Ethiopia 2012 (Muniappan R (2013). The spread of the insect over long distances is as a result of trade in infested tomato fruits and ability of the adult insects to move several kilometers by flying or drifting with the wind. Other factors that have led to the global rapid distribution of this pest include: Wide spread tomato cultivation, high number of insect pest generations (up-to 10) per year, Lack of specific natural enemies and fast development of resistance to insecticides. In view of this, Bayer EA Ltd and Kenya Agricultural and Livestock Organization partnered together with the aim evaluating Belt 480 SC (Flubendiamide 480 g/L) insecticide against this invasive tomato for registration as a pest control product (PCPB, 2012). No standard local check was used in the trial since there is no currently registered Soluble Concentrate (SC) product in the control of the tomato leaf miner in Kenya. 271 OBJECTIVES: i. To determine the efficacy of Belt 480 SC (Flubendiamide 480 g/L) against Tuta absoluta on tomatoes. ii. To assess the effect of Tuta absoluta on tomato yield and quality MATERIALS AND METHODS Trial site and study design The study was carried out at Mwea, Kirinyaga County, in two different locations - Redsoil and Kiorugari - between July 2014 and December 2014. Tomato variety Kilele was transplanted from a nursery and planted at a spacing of 90 x 45 cm in 4 x 4 metre plots (AEI (2011). The plots were separated with 2-metre paths between the plots and blocks. Agronomic practices including weeding and fertilizer application were applied as required. The treatments used were 3 rates of Belt 480 SC (Flubendiamide 480 g/L) and an untreated control as shown below (Table 1). The treatments were arranged in a randomized complete block design (RCBD) with 3 replications. All Belt 480 SC treatments were applied at a weekly interval using a lever-operated knapsack sprayer. Control plots were not treated with any product (Ciba Geigy, 1981). Table 7: Treatments used in evaluation of Belt 480 SC Product Active Ingredient Rate per 20 litre of water Rate/ha Belt 480 SC Flubendiamide 480 g/L 2.0 ml 0.05 L Belt 480 SC Flubendiamide 480 g/L 4.0 ml 0.10 L Belt 480 SC Flubendiamide 480 g/L 8.0 ml 0.20 L Control Nil 0.0 ml 0.00 L Sampling Data on tomato leaf miner was collected on weekly basis, starting from the 2nd week after transplanting. Five plants per plot were tagged and leaf damage (mined) as a result of Tuta absoluta infestation was thoroughly scrutinized and counts of mines were recorded. Live Tuta larvae on the treatment plots were recorded in well-structured data sheets. Yield data was collected through harvesting of fruits, which were then graded into marketable and unmarketable fruits i.e. undamaged and damaged fruits, respectively. DATA ANALYSIS Data on mines per leaf, total fruit yield and percentage marketable yield were processed by use of two-way Analysis of Variance (ANOVA) and Least Significant Difference was used for the separation of treatment means (Montogomery, 1976). SAS statistical package was used for the analysis (SAS, 2014). Data was transformed using square root formula to normalize it where it was found to be skewed, before carrying out the ANOVA. 272 RESULTS AND DISCUSSION Effect of Belt 480 SC against tomato leaf miner at Kiorugari There was a significant difference between treated and untreated plots with regard to the number of leaves showing Tuta damage from 2 weeks after the first insecticide application and this trend continued throughout the experimental period as recorded at 4, 6, 8 and 10 weeks after the first insecticide application (Table 1). Plots under all the 3 rates of Belt 480 SC (2.0, 4.0 and 8.0 ml/20 L of water) showed significantly few damaged (mined) leaves compared with the untreated control plots. The numbers of mined leaves in plots treated at the three rates of Belt 480 SC were not significantly different from each other at the various sampling times. Table 8: Larvae populations at various sampling times at Kiorugari No. of mined tomato leaves per plant (x), at various periods (Weeks after 1st insecticide application) Treatment Rate/20 L water 0 2 4 6 8 10 Pooled means Control 0.0 ml 2.82b 4.72a 5.87 a 5.40 a 5.15a 5.32a 5.04a Belt 480 SC 2.0 ml 3.79a 2.49b 1.09 b 0.94 b 0.88b 0.88b 1.55b Belt 480 SC 4.0 ml 3.31b 2.20b 1.16 b 0.97 b 1.13b 0.88b 1.51b Belt 480 SC 8.0 ml 2.88b 1.99b 1.13 b 0.97 b 0.88b 0.88b 1.37b LSD (p=0.05) - 0.83 1.21 0.58 0.48 0.66 1.04 0.54 P-value - 0.09 ‹0.00 1 ‹0.00 1 ‹0.00 1 ‹0.001 ‹0.001 ‹0.001 CV (%) - 13.01 34.90 20.5 8 19.1 8 26.94 26.14 44.80 Means in the same column followed by the same letter are not significantly different at p= 0.05 level. The figures presented in Table 2 are (√ x+0.5) = Square root transformation, (√ x+0.5), where x = No. of mined leaves/plant The number of live Tuta absoluta larvae on tomato leaves was significantly reduced 2 weeks after the first insecticide application. This trend was similar throughout the season as recorded in 4, 6, 8 and 10 weeks after the first insecticide application (Table 2). All the 3 rates of Belt 480 SC (2.0, 4.0 and 8.0 ml/20 L of water) showed significantly lower Tuta larvae population densities than the untreated control plots. The pooled (overall) means revealed that Belt 480 SC was effective against the tomato leaf miner at 273 all the three rates of application (2.0, 4.0 and 8.0 ml/20 L of water) as evidenced by low larvae infestation levels compared with the control. The Tuta population densities at the three application rates were not significantly different from each other. Table 9: Larvae populations under various treatments at Kiorugari, Mwea No. of live Tuta absoluta larvae per leaf (x), (Weeks after 1st insecticide application) Treatment Rate/20 L water 0 2 4 6 8 10 Pooled means Control 0.0 ml 1.88a 3.27a 3.80a 2.79a 2.58a 2.67 a 2.94a Belt 480 SC 2.0 ml 1.95a 1.09b 0.71 b 0.71 b 0.71b 0.71 b 0.93b Belt 480 SC 4.0 ml 1.61a 1.03b 0.71 b 0.71 b 0.71b 0.71 b 0.86b Belt 480 SC 8.0 ml 1.27a 0.95b 0.71 b 0.71 b 0.71b 0.71 b 0.81b LSD (p=0.05) - 1.21 0.75 0.18 0.44 0.13 0.29 0.26 P-Value - 0.544 ‹0.00 1 ‹0.00 1 ‹0.00 1 ‹0.001 ‹0.00 1 ‹0.001 CV (%) - 36.20 38.80 9.92 20.08 9.14 12.1 4 37.07 Means in the same column followed by the same letter are not significantly different at by LSD at p= 0.05 level. The figures presented in Table 3 are (√ x+0.5) = Square root transformation, (√ x+0.5), where x = Tuta absoluta larvae /leaf There was no significant difference in the total yield for tomatoes where Belt 480 SC was applied or not (Table 3). However, there were highly significant differences between plots with pesticide application and untreated controls with regard to the quantities of unmarketable and marketable yields. Untreated control plots had a significantly lower proportion of marketable yield, which was about 62% of the total yield, than the Belt- treated plots, which had high proportions of marketable yields, ranging from 92-96% (Table 3). The three different Belt 480 SC concentrations did not give statistically different yields. 274 Table 10: Tomato yields and marketability under Belt 480 SC treatments Treatment Rate/20 L water) Total yield/plot (g) Unmarketable yield (g) Proportion of marketable yield (%) Control 0.0 ml 3,440a 1,319.0a 61.6b Belt 480 SC 2.0 ml 2,673a 104.0b 96.1a Belt 480 SC 4.0 ml 2,849a 233.3b 91.8a Belt 480 SC 8.0 ml 3,103a 177.3b 94.3a LSD (p=0.05) - 843.6 332 10.1 P-Value - 0.305 ‹0.001 ‹0.001 CV (%) - 54.60 141.51 23.13 Means in the same column followed by the same letter are not significantly different at by LSD at p= 0.05 level Effects of Belt 480 SC against tomato leaf miner at Redsoil, Mwea There was significant difference in the number of leaves showing Tuta damage across the various treatments at Redsoil (Table 4). Plots treated with various concentrations of Belt 480 SC showed significantly lower number of infested leaves compared with control plots throughout the season. However, there was no statistical difference amongst the numbers of leaves exhibiting Tuta damage under the three Belt 480 SC concentrations. 275 Table 11: Number of tomato leaves with Tuta mines under various dosages of Belt 480 SC at Redsoil, Mwea No. of mined tomato leaves per plant (x), at various periods (Weeks after 1st insecticide application) Treatment Rate/20 L water 0 2 4 6 8 10 Pooled means Control 0.0 ml 2.68a 2.41a 3.87a 5.36a 5.17a 5.78a 4.52a Belt 480 SC 2.0 ml 2.06a b 1.22 b 0.97b 1.30b 1.26b 1.05b 1.16b Belt 480 SC 4.0 ml 1.84 b 1.22 b 0.86b 0.88b 0.97b 0.88b 0.96b Belt 480 SC 8.0 ml 2.27a b 1.16 b 0.97b 0.88b 0.80b 1.05b 0.97b LSD (p=0.05) - 0.69 1.03 0.71 0.70 0.58 0.61 0.45 P-Value - 0.104 0.054 ‹0.00 1 ‹0.00 1 ‹0.001 ‹0.001 ‹0.001 CV (%) - 15.61 56.31 35.42 27.32 23.19 23.00 45.81 Means in the same column followed by the same letter are not significantly different at by LSD at p= 0.05 level The figures presented in Table 5 are (√ x+0.5) = Square root transformation, (√ x+0.5), where x = No. of mined leaves/plant The number of live caterpillars recorded on tomato leaves was statistically different across the various treatments (Table 5). The three Belt 480 SC concentrations were not significantly different from each other but they were significantly different from the untreated control, which had the highest infestation levels. 276 Table 12: Mean number of live larvae per leaf on tomato plots grown under various treatments at Redsoil, Mwea No. of live Tuta absoluta larvae per leaf (x), (Weeks after 1st insecticide application) Treatment Rate/20 L water 0 2 4 6 8 10 Pooled means Control 0.0 ml 2.17a 1.37a 2.36a 2.78a 2.76a 2.89a 2.43a Belt 480 SC 2.0 ml 1.62a 0.71b 0.71b 0.71b 0.88b 0.71b 0.74b Belt 480 SC 4.0 ml 2.52a 0.71b 0.71b 0.71b 0.71b 0.80b 0.73b Belt 480 SC 8.0 ml 1.79a 0.80b 0.71b 0.71b 0.71b 0.71b 0.73b LSD (p=0.05) - 1.14 0.35 0.22 0.24 0.38 0.39 0.20 P-Value - 0.305 0.002 ‹0.001 ‹0.001 ‹0.001 ‹0.001 ‹0.001 CV (%) - 28.10 32.04 16.33 15.85 24.82 24.94 33.20 Means in the same column followed by the same letter are not significantly different at by LSD at p= 0.05 level. The figures presented in Table 6 are (√ x+0.5) = Square root transformation, (√ x+0.5), where x = Tuta absoluta larvae /leaf At Redsoil site, all the 3 Belt 480 SC applications resulted to production of 100% marketable yield compared with control, which recorded 76% marketable yield (Table 6). These results show that the pesticide application had significantly higher marketable yields compared with control. Table 13: Mean yield of tomatoes grown under various treatments at Redsoil, Mwea Treatment Rate/20 L water Yield/plot (g) Marketable yield (g) Percentage marketable yield Control 0.0 ml 9408a 6855b 72.86b Belt 480 SC 2.0 ml 9371a 9371ab 100.00a Belt 480 SC 4.0 ml 11517a 11514a 99.97a Belt 480 SC 8.0 ml 11429a 11429a 100.00a LSD (p=0.05) - 3915.6 3809.8 6.04 P-Value - 0.528 0.056 ‹0.001 CV (%) - 65.45 67.84 11.20 Means in the same column followed by the same letter are not significantly different at by LSD at p= 0.05 level 277 CONCLUSION The results obtained from the two trial sites, namely Redsoil and Kiorugari at Mwea, Kirinyaga County revealed that Belt 480 SC at the three tested rates (2.0, 4.0 and 8.0 ml/20 L of water) is effective against the tomato leaf miner, Tuta absoluta. The three different concentrations produced statistically similar results and led to significantly lower T. absoluta infestation levels than the untreated controls. Application of Belt 480 SC also led to significantly high proportions of marketable tomato fruit yields in comparison to untreated controls which registered large quantities of unmarketable fruits. The Kiorugari site had a relatively higher Tuta absoluta infestation than the Redsoil site. This translated to higher numbers of damaged tomato leaves, higher number of larvae and higher proportions of unmarketable fruits yields at Kiorugari than at Redsoil at comparable treatments. This shows that a high rate of application of Belt 480 SC would be appropriate under conditions of high infestation by this pest. Therefore, it is recommended that Belt 480 SC be applied at the rate of 2.0 to 4.0 ml in 20 litres of water for management of T. absoluta on tomatoes. ACKNOWLEDGEMENTS This trial was carried out with funding from the client, Bayer EA Ltd. We thank them for timely provision of resources, both finances and test materials. We thank KALRO management at Kabete and Mwea for facilitation, working space and granting of time dedicated for the trial. We acknowledge the role of PCPB staff in the implementation of the protocol. REFERENCES AEI (2011). How to grow tomatoes in Kenya. http:yagrein.blogspot.com/2011/11/how- to-grow-tomatoes.html Ciba Geigy (1981). Manual for field trials in Plant Protection. CIBA GEIGY 205 pp. HCDA (2013). Horticulture data validation for 2013. IPPC, (2014). New pest in Kenya: Preliminary surveillance report on Tuta absoluta - Preliminary report by Kenya. IPPC Official Pest Report, No. KEN-01/2. Rome, Italy: FAO. https://www.ippc.int/ Kambo C.M., Mbaka JN, Amata R and Kuria S (2014). Distribution of Tuta absoluta and its natural enemies in Kirinyaga and Embu County. KALRO National Sericulture Research Centre, Thika. 13 pp. Karanja T. (2014). Surveillance Report on Tuta absoluta (American) leafminer.Plant Protection Services Division. Powerpoint Presentation-32 slides Kasina M. and W. Lusike (2014). Tomato leafminer (Tuta absoluta). KARI-E-mimea Plant Clinic. Kenya Agricultural Research Institute- www.kari.org. 3pp. Koskei F. (2014). Press Release by Mr Felix Koskei, Cabinet Secretary, Ministry of Agriculture, Livestock and Fisheries on Tomato leafminer, Tuta absoluta in the country 5pp. 278 Montogomery, D C. (1976). Design and analysis of experiment. John Wiley & Sons Inc. 418p. Muniappan R (2013). Tuta absoluta: the tomato leafminer. Feed The Future Innovation Lab: Collaborative research on Integrated Pest Management IPM-IL Office of International Research, Education and Development, Virginia Tech. Power point presentation by Muniapan, Nov. 26, 2013, at International Hotel, Addis Ababa, Ethiopia. Nicholas D., Wajnberg E., Wyckhuys K.A.G., Burgio G., Arpaia S., Vasquez C.A.N., Cabrera D.C.R., Tabone E., Frandon J., Pizzol J., Poncet C., Cabello T. and Urbaneja A. (2010). Biological invasion of European tomato crops by Tuta absoluta: ecology, geographic expansion and prospects for biological control. J. Pest Sci. (2010) 83: 197-215. PCPB (2012). Pest Control Products registered for use in Kenya. Pp. 33 SAS (2014). SAS University Edition, SAS Institute Inc. World Hqts, SAS Campus Drive, CARY, NC, 27513. www.sas.com. Teresia Karanja T, (2014). Surveillance report on Tuta absoluta (American) Leaf Miner. Plant Protection Services Division. Ministry of Agriculture, Livestock and Fisheries MoALF, Kenya. 6.4 Developing an effective pest management strategy for French beans in Kenya to protect export market Matere1,2 NJ, S Ochola1, J Mulwa2, M Guantai2, C Kambo2, M Nyongesa3, M Otipa4, A Ndegwa5, J Ndungu5 L Wasilwa6, V Wasike7, V Kirigua6,C Gacheri8, R Barnaba8, S Njoroge9, and M Kasina2 1 Department of Environmental studies, Kenyatta University, P.O. Box 43844, GPO, Nairobi 2 National Sericulture Research Centre-KALRO, P.O. Box 7816 code 01000 Thika 3Kenya Agricultural and Livestock Research Organization (KALRO), Tigoni, P.O. Box 338, Limuru. 4KALRO Kabete, P.O. Box 14733 code 00800 Nairobi 5KALRO Kandara, P.O. Box 220 code 01000 Thika 6KALRO HQ, P.O. Box 57811, code 00200 Nairobi 7Genetic Resources Research Institute, P.O. Box 781, Kikuyu 8Dudutech Ltd, P.O Box 1927, code 20117, Naivasha 9Syngenta E.A, P.O BOX 30393, code 00100, Nairobi Correspondence: Muo.Kasina@kalro.org ABSTRACT In Kenya, French bean (Phaseolus vulgaris L.) production is majorly hampered by diseases and pests, which most farmers manage using synthetic pesticides, usually in a manner that results to exceeding maximum residue limits (MRLs) set by the import markets. Kenya has had challenges in managing these MRLs and thus this study was 279 carried out to provide best approach to the management of the pests that may result to meeting MRLs requirements and ensuring the product is of high quality. Treatments entailed planned use of synthetic pesticides, biological control (biocontrol) products and Integrated Pest Management (IPM) system. Trial was arranged in randomized complete block design with plots measuring 3m by 4m replicated four times. Data collected included recording number of pests and feeding marks on randomly selected plants per plot, every fortnight from germination till end of season. Yield data was collected twice weekly from 42 days after germination. Data was analysed using Genstat Discovery edition. Analysis of variance was performed at 95% confidence limit to significant effects across the treatments. The results show no significant difference in the number of thrips across various management practices in season 1 (P=0.286) and season 2 (P=0.503). Pesticide-applied plots had however higher infestation levels compared to the rest in both seasons. Aphids were significantly different across various management practices in season 1(P=0.000) and 2 (P=0.002). They were significantly lower in IPM applied plots and those under biocontrol compared with the pesticide-treated plots in both seasons. A highly significant (P<0.001) difference in the number of whiteflies was recorded in both seasons. Whiteflies were significantly lower in IPM plots compared with pesticide-controlled plots. Considering that IPM provided effective management of the various pests it is recommended that the system be optimized to ensure reliability and farmer adoption for French bean production. Key words: maximum residue limits, French beans, Agronets, biocontrol, pesticides INTRODUCTION French bean (Phaseolus vulgaris L.) is the second most important horticultural export crop after cut flowers in Kenya (Stetter and Folker, 2000). The beans though grown mainly for export (Nderitu et al., 2007) are also gaining popularity in domestic urban centers. Arthropod pests affecting French beans include thrips, aphids, red spider mites, white flies among others (Nderitu et al., 2007). Diseases include rust, fusarium wilt and blights (Monda et al, 2003). The abuse of pesticides has resulted in many negative externalities. Pesticides should be toxic to target organisms only, degrade naturally and within a short time (Rosell et al., 2008) but most pesticides kill non-targeted organisms and persist for long in the environment resulting in pollution and more stringent measures on allowed pesticide maximum residue levels. Pest management strategies tested in the past have mostly involved independent use of a strategy without considering a more integrated approach. This study seeks to provide a strategy that is environmentally safe with a guarantee of increased quality yield. MATERIALS AND METHODS Experiments were carried out at the Kenya Agricultural and Livestock Research Organisation research centres in Mwea, Kabete and National Sericulture Centre, Kenya, in two planting seasons. KALRO Mwea is found in Kirinyaga County lying between latitudes 37o13’E and 37o56’E and longitudes 0o10’S and 0o54’E. Its temperatures range from 16.50C to 29oC with an annual rainfall of 950mm (Kamanu et al., 2012). KALRO Kabete is located about 15 km to the West of Nairobi city and lies at Latitude 1o 15′S and 280 Longitude 36o 44′E (Sombroek et al., 1982). It has a bimodal distribution of rainfall, with mean annual temperature of 180 C (Mburu, 1996). The NSRC is located in Murang’a County and has an average temperature of 19.80 C. It receives an annual rainfall of 840 mm and is located at an altitude of 1499m ASL. (MOA Kandara, 2012). the study entailed a complete randomized block design layout with four replications. Pest data was collected every fortnight from germination date till the end of harvesting period. The plots measured 3m by 4m in all sites. The Serengeti French bean variety seeds supplied by Syngenta Company were treated as shown in Table 1. Data analysis were done using Genstat Discovery edition (Anonymous, 2015). Thrips data was square root transformed to fit assumption of analysis of variance (ANOVA). Where there was no significant difference, pooled analysis was carried out. TABLE 1: Treatment regimes Treatment Regime Description Control Control (no seed dressing except with the standard Thiram) Pesticide use Synthetic Pesticide use only Integrated Pest Management Integration of Pesticides, Bio-control and Agro-net technology Bio-control only Biological control use only RESULTS General outlook Results show no significant difference in the number of thrips across various management practices in both season 1 (P=0.286) and season 2 (P=0.503). This may have been due to the effectiveness of the various treatments in combating thrips. Aphids were significantly (P=0.000; P=0.002) different across various management practices in season 1 and 2, respectively. They were significantly lower in IPM plots and those under biocontrol compared with the pesticide-controlled plots in both seasons. A highly significant (P<0.001) difference in the number of whiteflies was recorded in both seasons. Whiteflies were significantly lower in plots under IPM compared with pesticide-controlled plots. Effects across the sites In both seasons whiteflies were significantly lower in plots under IPM and biocontrol compared with pesticide treated plots at Kabete but these were not significantly different at Mwea. Red spider mites were significantly lower in plots under biocontrol compared with IPM and pesticide treated plots in the first and second season, however in the second season they were quite high in IPM plots in Thika as compared to Mwea. 281 Table 2a: Mean number of pests infesting French beans at Mwea and Kabete Treatment Thrips Aphid colonies Whiteflies Kabete Mwea Kabete Mwea Kabete Mwea Control 8.08a 25.12 4.92b 1.46a 1.38a 15.17b Biocontrol only 18.12ab 22.00 2.38ab .42b 3.25b 12.46ab IPM 20.42b 16.04 0.88a .08b 4.92b 8.88a Pesticides only 26.62b 23.88 4.25b .38b 11.42c 14.96b P value .026 .422 .002 .000 .000 .004 N 24 24 24 24 24 24 Table 2 b Mean number of pests infesting French beans at Mwea and Thika Plots treated with Thrips Aphid colonies Whiteflies Thika Mwea Thika Mwea Thika Mwea Control 1.85 4.45 10.083a 2.746ab 19.79b 36c Biocontrol 1.82 4.05 5.250a 2.056a 15.54b 21.7b IPM 1.45 4.05 3.000a 00 a 1.686a 2.62a 4.25a Pesticides 1.96 0.9 5.958a 2.141a 15.62b 21.85b P value 0.4 0.587 0.002 <.001 <.001 N 20 20 20 20 20 Diseases During season one Root rot, Rust, Halo blight and viral diseases were present in low levels in both sites but Anthracnose and Pythium appeared only in Kabete. The diseases were not severe during the two trials as they were well managed by the different treatment applications. Root rot and rusts were recorded in the second season. Rust appeared at fruiting and harvesting period in both seasons. French beans yields grown under different treatments. IPM plots recorded significantly (37,192.411; p=.004 ) higher French bean yields in both seasons and in all sites. This was followed by those under pesticide treatment (28,302.96bc) and then biocontrol (13,836.13ab). CONCLUSIONS AND RECOMMENDATIONS From the study IPM was a better management strategy for the French bean pests as it entailed use of both pesticides and biocontrol products but with the inclusion of the Agronets. Other than physically barring pests from accessing the crop Agronets also modify the micro-climate conditions exposing plants to factors that stimulate faster growth. Improvement of IPM for example by registering biocontrol products that have 282 not been registered should be considered as seen from the results showing their great efficiency in pest control. Acknowledgements Authors thank their institutions for co-funding the study through provision of test products and staff time. The study was financially supported by KAPAP through KALRO REFERENCES Asfaw, S., Mithoefer, D. and Waibel, H. (2009). EU Food safety standards, Pesticide use and Farm-level productivity. The case of High-Value crops in Kenya. Journal Compilation, Agricultural Economics Society. 60 (3): 144-180. Nderitu, J.H, Kasina, M.J., Nyamasyo, G.N, Waturu, C.N and Aura, J. (2008). Management of Thrips (Thysanoptera: Thripidae) on French Beans (Fabaceae) in Kenya: Economics of Insecticide Applications. Journal of Entomology, 5: 148-155. Kamanu, K., Chemining’wa, G.N., Nderitu, J.H. and Ambuko, J. (2012). Snap bean plant response to different inorganic fertilizers in Kenya. Journal of Applied Biosciences 55: 3944– 3952. Monda, E.O., Munene, S., and Ndegua,A.(2003). French beans production in Kenya. Proceedings of the African Crop Science Conference. 6(2 ):683-687. Rosell, G., Quero, C., Col, l J., Guerrero, A.(2008). Bio-rational insecticides in pest management. Journal of Pesticide Science, 33(2):103-121. Wasonga, J.C., Pastor-Corrales, M.A., and Griffiths, D.P (2010).Targeting Gene Combinations for broad-spectrum Rust Resistance in Heat-tolerant Snap Beans Developed for Tropical Environments. Journal of American Society of Horticulture Science,135 (6): 521-532. 6.5 Evaluation of Tihan OD 175 (Flubendiamide 100 g/L + Spirotetramat 75 g/L) against sucking and chewing arthropod pests of cotton at Mwea in Kirinyaga County 1Kambo CM, 2Kurendi A and 1Njenga GK 1Kenya Agricultural and Livestock Research Organization 2Bayer East Africa Ltd Correspondence: Email: cmkambo@gmail.com , 0722836911 ABSTRACT A field experiment to evaluate the of efficacy of Tihan OD 175 (Flubendiamide 100 g/L + Spirotetramat 75 g/L) against sucking and chewing arthropod pests of cotton was carried out in the field at conducted at KARI-Mwea between October 2014 and June, 2015. Cotton variety, HART 89M, was grown under rain fed conditions using the currently recommended agronomic practices. Cotton variety, HART 89M was planted in a 283 randomized complete block design with 4 replicates. The crop was grown using recommended agronomic practices. The treatments used were 3 rates of Tihan OD 175, local check, Thunder 145 OD and untreated control. Tihan OD 175 at 0.20 and 0.25 L ha- 1 significantly reduced cotton aphids, African bollworms and mealybug populations as compared to the untreated control plots. The results obtained in this trial show that Tihan OD 175 at the rate of 0.20 - 0.25 litres in 500 litres of water per hectare was found to be effective against cotton aphids, bollworms and cotton mealybugs. Key words: Cotton variety, Tihan OD 175, aphids, African bollworm, mealybugs INTRODUCTION Cotton (Gossypium hirstum [L]) is one of the most important cash crops in Kenya. The crop was introduced in Kenya in the early 1900’s by the British colonial administration. It is grown in several agro ecological zones, mainly in the semi-arid regions of Eastern, Central, Nyanza, Coast, Western and Rift Valley provinces. Since its introduction, the crop has been characterized by fluctuating production trends. Between 1965 and 1984, the annual national lint production grew from 20,000 to 70,000 bales and the textile- apparel industry was the leading manufacturing industry in size and employment. The increase was due to both government and donor support aimed at reducing production constraints and achieving self-sufficiency. During the 1980’ peak production period, Bura and Hola Irrigation Schemes produced about 40% of the national cotton production. However by mid-1990’s the lint production declined to an average of 20,000 bales due to liberalization of the cotton industry. In 2005, the government invested a lot of funds for the revival of the cotton industry. From that time todate, there has been a gradual increase in cotton production and the current national production is on the range of 50,000 bales per year. Cotton is the most important fibre crops that provide fibre for the textile industries and is a source of income to resource-poor farmers. The seeds are also used as feed for livestock and as food to human beings (Lutrel, et al., 1994). In Kenya, up to 4000 kg haˉ¹ and 900 kg haˉ¹ of seed cotton have been recorded in research centres and farmers fields respectively (Ikitoo. et al,. 1989). This wide yield difference is probably due to poor agronomic practices, low soil fertility, rainfall patterns, disease and arthropod pests (Munro, 1987, Ikitoo et al., 1989). Cotton is attacked by several arthropod pests as early as from germination stage to the end of season. Some of the common early season pests are the cotton aphid (Aphis gossypii), red spider mite (Tetranychus telarius), thrips (Frankliniella sp), African bollworm (Helicoverpa armigera), tobacco whitefly (Bemisia tabaci) and cotton mealybug (Phenacoccus spp.). The African bollworm is the most serious pest, appearing at flowering stage from 8-9 weeks after plant emergence. Cotton stainer (Dysdercus sp) attack the crop during the opening of bolls, staining the lint which lowers the quality of seed cotton (Munro, 1987). The main management strategy currently is by application of synthetic pesticides. Many tested synthetic pesticides are effective against cotton pests. However, in order to reduce chances of insect developing resistance to the synthetic pesticides, there is need 284 to continually evaluate new pest control products (Munro, 1987). This would ensure that there are enough pesticides for control of cotton pests including the cotton mealbugs. Tihan OD 175 (Flubendiamide 100 g/L + Spirotetramat 75 g/L) is an Oil Dispersion, currently there is no OD that has the same active ingredients as Tihan OD 175. In view of these the author of this report chose Thunder 145 OD (Imidacloprid 100g/L + Betacyfluthrin 45 g/L) which is already in the market and controls a wide range of sucking pest as a local check. Thunder is registered (PCPB (CR) 0739) for the control of thrips, aphids and whiteflies in horticultural crops (Tomatoes and French beans (PCPB, 2012). OBJECTIVES To determine the effective dosage of Tihan OD 175 (Flubendiamide 100 g/L + Spirotetramat 75 g/L) against mealybugs, caterpillars, red spider mites, aphids, white flies, thrips, African bollworm and cotton stainers To compare the efficacy of Tihan OD 175 (Flubendiamide 100 g/L + Spirotetramat 75 g/L) with that of the local check, Thunder OD 145 (Imidacloprid 30g/L + betacyfluthrin-13.5 g/L To determine the effect of the treatments on the natural enemies and other non- target arthropods MATERIALS AND METHODS The experiment was conducted in the field at KARI-Mwea during the 2014/15 cotton growing season. Cotton variety, HART 89M was planted in October, 2014. One seed per hill was planted at a spacing of 30 cm within rows and 100 cm between rows, the crop was planted in plots measuring 5 by 5 m and arranged in a randomized complete block design with 4 replicates. All the plots were separated by a 2 m path between blocks and 1 m path between plots. The crop was grown under rain fed conditions using the currently recommended agronomic practices. Cotton arthropod pests were allowed to infest the crop naturally. To avoid bias during data collection all treatments were marked with number codes written on metallic labels and placed in each plot. Foliar treatments were applied on two weekly basis from 5th March, 2015 to 18th April, 2015. Foliar treatments used in the experiment included:- Treatments: Product Active Ingredient Product Rate/ha Tihan OD 175 Flubendiamide 100 g/L + Spirotetramat 75g/L 0.15 L Tihan OD 175 Flubendiamide 100 g/L + Spirotetramat 75g/L 0.20 L Tihan OD 175 Flubendiamide 100 g/L + Spirotetramat 75g/L 0.25 L Thunder OD 145 Imidacloprid 100 g/L+ betacyfluthrin 45 g/L 0.30 L Untreated Control Nil - 285 The foliar sprays were applied using a 15 L knapsack (solo) sprayer fitted with a hollow conical nozzle. During application, the crop was totally covered with the spray solution. Tihan OD 175 g/L was tested against cotton mealybugs, aphids, African bollworm and thrips. The effect of Tihan OD 175 g/L against non-target insects (lady bird beetles and attendant ants) was also assessed. Pre-treatment data was taken one day before the first spray and post-treatment data was taken on two weekly interval starting from 13th March, 2015 to 25th April, 2015. During sampling, aphid counts were done on the under side of the leaves taken from each of 10 randomly selected plants in each plot. Foliar pests were counted from the topmost 5 mainstem node leaves. During sampling the numbers of bollworm larvae per plant were recorded. Cotton mealybugs were collected from three randomly selected twigs (10 cm long) from five tagged plants in each plot. The three twigs were cut off the plants and put in a sampling polythene bag and tied properly to prevent the mealybugs and the attendant ants from escaping. The polythene bags containing the mealybugs and ants were taken to the laboratory and the number of mealybugs and attendant ants were recorded in well designed data sheets. Data on mealybugs, aphids, thrips and African bollworm was transformed using √x+0.5 (Montgomery, 1976) and subjected to 2- way analysis of variance (ANOVA). To determine the effect of different rates of Tihan OD 175 g/L and Thunder OD 145 at 0.30 L ha-1 as compared with the untreated control, was done by use of, SAS statistical package (SAS, 2015). Mean separation was done by LSD test at p=0.05 level of significance. RESULTS AND DISCUSSION Table 1-7 show the efficacy of Tihan OD 175 at 0.15, 0.20 and 0.25 L ha-1 and local check Thunder 145 OD at 0.30 L ha-1 as compared to the untreated control. Cotton aphid (Aphis gossypii) Results presented in Table 1 show that there was no significant difference in aphid population among all the treatments prior to insecticide application. However, cumulative aphid counts through out the experimental period show that there was no significant difference in aphid population recorded in plots treated with Thunder 145 OD at 0.30 L ha-1 and those treated with Tihan 175 OD at 0.15, 0.20 and 0.25 L ha-1 with each recording 5.75, 11.61, 7.34 and 10.32 aphids per plant respectively. However, the untreated control plots recorded significantly high aphid populations of 19.39 aphids per leaf as compared to Thunder 145 OD at 0.30 L ha-1 and Tihan OD 175 at 0.15, 0.20 and 0.25 L ha-1 which recorded 5.75, 11.61, 7.34, 10.32 aphids per leaf respectively. 286 Table 1: Treatment means showing the efficacy of Tihan OD 175 against cotton aphids, Aphis gossypii Pre- treatment Weeks after 1st Pesticide application Treatment Rate/ha 0 3 4 5 6 7 Pooled Means (√x+0.5) Aphids/leaf (x) Square root (x+0.5) Tihan OD 175 0.15 L 4.07a 4.00a 3.75a 2.37bc 5.34a 1.13bc 3.48ab 11.61 Tihan OD 175 0.20 L 2.33a 1.56b 1.91a 2.64bc 5.75a 2.48bc 2.80b 7.34 Tihan OD 175 0.25 L 3.19a 2.20ab 3.17a 3.16ab 6.25a 2.85b 3.29b 10.32 Thunder 145 OD 0.30 L 2.19a 1.34b 2.91a 1.19c 4.15a 0.71c 2.50b 5.75 Untreted control - 2.19a 4.3a 4.06a 4.61a 5.46a 6.82a 4.46a 19.39 LSD (p=0.05) - 1.96 2.34 2.56 1.75 2.63 2.1 1 P-Value - 0.221 0.044 0.454 0.013 0.548 0.0001 0.002 CV (%) - 46.45 57.92 53.65 41.6 32.38 49.81 54.46 Means in the same column followed by the same letter are not significantly different by LSD test at p=0.05 level Note: x=Number of aphids/leaf African bollworms (Helicoverpa armigera) Results in Table 2 show that there was no significant difference in African bollworm population density before the application of Tihan OD 175 (0.15, 0.20L, 0.25 L ha-1) and Thunder 145 OD at 0.30 L ha-1. The pooled means show that Tihan OD 175 (0.15L, 0.20L, 0.25 L ha-1) and Thunder 145 OD at 0.30 L ha-1 recorded significantly low bollworm populations as compared to the untreated control plots with each recording 0.00, 0.06, 0.00, 0.006 and 0.21 bollworms per plant respectively. However, there was no significant difference in bollworm populations in plots treated with the 3 rates of Tihan OD 175 (0.15L, 0.20L, and 0.25 L ha-1) and Thunder 145 OD at 0.30 L ha-1. 287 Table 2: Treatment means showing the efficacy of Tihan OD 175 against African bollworm, Helicoverpa armigera Pre- treatment Weeks after 1st Pesticide application Treatment Rate/ha 0 3 4 5 6 7 Pooled Means (√x+0.5) bollworms /plant (x) Square root (x+0.5) Tihan OD 175 0.15 L 0.71a 0.71a 0.71a 0.71a 0.71a 0.71a 0.71b 0.00 Tihan OD 175 0.20 L 0.71a 0.71a 0.71a 0.71a 0.71a 0.71a 0.75ab 0.06 Tihan OD 175 0.25 L 0.71a 0.71a 0.71a 0.71a 0.71a 0.71a 0.71b 0.00 Thunder 145 OD 0.30 L 0.71a 0.71a 0.71a 0.71a 0.71a 0.71a 0.75ab 0.06 Untreted control - 0.71a 0.84a 0.71a 0.71a 0.71a 0.84a 0.84a 0.21 LSD (p=0.05) - 0 0.17 0 0 0 0.17 0.1 P-Value - 0 0.736 0 0 0 0.43 0.068 CV (%) - 0 15.15 0 0 0 15.51 24.32 Means in the same column followed by the same letter are not significantly different by LSD test at p=0.05 level; Note: x=Number of bollworms/plant 4.3 Thrips Results presented in Table 3 show that no significant difference in thrips population density was noted before and after the application of the test pesticide. However, pooled means of thrips population density for the whole experimental period indicate all the 3 rates of Tihan 175 OD and also the local check, Thunder 145 OD at 0.30 L ha-1 had no significant effect in the management of thrips on cotton. 288 Table 3: Treatment means showing the efficacy of Tihan OD 175 against thrips Pre- treatment Weeks after 1st Pesticide application Treatment Rate/ha 0 3 4 5 6 7 Pooled Means (√x+0.5) thrips /leaf (x) Square root (x+0.5) Tihan OD 175 0.15 L 1.64a 1.39a 1.43a 1.80ab 2.69ab 2.04a 1.88a 3.03 Tihan OD 175 0.20 L 1.00a 1.06a 0.93a 1.92ab 1.71b 1.85a 1.81a 2.78 Tihan OD 175 0.25 L 0.84a 1.00a 1.06a 1.13b 1.90b 1.76 1.66a 2.26 Thunder 145 OD 0.30 L 1.00a 0.84a 0.93a 2.39a 3.38a 2.03a 1.97a 3.38 Untreted control - 1.39a 1.06a 0.84a 1.38ab 2.41ab 1.22 1.70a 2.39 LSD (p=0.05) - 1.22 0.99 0.84 1.15 1.4 1.36 0.58 P-Value - 0.631 0.826 0.611 0.214 0.139 0.699 0.836 CV (%) - 69.19 61.73 53.93 44.4 38.52 50.64 61.09 Means in the same column followed by the same letter are not significantly different by LSD test at p=0.05 level Note: x=Number of thrips /leaf Mealybugs Results presented in Table 4 show that no significant difference in mealybug population density was recorded in all the treatment plots including the untreated controls before the application of test products. However, cumulative pooled means indicate that a significantly low population of 2.16 mealybugs per leaf was recorded in plots treated with Tihan at 0.25 L ha-1 as compared to the untreated control plots which recorded a significantly high population of 3.34 mealy bugs per leaf. On the other hand, unusually high mealybug population of 5.07 mealybugs/leaf was recorded in plots treated with Thunder 145 OD at 0.30 L ha-1. The possible explanation for this is that Thunder 145 OD at 0.30 L ha-1 may have killed all the natural enemies resulting in tremendous increase of mealybug populations. 289 Table 4: Treatment means showing the efficacy of Tihan OD 175 against mealybugs Pre- treatmen t Weeks after 1st Pesticide application Treatmen t Rate/h a 0 3 4 5 6 7 Pooled Means (√x+0.5 ) mealybug s /leaf (x) Square root (x+0.5) Tihan OD 175 0.15 L 2.10a 2.11 a 2.06 a 1.71a b 2.21ab 1.61 b 2.03b 3.62 Tihan OD 175 0.20 L 1.79a 1.88 a 1.65 a 1.35b 1.67ab c 1.42 b 1.76bc 2.60 Tihan OD 175 0.25 L 2.20a 1.88 a 1.42 a 1.42b 1.47bc 1.06 b 1.63c 2.16 Thunder 145 OD 0.30 L 2.11a 2.42 a 2.12 a 2.31a 2.32a 2.69a 2.36a 5.07 Untreted control - 2.20a 2.06 a 1.85 a 2.04a b 1.45c 1.54 b 1.96b 3.34 LSD (p=0.05) - 0.69 0.69 0.73 0.79 0.75 0.76 0.25 P-Value - 0.699 0.46 4 0.27 1 0.095 0.069 0.005 <0.0001 CV (%) - 21.87 22.1 5 26.5 8 29.86 27.44 30.36 25.95 Means in the same column followed by the same letter are not significantly different by LSD test at p=0.05 level Note: x=Number of mealybugs /leaf Attendant ants Table 5 show the effect of Tihan 175 OD attendant ants. Results obtained show that a significantly high population density of attendant ants were recorded in plots treated with Tihan 175 OD at 0.15 L, 0.20 L and 0.25 L ha-1 and also in the untreated control plots with each recording 31.76, 19.66, 15.58 and 10.39 ants/leaf respectively. The results indicate that all the 3 rates of Tihan 175 OD used (0.15 L, 0.20 L and 0.25 L ha-1 ) had no detrimental effect on attendant ants. On the other hand, Thunder 145 OD at 0.30 L ha-1 found to reduce the population of attendant ants to very low levels which is an indication that, it is harmful to the attendant ants which are beneficial insects in a cotton ecosystem. 290 Table 5: Treatment means showing the effect of Tihan OD 175 attendant ant populations Pre- treatmen t Weeks after 1st Pesticide application Treatment Rate/h a 0 3 4 5 6 7 Pooled Means (√x+0.5) ants/leaf (x) Square root (x+0.5) Tihan OD 175 0.15 L 3.34 a 4.08a 5.06a 5.47a 4.68 a 3.79a 5.68a 31.76 Tihan OD 175 0.20 L 4.05 a 3.92a 4.41ab 4.37a b 5.42 a 4.21a 4.49ab 19.66 Tihan OD 175 0.25 L 2.57 a 3.03a 3.84ab 3.15a b 5.25 a 5.36a 4.01b 15.58 Thunder 145 OD 0.30 L 2.82 a 3.37a 2.00b 2.54b 2.83 a 1.94b 2.65c 6.52 Untreted control - 4.71 a 3.46a 3.63ab 3.65a b 3.60 a 1.49b 3.30bc 10.39 LSD (p=0.05) - 4.74 3.68 2.82 2.65 2.65 1.82 1.25 P-Value - 0.07 8 0.973 0.253 0.211 0.22 8 0.002 <0.0001 CV (%) - 73.2 1 68.32 49.37 45.88 40.3 3 36.04 58.7 Means in the same column followed by the same letter are not significantly different by LSD test at p=0.05 level Note: x=Number of ants/leaf Ladybird beetles Table 6 show that prior to pesticide treatment there was no significant difference between the ladybird beetle populations in all the treatment plots including the untreated control. However, cumulative pooled means show that there was no significant difference in the number of ladybird beetles recorded in plots treated with Tihan at 0.25 L ha-1 and the untreated control plots with each recording 0.67 and 1.24 ladybird beetles per plant respectively. This is an indication that Tihan at the rate of 0.25 L ha-1 is not harmful to the ladybird beetles which are important natural enemies of a wide range of cotton pests including and not limited to aphids, thrips and mites. 291 Table 6 Treatment means showing the efficacy of Tihan OD 175 against ladybird beetles Pre- treatme nt Weeks after 1st Pesticide application Treatment Rate/ ha 0 3 4 5 6 7 Pooled Means (√x+0.5 ) beetles / plant (x) Square root (x+0.5) Tihan OD 175 0.15 L 0.71a 1.09a 1.18ab 0.84a 0.93a 0.84b 1.01b 0.52 Tihan OD 175 0.20 L 1.06a 1.15a 0.84b 0.84a 0.84a 0.71b 0.89b 0.29 Tihan OD 175 0.25 L 0.71a 0.84a 0.84b 0.84a 1.18a 0.71b 1.08ab 0.67 Thunder 145 OD 0.30 L 0.71a 0.71a 0.84b 0.97a 1.06a 0.71b 0.97b 0.44 Untreted control - 0.97a 1.18a 1.59a 0.93a 1.28a 1.22a 1.32a 1.24 LSD (p=0.05) - 0.35 0.48 0.66 0.46 0.79 0.17 0.26 P-Value - 0.132 0.203 0.107 0.956 0.751 <0.000 1 0.015 CV (%) - 27.84 32.16 41.48 34.84 49.31 13.62 45.78 Means in the same column followed by the same letter are not significantly different by LSD test at p=0.05 level Note: x=Number of beetles/plant 4.7 Plant biomass, life plants, damage score, total seed cotton and percentage (% BR) Table 7 show the Biomass of cotton plants taken at harvest. Results presented show significantly, low biomass 7.00 kg per plot in Thunder 145 OD and significantly high Biomas in plots treated with Tihan 175 OD at 0.15, 0.20 and 0.25 L ha-1 and the control with 11.00, 11.25, 12.25 and 10.25 kg/plot respectively. In reference to damage, the highest score was recorded in plots treated with Thunder 145 OD at 0.30 L ha¹ while significantly low mealybug damage score was recorded in plots treated with Tihan 175 OD at 0.20 and 0.25 L ha-1 with each recording a score of 0.25 while slightly higher damage score were recorded in Tihan 175 OD at 0.15 L ha-1 and the untreated control with a score of 1.50 and 1.25 respectively. 292 Results presented in Table 11 show that there was no significant difference in the number of bolls per plant, total seed cotton yield and the weight of grade BR seed cotton between all the treatments. Table 7: Treatment means showing effect of Tihan OD 175 on mealybug damage, Yield and quality of seedcotton Treatment Rate/ha Boll/plant Seedcotton (kg/ha) Wt of BR Tihan OD 175 0.15 L 57.5a 1298.8a 63.8a Tihan OD 175 0.20 L 50.3a 1171.7a 35.5a Tihan OD 175 0.25 L 41.8a 1171.2a 47.8a Thunder 145 OD 0.30 L 52.5a 1342.5a 24.5a Untreted control - 44.3a 1145a 35.0a LSD (p=0.05) - 42.98 1318.4 53.06 P-Value - 0.936 0.602 0.587 CV (%) - 57.5a 50.02 43.75a Means in the same column followed by the same letter are not significantly different by LSD test at p=0.05 level CONCLUSION The results obtained in the current study show that Tihan 175 OD at the rate of 0.20 - 0.25 L ha-1 in 500 litre of water is effective against the cotton aphids, African bollworm and the cotton mealybug. The findings of the study also reveal that the application of Thunder 145 OD at 0.30 L ha-1 encouraged the build up of the cotton mealybug. The use of Thunder in the control of mealybugs is highly discouraged. Tihan 175 OD at 0.20 - 0.25 L ha-1 was found to be effective in the managent of cotton aphids, African bollworm and the cotton mealybug. Tihan 175 OD at the rate of 0.20 - 0.25 L ha-1 was found to conserve the natural enemies namely ladybird bettles and attendant ants in cotton fields. REFERENCES Ikitoo E.C, Onzere B.B., Karani E.W, and Maobe S.N (1989). Cotton agronomy research in Eastern Kenya and Kerio valley. In KARI-NFRC Annual report pp 23-35. KENGAP Publishing (2010). The Kenya Agrochemical Products Handbook. ISBN 9966 7035 27, pp 113. Luttrel, R.G., Fitt, G.P Ramaldho, F.S and Sugonyaer E.G (1994). Cotton pest management. Part 1 world Perspective. Annual Review of Entomology 39: 517-526. Montgomery, D.C. (1976). Design and analysis of experiments. John Wiley & sons Inc. 418pp. Munro, J.H. (1987). Cotton tropical Agricultural Series. Longman Group, UK limited 436 pp 293 PCPB (2012). List of Pest Control Products registered for use in Kenya. Revised Edition, 2012, pp. 179. SAS, (2015). SAS University Edition. SAS Institute Inc. World Headquarters. SAS Campus drive. CARY INC.27513. www.SAS.COM 294 7.Parallel Session 6: Crop Management, Biotechnology and Climate Change 7.1 Physicochemical characterization of selected rice (Oryza sativa L.) Genotypes based on gel consistency and alkali digestion Lagat R Chemutai1 , Mawia A Musyoki1, Wambua F Kioko1, Njagi S Mwenda1, Karau G. Muriira2, and Ngugi M Piero1 1Department of Biochemistry and Biotechnology, Kenyatta University, Nairobi, Kenya 2Molecular Laboratory, Kenya Bureau of Standards, Kenya *Correspondence: Email address: roselagat76@gmail.com , +254712848915 ABSTRACT Knowledge of rice genetic diversity is necessary to ascertain the germplasm conservation and the development of improved rice genotypes with good quality traits through various breeding programs. Blending of rice with good and poor quality traits by unscrupulous traders in order to make enormous profits causes a negative impact on the consumer preference and the rice trade. Genetic diversity was determined using the 8 SSR markers that are tightly linked to the quantitative trait loci’s (QTLs). Minitab 17.0 software package was used to analyse the gel consistency test results while the alkali digestion values were determined using standard evaluation system by International Rice Research Institute. PowerMarker version 3.25 was used to determine allele number, gene diversity and the polymorphic information content (PIC). GenAlEx version 6.41 was used to determine the principal coordinate analysis (PCoA) and analysis of molecular variance (AMOVA). DARwin 6.0.12 statistical software was used to determine Jaccards genetic dissimilarity matrix and unweighted neighbour joining tree. The rice genotypes were classified into low, intermediate and high alkali digestion values and also soft, hard and intermediate gel consistency. The number of alleles per locus ranged from 2 to 4 with an average of 2.75. Polymorphic information content (PIC) ranged from 0.5224 (RM577) to 0.1411 (RM85) with an average of 0.3673. Gene diversity ranged from 0.5764 (RM577) to 0.1528 (RM85) with an average of 0.4181 with one rare allele was detected at RM577 loci. Genetic dissimilarity ranged from 0.9333 to 0.1818 with the least being observed between IR 54 and BS 370 while the highest, 0.9333 between Saro 5 and IR 2793. Unweighted neighbour joining tree showed three major clusters and subsequent sub-clusters hence effectively differentiating the Kenyan and Tanzanian genotypes based on gel consistency and alkali digestion. Clustering was complemented with the findings in the principal coordinate analysis. INTRODUCTION Rice (Oryza sativa L.) is a monocotyledonous plant belonging to the grass family (Gramineae) and the genus Oryza. Rice is the major staple food for the 57% of the world’s population and it provides approximately 23% of daily caloric intake. It was introduced into Kenya in 1907 from Asia and is an important cereal crop after maize and wheat in the country (1, 2). The quality preferences of rice consumers have resulted 295 in a wide diversity of the rice genotypes specific to different localities. The physicochemical properties comprises of gel consistency (GC), alkali digestion, and amylose content (3). Gel consistency is used in distinguishing cooked rice texture of high amylose genotypes. Genotypes are grouped based on the gel lengths into; hard (< 40 mm), medium (41–60mm), and soft (>61 mm) (7). The gene coding for gel consistency is located within the Wx locus (4, 5). Rice with hard gel consistency hardens faster than genotypes with soft gel consistency. Alkali digestion and gelatinization temperature have a significant correlation. Both traits are controlled by alk locus, which encodes soluble starch synthase (SSII). The use of DNA markers is widespread among plant geneticists because of the substantial amount of useful information that can be gathered from these markers. Simple sequence repeats (SSR) markers have been applied in various studies to study rice genetic diversity, marker assisted selection and in mapping of the QTLs (6,7,8). MATERIALS AND METHODS Plant material A total of 500g of the rice grains of twelve selected Kenyan and Tanzanian rice genotypes were collected from two repositories; Mwea Irrigation and Agricultural Development (MIAD) in Mwea, Kirinyaga county, Kenya and Kilimanjaro Agricultural Training Centre (KATC) in Moshi, Tanzania. Samples were brought to Kenyatta University Plant Transformation Laboratory for molecular analysis. IR 64 was selected as a model genotype due to desirable combination of attributes such as the intermediate alkali digestion values, amylose content, and soft gel consistency (9). Molecular and physicochemical methods used Alkali digestion was determined using methodology described by (10). Gel consistency in rice genotypes was determined using protocol described by (11). Genomic DNA was extracted from leaf samples using cetyltrimethyl ammonium bromide (CTAB) extraction protocol method. SSR markers identified from available Gramene database (http://www.gramene.org). The PCR reactions were was carried out in 200 µl thin- walled PCR tubes in a thermal cycler (Applied Biosystem® 2720). Electrophoresis was carried for the PCR products in 2% agarose gel alongside 3μl 100bp DNA ladder. RESULTS Alkali digestion Based on the degree of alkali digestion observed, the rice genotypes were classified into three groups; low, intermediate and high alkali digestion values. Genotype with low degree of alkali digestion value was Red Afaa while four genotypes had intermediate alkali digestion namely; IR 2793, BS 217, BS 370 and IR 64. High alkali digestion was observed in 7 genotypes; Saro 5 and BW 196 had an alkali digestion value of 6.0 while Wahiwahi, ITA 310, IR 54, Kahogo and Kilombero had an alkali digestion value of 7.0 (Table 1). 296 Gel consistency (GC) Based on GC values shown in Table 4, the average GC values ranged from 31.50 mm in ITA 310 to 99.5 mm in IR 2793. Genotype BS 217 had the second highest GC values after IR 2793 with no significant differences amongst their mean values. There was significant difference between genotypes as shown in Table 2below. Assessment of SSR profiles A total of 22 alleles were detected at the loci of the 8 SSR markers across the 12 genotypes. The number of alleles per locus ranged from 2 in RM 341, RM 539 and RM 85 to 4 in RM 577, with an average of 2.750. The highest gene diversity was observed in RM577 at 0.5764, while RM85 had the lowest gene diversity of 0.1528 with an average of 0.4181. Polymorphic information content (PIC) showed an average of 0.3673 with RM577 showing the highest PIC value (0.5224) and RM85 showing the lowest values of 0.1411 (Table 3). A rare allele was observed in Marker RM 577 loci at 400bp (Table 3; Figure 1). Genetic dissimilarity A dissimilarity matrix based on the Jaccard’s dissimilarity index was used to determine the levels of relatedness of the 12 rice genotypes based on gel consistency and alkali digestion. The genetic dissimilarity ranged from 0.9333 to 0.1818. Saro 5 and IR 2793 had the highest genetic distance of 0.9333 while IR 54 and BS 370 had the least genetic distance of 0.1818 (Table 4). Clustering of rice genotypes Rice genotypes were clustered into major clusters and subsequent sub-clusters. Bootstrap values showed the confidence limits of the phylogenetic branches, where the sub cluster IIB and sub cluster IIA1 had the highest bootstrap value at 56%. The lowest bootstrap value of 19% was attained in major cluster II (Figure 2) Analysis of molecular variance (AMOVA) From the total variation, 96% (P<0.001) of variation was distributed within population while 4% (P<0.001) was distributed among populations (Table 3 below) Principal coordinate analysis (PCoA) The two dimensional plot shows that the first principal coordinate accounts for 29.43% of the total variation while the second coordinate accounts for 19.84% of the total variation. Genotypes were distributed in all the 4 quadrants (Figure 3 below). 297 Tables and figures Table 1. Degree of degradation, alkali digestion values of the 12 genotypes Genotype Degree of degradation Alkali digestion value (ADV) IR 2793 Intermediate 4.0 BS 217 Intermediate 5.0 BS 370 Intermediate 5.0 BW 196 High 6.0 ITA 310 High 7.0 Saro 5 High 6.0 IR 64 Intermediate 4.0 Kilombero High 7.0 Red Afaa Low 1.0 Kahogo High 7.0 IR 54 High 7.0 Wahiwahi High 7.0 Table 2. Analysis of variance of the gel consistency values of the 12 rice genotypes Genotype Gel consistency values (mm) IR 2793 99.50 ± 0.50a BS 217 97.00 ± 1.29ab BS 370 75.25 ± 3.90bcd BW 196 67.00 ± 3.39cd ITA 310 31.50 ± 4.48e Saro 5 66.50 ± 3.48cd IR 64 67.75 ± 3.42cd Kilombero 69.50 ± 3.07cd Red Afaa 78.50 ± 9.54abc Kahogo 67.50 ± 5.50cd IR 54 53.25 ± 7.09de Wahiwahi 54.25 ± 3.33de 298 Gel consistency values are expressed as Mean ± SEM. Values followed by same superscript letter are not significant different at P ≥ 0.5.One-way ANOVA followed by Tukey’s test. Table 3. Genetic diversity profiles of 8 SSR markers used in the study Marker Major allele frequency Allele number Gene diversity PIC RM333 0.6667 3.000 0.4861 0.4235 RM314 0.7500 2.000 0.3194 0.2545 RM577 0.5417 4.000 0.5764 0.5224 RM501 0.6111 3.000 0.5093 0.4552 RM50 0.5833 3.000 0.5370 0.4694 RM539 0.8333 2.000 0.2639 0.2229 RM347 0.6667 3.000 0.5000 0.4491 RM85 0.9167 2.000 0.1528 0.1411 Mean 0.6962 2.750 0.4181 0.3673 Table 4. Jaccard’s coefficient genetic dissimilarity matrix of the 12 genotypes Genoty pes IR 2793 BS 217 BS 370 BW 196 ITA 310 IR 64 IR 54 Saro 5 Kilomb ero Red Afaa Kaho go BS 217 0.727 3 BS 370 0.692 3 0.333 3 BW 196 0.538 5 0.583 3 0.461 5 ITA 310 0.636 4 0.700 0 0.545 5 0.615 4 IR 64 0.636 4 0.818 2 0.666 7 0.875 0 0.600 0 IR 54 0.733 3 0.454 5 0.181 8 0.533 3 0.615 4 0.71 43 Saro 5 0.933 3 0.833 3 0.583 3 0.812 5 0.750 0 0.63 64 0.64 29 Kilomb ero 0.538 5 0.583 3 0.571 4 0.533 3 0.500 0 0.61 54 0.53 33 0.88 24 Red Afaa 0.600 0 0.733 3 0.625 0 0.500 0 0.571 4 0.66 67 0.58 82 0.83 33 0.2857 Kahogo 0.600 0 0.733 3 0.625 0 0.500 0 0.461 5 0.66 67 0.50 00 0.83 33 0.2857 0.2667 Wahiw ahi 0.615 4 0.666 7 0.642 9 0.500 0 0.866 7 0.78 57 0.50 00 0.87 50 0.5000 0.4667 0.466 7 299 Table 5. Summary of analysis of molecular variance (AMOVA) Source of variation Df SS MSD Estimated Variation % variation P- Value Among Populations 2 9.667 4.833 0.193 4% Within Populations 9 37.000 4.111 4.111 96% Total 11 46.667 4.304 100% <0.001 M A B C D E F G H I J K L Figure 1. SSR marker RM 501 showing polymorphism pattern of the 12 rice genotypes in 2% agarose gel. A - IR 2793, B - BS 217, C - BS 370, D - BW 196, E - ITA 310, F - Saro 5, G - IR 64, H - Kilombero, I - Red Afaa, J - Kahogo, K -IR 54, L – Wahiwahi. M- 100bp molecular ladder. 300b p 200b p 100b p 300 Figure 2. Unweighted neighbour joining dendrogram of the 12 rice genotypes based on gel consistency and alkali digestion using 8 SSR markers. Figure 3. Principal coordinate analysis of the 12 rice genotypes DISCUSSION Hard gel consistency observed in this study was due to formation of rigid rice gels, which occur as a result association of starch polymers in the aqueous phase. Amylose polymers leach when the starch granules are heated and they subsequently form networks once the gel cools (12,13,14). Physicochemical characterization of double haploid rice population also reported on the same classification (14). The intermediate gel consistency observed in this study could be due to effects of minor genes, for instance, gene interaction between waxy and pullulanase or waxy and BEIII genes (14). The rice grains that were highly affected by alkali solution had a high ADV. This observation could be attributed to the presence of amylopectin with a high number of short chains (A and B1) with minimal number of long amylopectin chains. Genotypes that were least affected by the alkali solution had a low ADV, which could be attributed to the presence of more long amylopectin chains (B2 and B3) than the short (A and B1) amylopectin chains (14). The intermediate alkali digestion could be attributed to the presence of intermediate number of both the long and the short chains. The number of alleles per locus detected varied per the marker used. This could be attributed to heterozygosity in the varieties studied. Similar findings have been reported in Pakistan basmati and non-basmati genotypes (27). However, these results were less than what is reported by other studies using different genotypes and SSR markers (28, 29). This could be as a result of use of different rice germplasm accessions and the selection of SSR markers. Gene diversity also referred to as the expected heterozygosity (He) attained had an average of 0.4181. This value was higher compared to 0.358 reported by (30) and lower than an average of 0.5513 reported in Kenyan and Tanzanian genotypes (16) and 0.68 in IR 2793 BS 217 BS 370 BW 196 ITA 310 IR 64 IR 54 Saro 5 Kilombero Red afaa Kahogo Wahiwahi 1 9 .8 4 % 29.43% Kenya Phillipines Tanzania I IIIIV II 301 Asian rice accessions (31). Evaluation of gene diversity based on the attributes of the genotypes, showed that improved genotypes had higher gene diversity value of 0.3729 compared 0.3167 of the landrace genotypes. Therefore, the low gene diversity is attributed to presence of a common gene pool amongst the landrace varieties unlike the improved genotypes where new genes have been introduced during crop improvement programs. The difference in gene diversity attained in this study with those in published literature could be attributed to the use different set of primers and rice genotypes under study. Based on PIC classification, one marker (RM 577) was the most informative marker while 5 markers were reasonably informative markers. These informative markers can be utilised in marker assisted selection due to its ability to distinguish between genotypes. The mean PIC was less than 0.4680 observed by (33) and 0.31 in Indian basmati and non-basmati rice genotypes (34). The many rare alleles observed in a germplasm indicates that there is a unique source of genetic diversity amongst the genotypes. Presence of rare alleles are important in fingerprinting rice genotypes (36). The highest dissimilarity observed between genotypes could imply that these genotypes have an uncommon origin. The unweighted neighbour joining dendrogram was used in this study as it produces additive trees and does not assume identical evolutionary rates along its branches. The low bootstrap values attained in some branches could be attributed to the low redundancy in genetic dissimilarity dataset which is indicative of considerable variation within the genotypes (39). Hence, in order to understand the level of genetic variation contributing to the low bootstrap values, AMOVA analysis using the Euclidean distance matrix (40) was conducted within and among populations. The high within population variation in AMOVA table is confirmed by the large values attained in the dissimilarity matrix as well as the low bootstrap values that were obtained for each of the cluster nodes (41). The PCoA showed that genotypes that are located closer to one another have more similarity than those that are ordinated further away. CONCLUSIONS From this study, it can be concluded that physicochemical classification showed considerable variation based on gel consistency and alkali digestion. Molecular diversity shows that improved varieties had higher gene diversity compared to landrace genotypes. The PIC values obtained reveals that RM577 is the best marker for diversity studies on gel consistency and alkali digestion given it is highly informative compared to other markers. ACKNOWLEDGEMENTS I acknowledge Mwea Irrigation and Agricultural Development (MIAD) and Kilimanjaro Agricultural Training Centre (KATC) for providing the rice genotypes for study. Also my co-authors for their input into this study. 302 REFERENCES Somado, E.A, R. Guei R. and Keya S.O. (2008). NERICA. The New Rice for Africa- a Compendium. Africa Rice Center (WARDA), Rome, Italy. Ministry of Agriculture (2009). National Rice Development Strategy 2008-2018. Information and Documentation Services. Kenya Agriculture and Livestock Research Organization (KALRO), Nairobi. Siddiqui, S. U., Kummamaru, T., and Satoh, H. (2007). Pakistan rice genetic resources- I: Grain morphological diversity and its distribution. Pakistan Journal of Botany, 39(3), 841-848. He, Y., Han, Y., Jiang, L., Xu, C., Lu, J., and Xu, M. (2006). Functional analysis of starch synthesis genes in determining rice eating and cooking qualities. Molecular Breeding, 18(4), 277–290. Tian, R., Jiang, G. H., Shen, L. H., Wang, L. Q., and He, Y. Q. (2005).Mapping quantitative trait loci underlying the cooking and eating quality of rice using a DH population. Molecular Breeding, 15(2), 117–124. He, Y., Han, Y., Jiang, L., Xu, C., Lu, J., and Xu, M. (2006). Functional analysis of starch synthesis genes in determining rice eating and cooking qualities. Molecular Breeding, 18(4), 277–290. Tian, R., Jiang, G. H., Shen, L. H., Wang, L. Q., and He, Y. Q. (2005).Mapping quantitative trait loci underlying the cooking and eating quality of rice using a DH population. Molecular Breeding, 15(2), 117–124. Khush, G. S. (2005). What it will take to feed 5.0 billion rice consumers in 2030. Plant Molecular Biology, 59(1), 1-6. Cagampang G. B., Perez C. M., and Juliano B. O. (1973). A gel consistency test for eating quality of rice. Journal of Science of Food and Agriculture, 24(12), 1589– 1594. Graham, R. (2002). A proposal for IRRI to establish a grain quality and nutrition research center. Los Baños, IRRI. Laguna, 15. He, Y., Han, Y., Jiang, L., Xu, C., Lu, J., and Xu, M. (2006). Functional analysis of starch synthesis genes in determining rice eating and cooking qualities. Molecular Breeding, 18(4), 277–290. Ritika, B. Y., Khatkar, B. S., and Yadav, B. S. (2010). Physicochemical, morphological, thermal and pasting properties of starches isolated from rice cultivars grown in India. International Journal of Food Properties, 13(6), 1339-1354. Biswas, S. K., and Juliano, B. O. (1988). Laboratory parboiling procedures and properties of parboiled rice from varieties differing in starch properties. Cereal Chemistry, 65(5), 417-423. Singh, R. K., Singh, U. S., and Khush, G. S. (1997). Indigenous Aromatic Rices, of India: Present Scenario and Needs. Agricultural Situation in India, 54, 491-496. Nachimuthu, V. V., Muthurajan, R., Duraialaguraja, S., Sivakami, R., Pandian, B. A., Ponniah, G., Gunasekaran, K., Swaminathan, M., Suji, K. K., and Sabariappan, R. (2015). Analysis of Population Structure and Genetic Diversity in Rice 303 Germplasm Using SSR Markers: An Initiative towards Association Mapping of Agronomic Traits in Oryza sativa. Rice, 8(1), 1-25. Sivaranjani, A. K. P., Pandey, M. K., Sudharshan, I., Kumar, G. R., Madhav, M. S., and Sundara, R. M. (2010).Assessment of genetic diversity among basmati and non- basmati aromatic rices of India using SSR markers. Current Science, 99(2), 221-226. Jain, S., Jain, R. K., and McCouch, S. R. (2004).Genetic analysis of Indian aromatic and quality rice (Oryza sativa L.) germplasm using panels of fluorescently-labelled microsatellite markers. Theoretical and Applied Genetics, 109(5), 965-977. Singh, Y. (2011). Molecular approaches to assess genetic divergence in rice. GERF Bulletin of Biosciences, 2(1), 41-48. Mohanty, S., Wassmann, R., Nelson, A., Moya, P., and Jagadish, S. V. K. (2013). Rice and climate change: significance for food security and vulnerability. International Rice Research InstituteDiscussion Paper, 49, (pp 14), Los Baños, Philippines. Schneider, S., Roessli, D., and Excoffier, L. (2000).Arlequin: a software for population genetic analysis. Genetics and Biometry Laboratory, University of Geneva, Geneva, Switzerland. Ansari, S. A., Narayanan, C., Wali, S. A., Kumar, R., Shukla, N., Kumar, S. R., and Dar, G. H. (2012).ISSR markers for analysis of molecular diversity and genetic structure of Indian teak (Tectonagrandis Lf) populations. Annals of Forest , 55(1),3. 7.2 Impact of nitrogen fertilizer application on groundwater nitrate and ammonium concentration in different farm systems in Ainabkoi, Uasin Gishu County. Kenya. Kitonga, l.M.1, Moturi, W.N. and Mwonga, S.M. and Taabu, F. 1Egerton University Correpondence: Email: lydiakitonga@gmail.com ,Tel: +254-726478884 ABSTRACT Excessive nitrogen fertilizer application and its mobility as nitrates (NO3-N) and ammonium (NH4-N) down the soil profile makes it the major contributor of non-point source pollution of groundwater. Nitrogen pollution of groundwater in Uasin Gishu is of particular concern because it is a high-potential maize agro-ecological zone, has high levels of fertilizer consumption and the dosage rates seem to be increasing. The objective of the study was to assess the potential risk of fertilizer application on ammonium and nitrate contamination of groundwater in different farms systems. Farm were purposively stratified into three non-overlapping strata (location) Ainabkoi, (large >40 acres), Kispinende (medium 10-40 acres) and Olare (small 2-10acres) mixed farm systems and well water evaluated in a longitudinal and diagnostic study. Fertilizer 304 application rates at planting ranged 22 – 44 Kg N /Ha and at top dressing 32-48 Kg N/ha CAN (26%). Groundwater NH4-N and NO3-N concentrations ranged from of 1.85-542.38µg/l and 1.61-246.18 µg/l respectively. There were no significant differences in the ammonium and nitrate concentrations between the farms, but the highest concentrations were found in wells within the large farm systems. There were positive linear relationships between N fertilizer and NH4-N (Y = 0.586x + 55.729 R2 = 0.27) and NO3-N (Y = 3.74x - 463.99 R2 = 0.31) groundwater concentration at topdressing. Therefore timing of N fertilizer application is a critical aspect in determining the potential to pollute groundwater. Keywords: Groundwater, Farm Systems, Nitrogen Fertilizers, Ammonium-nitrogen, Nitrate-nitrogen INTRODUCTION Throughout history the increase and intensive use of nitrogen-based fertilizers to achieve higher yields or productivity in agricultural systems has been well documented in Japan (Kumazawa, 2002 and Inoue et al., 2012), in China (Liu et al., 2005; Liu et al., 2005 and Ju et al., 2006), USA (Kraft, G.J. and Stites, W. 2003), Australia (Keating et al. 1996) and New Zealand (McLay, et al., 2001). It is accepted that the use of fertilizer nitrogen is expanding globally in order to meet the needs for food, fibre, and fuel of a growing world population (Zhang et al., 1996 and Snyder et al, 2009). Nitrate contamination of groundwater usually exceeds the standard for drinking water in the regions where N fertilizer application rates are above 500 N ha-1 and N use efficiency is less than 40% (Zhang et al, 1996). The main diffuse source of nitrates in groundwater has been identified as nonpoint source from excessive use of nitrogen based fertilizers in agricultural and livestock production systems (Novotny and Chesters, 1981, Spalding and Exner, 1993, Hill, 1996, Kumazawa, 2002 and Ju et al., 2006). In addition, large one-off applications of soluble fertilizer in excess of immediate crop needs can lead to substantial nutrient losses leaching (Crush et al 1997). This is because nitrates are soluble, highly mobile and loosely bound therefore leach out of the soil with percolating water making it unavailable for plant uptake. When N fertilizer application rates exceed the plant demand and the denitrification capacity of the soil, the nitrogen leaches out into groundwater usually as nitrate-N (Lord and Anthony, 2002 and Schroeder et al., 2004). According to Daberkow et al., (2001) the recommended N fertilizer rate is usually based on the amount of N required to meet the expected yield minus the amount of nitrogen in the soil, irrigation water and previous legume crop. The impact of fertilizer on groundwater has been witnessed in China which has had a strongly increasing trend towards the growth of crops of high economic value mainly vegetables and fruits over the last twenty years (Ju, et al 2006). Nitrogen fertilizer consumption has increased in China and is the main cause of increase in nitrate content in groundwater. The annual fertilizer application has risen gradually from 2.4 kg/HA 305 in the 1950s’ through to 211 N kg/Ha in the 1990’s in order to meet increasing food requirement by an ever increasing population and reducing arable land (Zhang et al., 1996). A study in Shandong province found that the average N fertilizer rate was 280 kg N ha-1 in winter wheat fields, 280 kg N ha-1 in maize fields, 1700 kg N ha-1 per crop in protected vegetable fields (plastic film greenhouses) and 848 kg N ha-1 in apple orchards (Ma, 1999). Townsend et al. (1996) reported high nitrate-N concentrations ranging between 12-60mg N L-1 in groundwater in Kansas caused by high application rates of N fertilizer to sugar beet fields. In the Chinese provinces of Beijing, Tianjin, Hebei, Shandong and Shanxi about 45% of 600 groundwater samples exceeded the WHO and European limit for nitrate in drinking water of 11.3 mg NO3-N (50 mg NO3-1 L-1), with the highest nitrate-N concentration reaching 11.3 mg L-1 (Zhang et al., 2004). The human health risk associated with consumption of water containing nitrate is due to the reduction of nitrate to nitrite in the human gut which causes methaemoglobinaemia or ‘blue-baby’ in infants and stomach cancers in adults (Spalding and Exner 1993, Hudak 1999, EPA 2002, (Lord and Anthony, 2002, Schroeder et al., 2004 and WHO, 2008). The general or broad fertilizer application rates for maize in Uasin Gishu county, Kenya is about 52-70 kg of DAP per acre at planting and 58 kg of CAN or 92 kg ASN per acre at top dressing depending on the soil fertility. In Kenya the fertilizer dose rates applied on maize fields during the main season has increased from 65 kg/acre in 1997 to 75 kg/acre in 2007 (Ariga et al., 2008) and the intensity of fertilizer application has increased dramatically on the intercropped fields. The planting season begins in the month of February. Farmers normally broadcast Di-ammonium Phosphate (DAP 18%) fertilizer at the rate of one 50 kg DAP bag /Acre, equivalent to 9 kg N/Acre (22 kg N/ha) during the last field harrowing procedure. Maize planting was done between March-April in 2012 and 2013 at the onset of the long rains. Top dressing was done when the crop was knee high or 45-60cm high, with Calcium Ammonium Nitrate (CAN 26%) at the rate of one 50 kg CAN bag /Acre equivalent to 13 kg N/Acre (32 Kg N/ha) as recommended by the Agriculture Extension officers. However it was observed that farmers tended to increase the recommended rate by 1½ to 2 times with the aim of increasing crop yields. Therefore excessive N fertilizer application is very common and may cause groundwater pollution. The focus today is worldwide is on sustainable agriculture which aims at developing environmentally friendly, ecologically sound and economically profitable agricultural management systems. This means that fertilizer management practices must be geared towards achieving high yields and high-quality crops with the efficient use of water, fertilizers, agrochemicals and energy with minimal environmental impact. The recommended threshold or maximum allowable concentration (MAC) by WHO and Government of Kenya (GOK) for nitrate-nitrogen in drinking water is 10 mg/l NO3-N (US EPA, 2000, WHO, 2008) and for Ammonium-Nitrogen is 0.5 mg/L NH4+-N (GOK, 2006). To evaluate/access and compare the nitrogen loading and groundwater 306 nitrate, nitrite and ammonium accumulation in different farm systems in Ainabkoi Sub- County of Uasin Gishu County, Kenya. METHODOLOGY The study was carried out in Uasin Gishu County which lies between longitudes 340 50´ East and 350 37´ West, and latitudes 00 03´South and 00 55´ North covering a total area coverage of the county is 3,345.2 km2 (UGCIDP, 2013 and UGDP, 1997). Ainaibkoi sub- county is considered an intensive and extensive agricultural area (Jaetzold and Schmidt, 1983). The main crops produced in the small farm sectors include maize, beans, wheat, vegetables, pyrethrum and flowers and livestock production. It is in the upper highlands zones 1 and 2 (UH1 and UH2), with a very long cropping season and intermediate rains which can be divided in to two variable cropping seasons. Purposive stratified random sampling technique was applied in selection of the representative farms that had access to a well within the farm or a centrally communal well, for evaluation of the groundwater sources. Each location was considered as a stratum of homogenous farms such that farms in Ainabkoi location were mainly large, family-generations-owned mixed farm systems and ranged more than 40 acres in size (>40 acres). In Kipsinende location farms were medium sized (10-40 acres) mixed farm system with privately owned wells while in Olare the mixed farm system were ranged from 2-10 acres in size with communally owned wells in Olare. A combination of: longitudinal survey, observational/descriptive and diagnostic study designs were used to evaluate the groundwater wells at different times throughout the crop production cycle over a period of two years (2011-2013). Farm systems, as defined by Garrity et al., (2012) are individual farms, with their own specific characteristics arising from variations in resource endowments and family circumstances which are translated into productive activities, and household consumption and decision making activities. Groundwater samples were collected at least every week from just before planting, in January to March, during planting in April and through to two weeks after topdressing in June-July and thereafter at least once a month until after harvesting in October December in 2012 and 2013. Sampling of groundwater was also done at least once a month during off production season in the months of November to January in 2013. Nitrate-nitrogen (NO3-N) and ammonium-nitrogen (NH4-N) were quantitatively determined by methods described in APHA, 1995. The primary data were summarized, organized and presented using the descriptive statistics, table means and graphs. Regression analysis was done by using the SAS software package (Version 6.12, 1997; SAS Institute, Inc., Cary, NC, USA). RESULTS AND DISCUSSIONS Maize production is the most important agronomic crop in Ainabkoi sub-county. The Agricultural Extension officers and Kenya seed company recommend fertilizer application rates for the region at 123.56 kg DAP/Ha (22.24 kg N/ha) at planting and 123.56-185 kg/ha CAN (26%) (32.12-48.2 kg N/ha) at topdressing. However, some farmers apply one and a half to twice the recommended rate of DAP and CAN. This 307 results in the application of 185.33-247.11 Kg DAP/ha (33.36-44.48kg N/ha) at planting and 185.33 kg CAN/ha (32.12 kg N/ha) at topdressing. The amount of fertilizer applied as total N (Kg N) in each farm system through the production cycle during planting and at top-dressing during the study period of 2012-2013 is shown in Table 1. This higher application rate was driven by farmer accessibility to higher disposable income that is common with the medium and the large farm systems. The fertilizer application rate in Ainabkoi county closely compares with that of sweet corn production in the state of Illinois, United States (US) which ranges from 84-336 kg N/Ha depending on the soil fertility (Mwanza et al 2011) and in Lithuania where the N fertilization rate for cereals and annual grasses was 90 and 180 kg/ha. Lack of enabling policies for the private industry, poor infrastructure (access to fertilizer), and low demand by fertilizer consumers, especially in rural areas of Sub-Saharan Africa (SSA), are three major causes of low consumption (Ariga et al., 2007). The total N applied during the maize production season varied significantly with the farm systems because the area under crop production varied in the different farms systems such that the large farm systems had crop acreage ranging from 8-21 ha, the medium sized from 1.2-10.1 ha and the small farm systems ranging from 0.6-2.8 ha during the study period of 2012 and 2013. However, these N-fertilizer rates are much lower than those reported in several parts of the world’s farm production areas, such as the average application of N to maize in China of 249 kg N/ha with an application range of 50-600 N/Ha (Cui, 2005). Farmers in Ainabkoi tend to apply more than the recommended fertilizer rates as is common with many other farmers in other parts of the world. Although the fertilizer recommendation rate for sweet corn was 168 kg N/ha and for potato was 258 kg N/ha in Wisconsin, USA, producers generally apply 250 kg N /ha and 297-357 kg N/ha respectively (Kraft and Stites, 2003) and the groundwater nitrate loading increased with increasing fertilizer. Reports have shown that between 1997 and 2007 fertilizer dose rates applied on maize fields in the area during the main season increased from 65 kg/acre to 75 kg/acre (Ariga et al., 2008). At the time of the study the rates had increase to 92 kg N /ha and the farmers express a desire to have more fertile fields through application of more fertilizer. Research has linked the evolution of increasing nitrates in groundwater to increasing fertilizer use in many parts of the agricultural world such as in Central Lithuania (Adomaitis et al., 2008); Vietnam (Kurosawa et al., 2008), China (Ju et al., 2006), Australia (Thorburn, et al., 2003); USA ((Kraft and Stites, 2003; Hallberg et al, 1989), and Japan (Kumazawa, 2002). Several research outputs have also confirmed that groundwater N pollution generally increases with the amount of N-fertilizer application (Jordan and Smith, 2005, Dunn et al., 2005, Liu et al., 2005, Thorburn, 2003, Kuo et al., 2001, Owens et al., 1999, Errebhi et al., 1998, Zhang et al., 1996, and Halberg, 1989). In Central Lithuania, the nitrate concentration in lysimeter water depended mainly on the nitrate fertilizers application rate (Adomaitis et al., 2008), such that fertilization of agricultural crops with 112 kg N/Ha increased nitrate concentration in lysimeter water by 67.1 mg /L to 112.1 mg /L and an N-fertilization rate of 224 kg N /ha increased nitrate concentrations by 139.1 mg/L to 187.2 mg/L. The results from this 308 study did not indicate such large increases in NO3-N concentration increase with increase in fertilizer nitrogen application due to the fact that fertilizer input levels in Ainabkoi of range of 54-92 Kg N/Ha are still comparatively low. The groundwater ammonium-N, and nitrate-N concentrations were determined at the time of planting in the months of March and April of 2012 and 2013 (Figure A. B and C). The ammonium-N was generally higher than nitrates-N concentrations in both 2012 and 2013 in the three farms systems. Figures D, E and F show the change in concentration of NH4-N and NO3-N in groundwater at the time of top dressing and two weeks thereafter in 2012 and 2013 in the large, medium and small mixed farm systems respectively. The average total amount of N fertilizer applied during top dressing in the large farm systems ranged from 390 N kg to 1053 N kg in 2012 and 2013 which averaged to 598 N kg in 2012 and 657 N kg in 2013. The NH4-N concentration gradually decreased from June through August. Nitrate concentration in groundwater increased from June through to August in both 2012 and 2013. However in July of 2012 there was a significant increase by 94% in NO3-N concentration in groundwater. In 2013 the concentration increased by 44% between June and July and then reduced in August. Analysis did not show a significant positive regression between the N kg/ha applied at planting and the concentration of N-compounds in groundwater in all the three farm systems. However regression analysis at two weeks after top dressing with CAN fertilizer, showed that there were positive linear relationships between fertilizer application at top dressing and NO3-N concentration (Y = 3.74x - 463.99 R2 = 0.31) (Figure G), and NH4-N concentration (Y = 0.0586x + 55.729 R2 = 0.27) in groundwater (Figure H). Therefore this indicates that there is a relationship between the timing of application of fertilizer nitrogen and the loading of N in groundwater. The nitrate-N concentration in groundwater in the different farm systems ranged from a minimum of 0.129 mg/L to a maximum of 0.246 mg/L. These values are less than the maximum allowable concentration of 10 mg/L. Ammonium concentration ranged from 26.43µg/L to 90 µg/L (0.026mg/L to 0.09 mg/L) which was lower than the maximum allowable limit of 0.5 mg/L (500µg/L) in Kenya (GOK, 2006) According to WHO (1998), nitrate concentration in groundwater is normally low, and can reach high levels due to agricultural runoff and infiltration which was also recorded in the study area. These low values in groundwater nitrate concentration were also recorded in areas of high intensive agriculture in New Zealand where the groundwater nitrate concentration averaged over three years of data collection, ranged from 0.02 mgNO3-N/ml. Inoue, (2012) reported a linear impact of nitrogen laod from fertilizer on groundwater NO3-N over a four year period (2000-2004) and attributed the large differential over the regression line to non-fertiliser sources of nitrogen, such as the livestock industry and human sewage. In this research this can be attributed to the same sources of livestock waste, pit latrines and decomposing plant organic matter in the vicinity of the wells. The NH4+ that is relatively immobile but is also carried down the soil profile. The concentration of NH4-N concentration levels in the groundwater seemed to be higher compared with the levels of NO3-N. This has also been reported in several studies 309 (Kurosawa et al., 2008) in farming villages in northern Vietnam (2002-2006), whereby 380-420 Kg N/Ha of inorganic N fertilizer application resulted in high NH4-N concentration in drinking water and low NO3-N concentrations (Kurosawa et al., 2008). These results from this study showed that although the N concentration in groundwater did not exceed the recommended maximum concentration, but the application of fertilizer nitrogen has the potential to pollute groundwater systems. This is a common practice in rain fed agriculture in several parts of the country such as Nandi South district tea plantations which is done during the rainy seasons in May and October and causes river eutrophication and nitrate pollution (Maghanga, et al. 2013). Several studies have reported that the rapid increase in nitrogen fertilizer application in order to achieve higher yields and profits in intensive farming is the main cause of increased nitrate concentration in groundwater such as in Japan (Kumazawa, 2002), in Platte Valley of Nebraska, USA (Daberkow et al., 2001) and in Northern China. (Zhang, et al., (1996). CONCLUSION The results from this research showed that none of the wells in Ainabkoi have NH4-N and NO3-N concentration in groundwater greater than the Kenya Maximum allowable limits for drinking water of 0.5 mg/l and 10 mg/L respectively. However the variability in N concentrations indicates these wells are vulnerable to impacts of fertilizer N rates and farm systems. The NH4-N concentration in groundwater was in the range of 3.25- 382 µg/l and 1.85-542.38 µg/l in 2012 and 2013 respectively. The highest NH4-N concentration in groundwater was in wells in the large farm systems with an average of 464 µg/l in 2012 and 2013. The NO3-N concentration in groundwater was in the range of 1.61-246.18 µg/l and 2.56-177.56 µg/l in 2012 and 2013 respectively. The groundwater NO3-N concentration was highest in the large farm systems with an average of 211.87 µg/l in 2012 and 2013. The potential for Ainabkoi farmers to increase the field outputs by improving on their fertilizer application rates exists as farmers chose to shift to horticultural production and increase in N fertilizer application in cereal production. The potential impact of fertilizer nitrogen application on the concentration of nitrate in groundwater brings into perspective the importance of timing and splitting of N fertilizer application. Timing of nitrogen inputs is critical because the concentration of nitrates in the drainage water depends on the balance and timing of nitrogen inputs and outputs to and from the soil and on the rates of nitrification and removal of nitrates from the soil solution (Brady and Weil, 1999). Farmers in Ainabkoi sub-county normally do a one-off application of top dress during the peak rainfall period of the production season, which is usually in July. This exposes the nutrients to runoff and leaching especially nitrogen due to its high solubility. Notably the, the causal relationship between nitrogen fertiliser and pollution is obscure and uncertain because of the unknown processes of transportation from non-point source to specific monitoring points (Inoue, 2012). This means that most changes occurring on farmland surface cannot be directly related to the concurrent status of 310 groundwater, but that can be related to past changes on farmland and groundwater flow dilution. RECOMMENDATION Agriculture must co-exist with environmental concerns hence there is need to quantify the leaching losses of N from a range of fertiliser practices over several seasons in order to develop environmentally and economical fertilizer application rates. The absence of any significant contamination of groundwater in this study does not preclude it occurring in the future. There is a gradual shift to more intensive agriculture with the production of cut-flowers and vegetables for local and export markets, along with increased N-fertilizer application rates. These are known prerequisite conditions for groundwater pollution. Therefore there is need to advocated for fertilizer management strategies, such as split fertilizer application (pre- and post-emergence) in order to synchronize N supply with crop seasonal demand and soil and plant tissue nutrient analysis before application of N fertilizer TABLE I: The total amount of inorganic nitrogen fertilizer (KG N) applied in individual farms and total averages in 2012 and 2013. Farm System Total Kg N Fertilizer 2012 2013 Planting Top- dressing Total Kg N Planting Top- Dressing Total Kg N Large Mixed 288 624 912 225 487.5 712.5 Large Mixed 360 780 1140 486 1053 1539 Large Mixed 180 390 570 198 429 627 Average N Kg 276 598 874 303 656.5 959.5 Medium Mixed 27 39 66 54 78 132 Medium Mixed 90 97.5 187.5 108 117 225 Medium Mixed 45 65 110 63 91 154 Medium Mixed 360 390 750 450 487.5 937.5 Medium Mixed 202.5 292.5 495 202.5 292.5 495 Average N Kg 144.9 176.7 321.7 175.5 213.2 388.7 Small mixed 27 39 66 27 39 66 Small mixed 36 52 88 27 39 66 Small mixed 18 26 44 18 26 44 Small mixed 13.5 19.5 33 22.5 32.5 55 Small mixed 63 91 154 63 91 154 Average N Kg 31.5 45.5 77 31.5 45.5 77 311 Figure A: Groundwater concentration of nitrogen ions at planting time (March & April) of 2012 and 2013 in the Large Mixed Farm Systems Figure B: Groundwater concentration of nitrogen ions at planting time (March & April) of 2012 and 2013 in the Medium Mixed Farm Systems Figure C: Groundwater concentration of nitrogen ions at planting time (March & April) of 2012 and 2013 in the Small Mixed Farm Systems. 0 20 40 60 80 100 120 2012 2013 2012 2013 Ammonium NITRATE µ g /L March April 0 20 40 60 80 100 2012 2013 2012 2013 Ammonium NITRATE µ g /L March April 0 20 40 60 80 100 120 2012 2013 2012 2013 Ammonium NITRATE µ g /L March April 312 Figure D. Nitrogen concentration in groundwater around the time of top dressing (June, July and August) in the large farm systems in 2012 and 2013. Figure E: Nitrogen concentration in groundwater during top dressing time in the medium Farm systems in 2012 and 2013 Figure F: Nitrogen concentration in groundwater during top dressing time in the small Farm systems in 2012 and 2013 0 50 100 150 200 250 2012 2013 2012 2013 NH4-N NO3-N µ g /L June July Aug 0 50 100 150 200 250 2012 2013 2012 2013 NH4-N NO3-N µ g /L june july Aug 0 20 40 60 80 100 120 140 160 180 2012 2013 2012 2013 NH4-N NO3-N µ g /L june july Aug 313 Figure G: Relationship between Nitrate Concentration in groundwater and Fertilizer two weeks after Top Dressing in 2012. Figure H: Relationship between Ammonium concentration in groundwater and fertilizer Nitrogen two weeks after top dressing in 2012 REFERENCES Adomaitis, T. V. 2008. Influence of mineral fertilizer on nitrogen leaching. Acta Agriculturae Scandinavica Section B-Soil and Plant Science, 58:199-207. APHA. (1995). Standard methods for the Examination of Water and Wastewater. 19th Edition. Washington D.C. U.S.A.: American Health Association. Ariga J, Jayne T.S., Kibaara B., and Nyoro J. K. 2008. Trends and Patterns Fertilizer Use by Smallholder Farmers in Kenya, 1997-2007. Tegemeo Institute of Agricultural Policy and Development. Crush, J. R., Catheart, S.N., Singleton, P., Longhurst, R.D., Bake, M. 1997. Proceedings of the NewZealand Grass. Ass. 59:55-58.Environmental Protection Agency (EPA), y = 0.0836x + 165.18 R² = 0.3124 120 140 160 180 200 220 240 0 200 400 600 800 1000 To ta l Fe rt ili ze r in p u t (k g N ) Nitrate (µg/l) y = 0.0586x + 55.729 R² = 0.2695 20 30 40 50 60 70 80 90 100 110 0 200 400 600 800 To ta l N F er ti li ze r (K g N ) Ammonium (µg/l) 314 (2002). List of Drinking Water Contaminants and MCLs. Health and Aesthetic Aspects of Water Quality. In: Kumazawa K. 2002. Nitrogen fertilization and nitrate pollution in groundwater in Japan: Present status and measures for sustainable agriculture. Nutrient Cycling in Agroecosystems, 63:129- 137. Garrity D., Dixon, J and Boffa J. 2012. Understanding African farming systems: Science and Policy. Prepared for Food security in Africa: Bridging Research and Practice. Australian International Food Security Centre. Jaetzold, R and Schmidt, H. 1983. Farm Management Handbook of Kenya. Vol. II. Natural conditions and Farm Management Information. Central Kenya (Rift valley and Central Provinces). Hallberg, G.R. 1989. Nitrate in Ground Water in the United States. ln Nitrogen Management and G roundwater Protection, Developments in Agricultural and Managed Forest Ecology 27. New York: Elesevier Science Publishing ComPanY Inc. Hill, A.R. 1996. Nitrate Removal in Stream Rioarian Zones. Journal of Environmental Quality, 25:743-755. Inoue, Y. D.-Z. (2012). Synoptic assessment of environmental impact of agricultural management: a case study on nitrogen fertiliser impact on groundwater quality, using a finescale geoinformation system. International Journal of Environmental Studies, Vol. 69(3):443-460. Ju .X.T., Kou, C.L., Zhang, F.S. and Christie, P., (2006). Nitrogen balance and groundwater nitrate contamination: Comparison among three intensive cropping systems in North China Plain. Environmental pollution, 143:117- 125. Keating, B.A., Bauld, J., Hillier, J., Ellis, R., Weier, K.L., Sunners, F., Connell, D., 1996. Leaching of nutrients and pesticides to Queensland groundwaters. In: Hunter, H.M. Eyles, A.G., Rayment, G.E. (Eds.), Downstream Effects of Land-use. Department of Natural Resources, Brisbane, pp. 151–163. Kraft, G.J. and Stites, W. 2003. Nitrate impacts on groundwater from irrigated-vegetable systems in a humid north-central US sand plain. Agriculture, Ecosystem and Environment100:63-74. Kumazawa, K. 2002. Nitrogen fertilization and nitrate pollution in gorundwater in Japan: Present status and measures for sustainable agriculture. Nutrient Cycling in Agroecosystems, 63:129-137. Kuo, S Huang, B. and Bembenek, R. 2001. Effect of winter cover crops on soil nitrogen availability, corn yield and nitrate leaching. The scientific world journal. 1: 22-29 Kurosawa, K. H. 2008. Excessive Level of Inorganic Nitrogen in Groundwater in the Intensively Farmed Areas of Northern Vietnam. Communications in Soil Science and Plant Analysis, 39:2053-2067. Liu, G.D., Wu, W.L. and Zhang, J. 2005. Regional differentiation of non-point source pollution of agriculture-derived nitrate nitrogen in groundwater in northern China. Agriculture, Ecosystems and Environment. 107: 211-220 315 Lord and Anthony 2002 Lord I., A. S. (2002). Agricultural nitrogen balance and water quality in the UK. Soil Use and Management, 18(4):363-369. Ma, L.S. and Qian, M.R. (1987). Contamination of nitrate and nitrite of water environment in Taihu Lake watershed. Environmental Science 8:60-65. In: Liu, G.D., Wu, W.L., and Zhang, J., (2005). Regional differentiation of non- point source pollution of agriculture-derived nitrate nitrogen in groundwater in northern China. Agriculture, Ecosystems and Environment, 107:211-220. Maghanga, J. K., Kituyi, J. L., and Kisinyo, P. O. and Ng’etich, W.K. 2013. Impact of nitrogen fertilizer applications on surface water nitrate levels within a Kenyan tea plantation. Journal of Chemistry. McLay, C. D. 2001. Predicting groundwater nitrate concentrations in a region of mixed agricultural land use: a comparison of three approaches. Environmental pollution, 115:191-204. Mwanza, L.M., Swiader J. and Mulwa R. M. 2011. Evaluation of SPAD Chlorophyll fluorescence for onsite Nitrogen assessment in drip fertigated sweet corn. Journal of Applied Horticulture. Vol.13(1). Novotny, V. and G. Chesters. 1981 .Handbook of Nonpoint Pollution: Sources and Management. Yan Nostrand Reinhold Environmental Engineering Services. New York: Van Nostrand Reinhold Company. Owens, L. B., Edwards, W. M., & van Keuren, R. W. 1999. Nitrate leaching from lysimeters under grass treated with ammonium nitrate or slow-release nitrogen fertilizer. Journal of Environmental Quality, 28, 1810_1816. SAS Institute Inc. SAS User’s Guide:Statistics, 1997. Edition. Cary, NC: SAS Institute Inc., 1997. Schroeder J.J., S. D. (2004). The effect of nutrient losses from agriculture on ground and surface water quality: the position of science in developing indicators for regulation. Environmental Science and Policy, 7:15-23. Snyder, C. B. (2009). Review of greenhouse gas emissions from crop production systems and fertilizer management effects. Agriculture, Ecosystems and Environment, 133:247-266. Sommer, R., Bossio, D., Desta, L. Dimes, J., Kihara, J., Koala, S., Mago, N., Rodriguez, D., Thierfelder, C and Winowiecki, L. 2013. Profitable and Sustainable Nutrient Management Systems for East and Southern African Smallholder Farming Systems Challenges and Opportunities: A synthesis of the Eastern and Southern Africa situation in terms of past experiences, present and future opportunities in promoting nutrients use in Africa. CIAT, CIMMYT, QAAFI and University of Queensland. Spalding, R.F., and Exner, M.E., (1993). Occurrence of nitrate in groundwater-a review. Journal of Environmental Quality, 22(3):392-402. Thorburn, P. B. (2003). Nitrate in groundwaters of intensive agricultural areas in coastal Northeastern Australia. Agriculture, Ecosystems and Envirnment, 94: 49-58. 316 Uasin Gishu District Development Plan, (1997-2001). Government of Kenya Printers. Kenya Rural Planning Development office of the Vice President and Ministry of Planning and National Development. Uasin Gishu County Integrated Development Plan (UGCIDP) 2013-2018. WHO. (2008). Guidelines for Drinking-Water Quality. Third Edition. Geneva, Switzerland: WHO Press. Zhang, W. T. (1996). Nitrate pollution of groundwater in Northern China. Agriculture, Ecosystems and Environment, 59:223-231. Zhang, W.L., Wu, S.X., Ji, H.J., Kolbe, (2004). The current situation and controlling manures of non-point source pollution in China. In: Ju .X.T., Kou, C.L., Zhang, F.S. and Christie, P., (2006). Nitrogen balance and groundwater nitrate contamination: Comparison among three intensive cropping systems in North China Plain. Environmental pollution, 143:117-125. 7.3 Supplementing households income through sericulture technology. Hannah Gitonga*1, Ann W. Mburu2, Muo Kasina1, Everlyn Nguku3 Joyce Wainaina1 1 National Sericulture Research Centre, P.O BOX 7816-01000 Thika 2 Kenya Industrial Research and Development Institute, P.O BOX 30650 (00100), Nairobi 3International Centre for Insect Physiology and Entomology P.O BOX 30772(00100), Nairobi *Corresponding author-Email: gitongaw0@gmail.com, Tel. No. +254-722892297 ABSTRACT Sericulture value chain comprises feeding of silkworms for cocoon production and post-cocoon activities for production of silk yarn. The silk industry by its nature is a cluster of many non-farm and on-farm activities and has a strong need for many forward and backward linkages. It is therefore suited as a tool for rural employment and poverty alleviation in populous developing countries. The venture, if embraced in Kenya can be an important catalyzer of economic development for the value chain stakeholders in contributing towards government’s strategy to grow the economy by 10% through the agriculture sector. A study was therefore conducted by Kenya National Sericulture Research Centre (NSRC) with an aim to establish the industry’s capacity in supplementing household incomes in Kenya. Silkworm farmers and silk processors were sampled from secondary data obtained from NSRC and International Centre of Insect Physiology and Ecology (ICIPE) records for the past ten years. Incomes realized by the farmers were analyzed for comparison with respective projections. It is projected that under ideal conditions, domesticated sericulture in an acre can generate about Ksh.200, 000 from cocoon sales in a year. It was however established that though 40% of the farmers sampled were able to realize these projections, there is huge discrepancies in cocoon productions realized by farmers rearing silkworms in the same 317 land sizes. To aid the farmers fetch better returns, the NSRC has devised a model where farmers collectively process their cocoons and get higher profit due to economies of scale and enable them remain actors in the sericulture value chain. It was therefore concluded that the sericulture industry has the potential to supplement household incomes in Kenya. However, proper motivation of farmers through customized monitoring and evaluation as well as adequate training among other measures, are essential facilitators. Key words; sericulture, farmers, cocoons, silk products, income INTRODUCTION The vast sericulture value chain comprises provision of feeding silkworms for cocoon production and post-cocoon activities for production of silk yarn. The silk industry by its nature is a cluster of many non-farm and on-farm activities and has a strong need for many forward and backward linkages. It is therefore suited as a tool for rural employment and poverty alleviation in populous developing countries (JAICAF, 2007 & NSRC Manual, 2008). The venture, if embraced in Kenya can be an important catalyzer of economic development for the smallholder, contributing to the government’s strategy to grow the economy by 10% through the agriculture sector (MTP II, 2013- 2017). The present global silk production is fluctuating between 70,000 to 90,000 Metric Tonnes and the demand for silk is annually increasing by 5%. These figures are bound to rise even more due to increasing demand for fashionable items as a result of fast changing fashion trends (Shekar and Hardingham, 1995). This scenario opens a window for the rural community in developing world to exploit the potential for silkworm farming for economic advancement. Silkworm farming is widely unexploited in Africa yet financial prospects are high given the global rising demand for silk items or finished products. However, due to low local silk productivity and products quality, the local market remains unsecured while the international market competitiveness is high (JAICAF, 2007). According to Nyamu et al. (2013), domesticated, bombyx-mori, and wild silkworm rearing are technologies that can be adopted in Kenya to help revamp the fizzling silk spinning and weaving cottage industry through provision of raw material. Given the right support by the government and institutions such as NSRC and ICIPE sericulture industry is highly prospective and can act as a tool for strengthening local economy (Mburu et al., 2013). The raw materials such as fibre and natural dyes produced locally are used in the local spinning, weaving and dyeing cottage industry hence reducing outflow of money. Sericulture by-products which include mulberry leaf products have important health benefits that supplement domestic incomes as well as contribute to the household health wellbeing. Under ideal conditions, domesticated sericulture in an acre can generate at least Ksh. 200,000 from cocoon sales in a year (NSRC manual, 2008). This will have generated employment for approximately 419 man days. To ensure the sector elicits farmers’ interests, an outlet for the farmers’ products has to be available. Secondary data collected from NSRC and ICIPE, the two major cocoon buyers and silk processors in Kenya, was used to establish the capacity of the sericulture value chain in supplementing households’ incomes in 318 Kenya. NSRC and ICIPE also offer monitoring and evaluation services to the silkworm farmers to ensure optimal returns on the investment. METHODS Secondary data from the various sericulture value chain stakeholders collected from NSRC and ICIPE records was used for this study. The period of the data gathered ranged from 2007 to date. Study Area The silkworm farmers whose data was used were from Murang’a, Nyeri, Kiambu, Kirinyaga, Meru, Kakamega, Narok, Nakuru, Busia amongst other sericulture practicing counties. Data Analysis The data was grouped into the following categories for ease of analysis 2.2.1 Economies of Silk Production in One Acre Table i gives a summary of the projected financial analysis of silk cocoon production and processing in one acre piece of land. Assumptions; When cocoons are dried there is an increase in profit by Ksh. 269,875 (448,189-178,314) in an acre of mulberry from 3rd year. On processing the dried cocoons, profit increases by Ksh. 80,371 (528,560-448,189) per acre in one year. Actual Versus Projected Returns of the Silkworm Farmers Data of 15 sample farmers and the respective land sizes under cocoon production against the dry weight cocoon production from each was given by (ICIPE) for a comparative analysis against the projected productions. Table ii gives the comparative variance between the actual and projected returns of cocoon production in different land sizes. Assumptions; Year one profits (Ksh. 55,574) projections were used for ideal income calculations for the different land sizes. Average Number of worms reared (0.25acre=50,000,0.125acre=30,000), 0.0625=10,000,0.0313=1,000 A scatter plot illustrating the discrepancies of cocoon production in the different land sizes was generated from Table ii as shown in Figure 1. A bar chart illustrating the comparative variance between the actual and projected income from the different sampled land sizes is also shown in Figure 2. Cocoon Production Versus Income Generated from Various Areas The data for cocoon production capacity versus the income generated by 23 sampled farmers from various counties in Kenya between years 2007 to 2010 was gathered from 319 NSRC records for analysis as shown in Table iii. A bar chart representing the data in Table iii was drawn for demonstration as shown in Figure 3. Market analysis of the various silk products was done based on the data of various buyers as recorded in NSRC from 2007 to date records. Table iv gives the amount of the various silk products sold by NSRC to processors and merchandisers while Figure 4 gives the respective proportions of each product. RESULTS AND DISCUSSION The projected financial analysis of the cocoon production as shown in Table i gives a clear indication of expected ideal returns from cocoon production and/or processing when all factors are held constant. However a significant variation in cocoon output from farmers with same land size under silkworm farming is noted by the wide spread of values around the regression line as shown in the Figure 1 linear scatter plot generated from Table ii. This is further supported statistically by the coefficient of determination that is above 50% (R2=56.18%) in the same Figure 1. Despite the variations in the cocoon output, it’s worthwhile noting that 40% of the sampled 15 silkworm farmers were found to have been able to realize and/or surpass the expected returns as shown in the actual versus expected cocoon output comparative chart in Figure 2, with the assumption that all the sampled farmers reared equal number of silkworms on the same land size. The variations in the cocoon outputs by the sampled farmers could be attributed to insufficient technical knowhow on mulberry management and silkworm rearing practices and capital investment as deduced by a survey conducted by Mburu et al. (2013). Farmers from Thika County have been found to have produced the highest cocoon output over the years sampled as shown in the Figure 2 comparative bar chart generated from Table iii. This performance can be attributed to favorable weather conditions as well as proximity to the sericulture monitors and evaluators, NSRC in Thika and ICIPE in Nairobi. Kirinyaga, Nanyuki, Meru and Mathioya farmers gave the least cocoon outputs. These discrepancies could have been contributed by harsh weather conditions, given they are all located in cold areas where silkworm rearing is challenging. Figure 4 gives an indication that silk fabric and cocoons have the largest market share respectively. However, the amounts sold are too low, as shown in Table iv, therefore indicating serious challenges in the marketing value chain which could be traced back to low product quality. 320 TABLES AND FIGURES 1. Table i: Projected silk production and processing financial analysis 2. Table ii: Actual and projected income from cocoon production in different land sizes Sample Cases Cocoon production (kg dry weight) Land size(Acre) Actual Income (ksh. 500/kg) Ideal Variance Case 1 16.8 0.25 8,400 13894 5494 Case 2 18.9 0.25 9, 450 13894 4444 Case 4 21 0.25 10,500 13894 3394 Case 5 25 0.125 12,500 6947 -5553 Case 6 10 0.125 5,000 6947 1947 Case 7 10 0.125 5,000 6947 1947 Case 8 18 0.125 9,000 6947 -2053 Case 3 8.4 0.0625 4,200 3473 -727 Case 9 5 0.0625 2,500 3473 973 Case 10 0.42 0.0625 210 3473 3263 Case 11 5 0.0625 2,500 3473 973 Case 12 10 0.0625 5,000 3473 -1527 Case 13 2.1 0.0313 1,050 1739 689 Case 14 8 0.0313 4,000 1739 -2261 Case 15 7 0.0313 3,500 1739 -1761 Item Description Cost (Ksh.) Year 1 Year2 Year 3 Cocoon Drying Cocoon drier (60kg capacity) 60,000 Total variable cost (fuel and labor) 8,930 51,180 67,840 Yield (42% fresh weight) 64,504 387,022 516,029 Return (Profit @ Ksh.1097/kg) 55,574 335,842 448,189 Raw Silk Reeling Multi end reeling machine (4kg dry cocoons/day: 1.52kg raw silk) 170,000 Pressurized cooking machine 270,000 Booking machine 15,000 Electric skein twisting machine 50,000 Total 565,000 Total variable cost(skilled labour) 23,930 139,180 185,840 Yield (Approx. 38% dry cocoons:2.63/kg raw silk) 88,000 536,000 714,400 Income @ Ksh. 4000/kg 34,070 396,820 528,560 321 3. Table iii: Different years cocoon production/income of farmers from different counties Year County Cocoon production(dry weight) Income (Ksh.) 2007 Muranga 8.82 7938 2007 Thika 8.4 7560 2007 Isinya 16.17 14553 2008 Meru 0.42 378 2008 Nanyuki 0.84 756 2008 Nakuru 3.36 3024 2008 Thika 3.36 3024 2008 Taita 0.525 472.5 2008 Mathioya 0.126 113.4 2009 Kirinyaga 0.672 604.8 2009 Narumoru 8.4 7560 2009 Meru 1.26 1512 2009 Laikipia 1.05 945 2009 muranga 4.41 3969 2009 Kijabe 1.68 1512 2009 Ikoloman 4.41 3969 2009 Thika 29.484 26535.6 2009 Karatina 1.47 1323 2010 Kijabe 10.122 9109.8 2010 Meru 5.04 4536 2010 Thika 22.26 20034 2010 Kirinyaga 5.04 4536 2010 Busia 1.47 1323 4. Table iv: Silk products sold Silk Product Amount (kgs) shells 12 Floss 21 Pupa 7 Cocoons 25 Degummed fibre 3 Raw silk 0.43 Fabric (m) 49 322 Figure 1: Cocoon production versus land size scatter plot Figure 2: Actual and ideal cocoon production income chart 0 2000 4000 6000 8000 10000 12000 14000 16000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Ideal income Actual Income R² = 0.5618 0 0.05 0.1 0.15 0.2 0.25 0.3 0 5 10 15 20 25 30 L a n d s iz e (A cr e) Cocoon Production (kg) Cocoon Production Vs Land size 323 Figure 3: Different farmers/areas dry cocoons income chart 7938 7560 14553 378 756 3024 3024 472.5 113.4 604.8 7560 1512 945 3969 1512 3969 26535.6 1323 9109.8 4536 20034 4536 1323 0 5000 10000 15000 20000 25000 30000 muranga Thika Isinya Meru Nanyuki Nakuru Thika Taita Mathioya Kirinyaga Narumoru Meru Laikipia muranga Kijabe Ikoloman Thika Karatina Kijabe Meru Thika Kirinyaga Busia 2 0 0 7 2 0 0 7 2 0 0 7 2 0 0 8 2 0 0 8 2 0 0 8 2 0 0 8 2 0 0 8 2 0 0 8 2 0 0 9 2 0 0 9 2 0 0 9 2 0 0 9 2 0 0 9 2 0 0 9 2 0 0 9 2 0 0 9 2 0 0 9 2 0 1 0 2 0 1 0 2 0 1 0 2 0 1 0 2 0 1 0 Income (Ksh.) Y ea r/ A re a Dry Cocoons Income 324 Figure 4: Proportions of silk products sold by NSRC (2007-2016) CONCLUSION AND RECOMMENDATION The sericulture industry in Kenya has therefore been found to have the capacity to supplement households’ incomes in Kenya as per the projected financial analysis. This has further been supported by the few farmers who have been found to realize these projections. It can therefore be recommended that farmers from different regions in Kenya be given customized attention by the industry monitors and evaluators, ICIPE and NSRC to motivate them realize fully the expected returns and hence supplement their household incomes. The government should also set clear sericulture - specific shells 10% Floss 18% Pupa 6% Cocoons 21% Degummed fibre 3% Raw silk 0% Fabric (m) 42% Silk products sold (2010-2016) 325 policies and strategies that aim at promoting the industry. The processors should also be equipped with the right training on how to produce competitive and adequate products that can easily be absorbed in the market. To address this, the NSRC has devised a marketing model where farmers collectively process the raw materials and offload products to the market with NSRC assistance. For example, payment to the farmer for their raw material will be based on the quantity and quality of the value added silk fibre. This model will ensure that the farmers get high profit due to economies of scale in processing, making it an incentive for the farmer to remain as an active actor in the sericulture value chain. The center also sells different silk raw materials and products to processors and merchandisers. This enables easier linkage of the silk worm farmers and silk consumers. Silk processors such as weavers will be assured of quality raw materials for advancement of the industry. ACKNOWLEDGEMENTS We are indebted to ICIPE for their kind sharing of the data that supplemented the data that was used to conduct this study. We are also grateful to the NSRC, for allowing us to use its data records for the purposes of this study. REFERENCES Japan Association for International Collaboration for Agriculture and Forestry (JAICAF). Sericulture in East Africa, Available from www.jaicaf.or.jp/publications/report- 2007_1_e.pd, Accessed: April 2013 National Sericulture Station, Manual guide to Sericulture practices in Kenya, (2008). Nyamu, K.R, Wainaina N.J. & Onyango A.J. Sericulture as a Business, Available from http://www.infonet-biovision.org/default/ct/719/products, Accessed: March 2013. Second Medium Term Plan, 2013 – 2017 “Transforming Kenya: Pathway to Devolution, Socio- Economic Development, Equity and National Unity”. Kenya Vision 2030. Mburu, A., Kisoryo, A., Kinyanjui J., & Waithaka, A. (2013). “Challenges and Prospects of Sericulture in Kenya”. A paper presented at the 2nd international Textiles and Consume Congress, Department of Textile Science, Faculty of Agro-Industry, Kasetsart University, Bangkok, Thailand, 28-29 October 2013. Shekar, P. and Hardingham, M. (1995). Sericulture and Silk Production Handbook, Intermediate Technology Publications, London. 326 7.4 Seed certification as a means of curbing emerging diseases: A study of Maize Lethal Necrosis disease in Kenya Peter Shango1*, Charles Onyango 1, James K. Oganda1, Isaac Macharia1, George Kariuki2 1. Kenya Plant Health Inspectorate Service (KEPHIS), P.O BOX 49592-00100, Nairobi 2. Department of Agriculture and Technology Kenyatta University *Corresponding author : Peter Shango (pshango@kephis.org). ABSTRACT Maize lethal necrosis disease (MLND) is an important constraint in maize production in sub-Sahara Africa that threatens food security and poses challenge in trade. It was first reported in Kenya in 2011 and has since spread to other countries in the region. The disease is caused by a combination of Sugarcane mosaic virus (SCMV) and Maize chlorotic mottle virus (MCMV), which are vectored by aphids and thrips, respectively. Maize is the main staple food in Kenya and therefore, with the advent of the MLND, there was need to establish mechanisms aimed at combating the spread of the disease. This resulted in the amendment of seed certification protocol which included testing of seed. In consultation with seed stakeholders, KEPHIS incorporated guidelines for MLND inspection in maize seed certification program where all maize seed crops were inspected three times including preliminary, first, second and third inspections. A seed sample was taken before seed dressing for laboratory test to ensure the lot is free from MLND. Imported seeds were also tested for MLND before being accepted into the country. Furthermore, there have been concerted efforts by breeders and researchers to develop and screen maize lines for resistance to MLND. Other strategies that have been put in place include control of vectors and use of certified seed which are free from MLND. Amendments in seed certification program that were implemented in Kenya have led to drastic reduction of MLND. It is further hoped that resistant maize lines from the breeding program will further support the effort to manage the disease. Keywords: Maize, MLND, Resistance, Screening, Seed certification, seed testing INTRODUCTION Maize is the most important cereal crop in both Sub-Saharan Africa and Latin America. It is an important staple food for more than 1.2 billion people in Sub Sahara Africa. All parts of the plant can be used for food and non-food products. In industrialized countries, maize is largely used as livestock feed and as a raw material for industrial products. According to 2007 FAO estimates, 158 million hectares of maize are harvested worldwide of which 29 million hectares are harvested in Africa, with Nigeria, the largest producer in SSA, harvesting 3%, followed by Tanzania. Maize is widely grown in Kenya and ranks highly in food security, meeting dietary preferences of many communities. 327 Although many production constraints have been reported on Maize, MLND has emerged as an important constraint in maize production in sub-Sahara Africa that threatens food security and poses challenge in trade (FAO June 2013). The disease is caused by a synergistic interaction of Maize Chlorotic Mottle Virus (MCMV) and Sugarcane Mosaic Virus (SCMV) or other cereal viruses like Maize Dwarf Mosaic Virus (MDMV) or Wheat Streak Mosaic Virus (WSMV) (Morales et al 1999) In Kenya, the disease has been reported to be caused by a combination of MCMV and SCMV although the two viruses independently are capable of causing a significant damage. Sugarcane Mosaic Virus has been in Kenya for a long time thus has not been a major threat in maize production. The entry of MCMV in 2011 has been the source of major threat to maize farming (Wangai et al 2012). Maize chlorotic virus (MCMV) is the only established member of the genus Machlomovirus in the family Tombusviridae (Fauquet et al.; 2005) which was first established in maize from Peru (Castillo and Hebert, 1974). MLND TRANSMISSION. MLND which is a combination of Sugarcane mosaic virus (SCMV) and Maize chlorotic mottle virus (MCMV),are vectored by aphids and thrips, rootworms and leaf beetles, respectively. Severe infection has been reported in areas where maize is grown continuously due to continuous availability of inoculum. The build-up of vector that transmit the virus is also evident during continuous cropping. MCMV has been reported to survive in maize residues in the field for long (Li et al., 2011). MCMV has also been reported to be either seed-borne where seed produced from infected plant may carry the virus or seed-transmitted. Although a seed transmission rate of 0.3% was been reported (Li et al., 2011; David et al., 2012), the effect is severely amplified when the small inoculum is taken up by vectors in the field. There has been no quantitative assessment of seed transmission of MCMV that has been published recently but evidence in Eastern Africa so far indicate that these virus among others can be carried through seed more so where seed production fields had high incidence of MCMV. A small source of inoculum of the virus in the field either from the seed surface contamination or seed-borne will have a severe effect in the entire field when the vectors takes up the virus and spread the virus in wider areas especially where no proper phytosanitary measures has been established to control the vectors. Whether the virus occurs on the outer surface of the maize seed or inside the maize seed, chances of the virus being transmitted to the new generation of crop in the field stand high (Stanley et al,; 1991). MLND FREE SEEDS PRODUCTION In Kenya, a concerted effort was initiated by the government through various institutions including both governmental and non-governmental bodies. Their target was to find ways of curbing the MLND menace that was threatening maize production. The team involved included higher learning institutions, research organizations, Ministry of Agriculture and KEPHIS. Among many strategies that were adopted, disease free seed production was taken up as one of the most viable target to ensure the 328 disease cycle interrupted.. This led to amendments in maize seed field inspections guidelines. MLND was given a higher weight and priority in scoring during inspection with a 0% tolerance level being adopted for field and laboratory stages. Seed dressing with appropriate systemic insecticides was also recommended for the control of vectors. Seed dressing was strategically scheduled to be done after final laboratory results have been released so as to minimize loss where the seed lot is tested and found positive. Such lots were recommended for milling to reduce the losses incurred by the seed companies.. Rejection during field inspection would enable the seed company to make a decision on whether the seed crop could be converted to commercial crop or even the maize could be sold for consumption while still green in the field. KEPHIS has been utilising these strategies to ensure MLND free seeds are availed to the market for both imported and locally produced seeds. Since as a general rule, the quality of your input ultimately determine the final yield, planting healthy, certified and treated seeds is generally the first step in production of healthy crop. When MCMV virus is introduced through seed to an area where it is has not been previously known to occur, the insect vectors immediately picks up the small inoculum and spread the virus rapidly if the virus is not diagnosed and infected crops rogued in time. In the long run, the virus will establish itself and management at later stages becomes difficult to achieve. In order to curb this menace, KEPHIS has from its mission of ensuring quality agricultural input strived to ensure that all maize seeds passes through rigorous tests to ensure the final quality that will reach the farmers is of high quality and disease free. The major objective of this work was to ascertain whether the measures that were adopted by KEPHIS through seed certification process are effective. The volumes as well as percentage of rejections at different levels in the certification chain were also established. MATERIALS AND METHODS This work has mainly relied on the general operations carried out in KEPHIS especially Nakuru Regional office to curb the spread of MLND. Activities ranging from site selection, seed treatment, field inspections and laboratory testing.. Most of the field activities observed during this study were in Central Rift Valley Kenya. Laboratory samples were taken from both field samples and also from import consignments at the port of entry. MLND SEED SAMPLING PROCEDURE Field inspection of maize crop after amendments of seed certification protocol is done three times. For 1 to 2.5 Hectares, 2000 maize plants are counted;. At final level inspection, MLND infected plant is not supposed to be present in the field during inspection. First inspection is done one month after planting, where a tolerance of not more than 1% is accepted but with recommendations for rouging infected plants. Second inspection is done six weeks after planting or two weeks after first inspection. 329 At this stage, the crop is almost flowering, de-tasseling and off type rouging underway. MLND infection of more than 0.9% is an automatic rejection. Rouging is recommended at this stage. Second inspection rouging is allowed for 0.9% infection and below. For 3rd and final inspection MLND tolerance is curbed at one per cent although the tolerance percentage is expected to be lowered to zero per cent at this stage. The 3rd and final inspection is done normally one week after 2nd inspection. During this inspection, whenever MLND is spotted in the field under inspection, an outright rejection is done, tolerance is pegged at zero percent. SAMPLING PROCESS FOR LABORATORY CONFIRMATION FOR MLND. Each seed crop after harvesting and shelling was sampled for MLND laboratory confirmatory test. For imported maize seeds, a sample was taken at the port of entry where the seeds are held while awaiting the outcome of the laboratory results. For locally produced seeds, immediately after shelling and before transportation, an inspector took a sample using a sampling probe. Labelling of the sample was done by the inspector filling all the sample details in the official KEPHIS SR9 forms before the sample being dispatched to the KEPHIS Molecular laboratory for testing. LABORATORY DIAGNOSIS OF MLND KEPHIS has equipped three molecular testing laboratories that carry out MLND diagnosis. These are the Molecular laboratory at Plant quarantine and Biosafety Station, KEPHIS headquarter molecular laboratory and Molecular Testing Laboratory at KEPHIS Nakuru. Official samples are received accompanied by KEPHIS Form SR9 forms which contain all the details of the sample. The information on the Form SR9 is captured in the database and the sample bags are coded with an identity number for use in the laboratory. Usually, 400 hundred seeds per sample are planted in sterile media (Current ISTA Rules 2016) and incubated at 200C to 30 0C for at seven days until the sample is at two leaf stage. The sample is then harvested by cutting the tips of each seedling of the sample into a labelled sample bag for crushing. RNA is extracted from the leaf using modified CTAB method. Positive control, blank (EB), and water are used as control in the process of real time PCR. The primers indicated in Table (i) are used. The results are collected basing on the CT Values of each sample which correspond to viral load in each sample. The laboratory results are communicated back to the merchant and the sampler advising whether to process or reject the seed. The crop rejected in the field depending on the age of the crop is not harvested until other seed crops have been harvested. 330 RESULTS Field Rejections Between 2013 and 2016 in Kenya are illustrated in Table (ii). Real-time PCR laboratory results since the year 2013 are illustrated in Table (iii) Tables and Figures Table i Primers for MCMV and SCMV MCMV-F 5’ – CCG GTC TAC CCG AGG TAG AAA - 3’ MCMV-R 5’ – TGG CTC GAA TAG CTC TGGB ATT T- 3’ MCMV-P 5’– CAG CGC GGA CGT AGC GTG GA- 3’ SCMV-F 5’ – CCA GGC CAA CTT GTA ACA AAG C- 3’ SCMV-R 5’ –CAT CAT GTG TGG ATA AAT ACA GTT GAA- 3’ SCMV-P 5’ -– TGT CGT TAA AGG CCC ATG TCC GCA ’ Table ii. Field rejections due to MLND Number Region Area rejected in Ha. 1 North Rift( Kitale Region) 154.7 2 Central Rift (Nakuru Region) 408 3 Embu 4.5 4 Nairobi 1.9 5 Kisumu 0.0 6 Mombasa (Bura and Taveta) 0.0 Total 569.1 Table iii: Real Time PCR Laboratory Results Since 2013 No. Period No. of MLND samples received No. of MLND samples Tested Negative No. of MLND samples Tested Positive 1 2014 1444 1406 38 (2.6%) 2 2015 2088 2040 48(2.3%) 3 2016 784 776 8 (1.02%) Total 4316 4222 91 (2.1%) DISCUSSION On average, one hectare of maize crop seed can yield up to 2,940Kgs. From the field rejections above, a total of 569Ha of seed crop have so far been rejected. These translate to 1,672, 860Kgs of maize seed crop that was rejected in the field. Normally, for small scale farmers, the seed companies package their maize seed in 2kg packet. A two 331 kilogram packet of maize seed retails at an average price of US $ 3.6. Working with the above figures projects to seed companies having lost up to US $ 3,011,148 from the field rejections alone. Calculations from the rejections done at the laboratory level, the 1.5M Kg of seeds that were rejected could otherwise have fetched up to US Dollars 2,700 000. In total on average, seed companies have lost more than US Dollars 5,711,148 this amount is excluding losses incurred from other cost of production that were already invested by this seed companies during the process of this seed production to the levels at which rejections were done. The results above include samples that were from both local and imported consignments sampled during the period. From the above results, 2.3% of total samples that were tested during the period between 2013 and 2015 tested positive for MLND. Each sample represents a seed crop harvested from a field of not more than 2.5Ha. On average, a seed crop would range between 5000kgs to 40,000kg of maize. An average of 1.5million Kilograms of MLND infected maize seed were prevented from reaching the farmers in Kenya during that period. This amount of seeds plus more others rejected during field inspection together would have provided a huge amount of inoculum for spreading MLND. This would have been a big blow to the fight against MLND especially where the seeds would have been purchased and planted in many parts of the country. Taveta and Bura have had no positive samples properly due to their geographical isolation from maize growing areas in Kenya. On the other hand, the central Rift has had the highest rejections considering that this is the area where the disease was first reported in Kenya. Furthermore, cultural practice for maize farmers in this region could have immensely contributed to this since small scale farmers had maize on their farms throughout the year. The climate in the central and south Rift could also favour the disease since it is almost evergreen throughout the year. Trend analysis of the above results indicates that the disease severity has been on the decline since 2012. CONCLUSION Embracing the idea of disease exclusion has been and is still the best ever approach in disease management MLND not being an exception. From the data above, a long stride was made in management of MLND in Kenya through the amendment of seed certification guidelines. Laboratory testing plays an important role in management of disease especially where many seed lots that were approved at field level failed at the laboratory stage. The overall multiplier effect of a single infected plant in a seed crop field is generally large when not controlled. Striving at ensuring high quality of agricultural input, if well implemented can ultimately help in ensuring food security globally. MLND is still a disease in Maize that still requires monitoring in irrigated field where there is no observation of closed seasons for maize. 332 RECOMMENDATION Evaluations for new maize varieties for resistance or tolerance to MLND should be augmented in management of MLND. Seed certification measures that have led to the reduction of the disease should be entrenched in the seed laws and implemented fully. Irrigation schemes and other seed growers should be guided to observe closed maize season to reduce disease inoculum. Phytosanitary measures should be observed strictly even for seed material to exclude MLND from countries where it is not known to occur. A full proof of laboratory test for MLND should be demanded at all border points to ensure the disease is managed and spread to disease free countries is controlled. National laboratories should be equipped to produce more efficient and accurate results. LIST OF ABBREVIATIONS Form SR9: Seed Regulation 9 ISTA: International Seed Testing Association KEPHIS: Kenya Plant Health Inspectorate Service MCMV: Maize chlorotic mottle virus MLN: Maize Lethal Necrosis MLND: Maize lethal necrosis disease PCR: Polymerase Chain Reaction SCMV: Sugarcane mosaic virus ACKNOWLEDGEMENTS I wish to acknowledge the assistance of KEPHIS Nakuru staff at large who provided the much needed information while writing this paper. Special thanks to the KEPHIS Muguga team at large for their help in provision of data. My special thanks to my wife Eddah for her personal editing of the work. Thank you all. REFERENCES Fauquet et al; 2005; Virus taxonomy; classification and nomenclature of viruses. Elsevier/Academic Press, Amsterdam Goldberg, K.B., Brakke, M.K., 1987. Concentration of maize chlorotic mottle virus in mixed infections with maize dwarf mosaic virus, strain B. Phytopathology 77, 162- 167 http://www.fao.org/fileadmin/user_upload/emergencies/docs/MLND%20Snapshot FINAL.pdf. I.P Adams et al., 2014. First Report of Maize Lethal necrosis disease in Rwanda. New Disease Reports 29, 22 htt:dx.doi.org/10.5197/j.2044-0588.2014.029.022 IITA http://www.iita.org/maize ISTA Rules 2016; Current version; CH-8303 Bassersdorf, Switzerland 333 K. Subramanya Sastry. 2013 Seed-borne Plant Virus Diseases K.S Sastry. 2013. Methods of combating seed-Transmitted Virus Disease Pg 186 Kumari, S., & Makkouk, K. (1996). Inactivation of broad bean stain comovirus in lentil seeds by dry heat treatment. Phytopathologia Mediterranea, 35(2), 124-126. Retrieved from http://www.jstor.org/stable/42685255Morales, F.J., Arroyave,J.A., Catillo, J., Leon, C.D., 1999. Cytopathology of Maize chlorotic mottle virus in Zea mays L. Maydica 44, 231-235 Scot Nelson, James Brewbaker, and John Hu; 2011; Maize Chlorotic Mottle. Journal paper 79 of College of Tropical Agriculture and Human Resources, University of Hawai’i at M.G.Redinbaugh, Z.M.Kinyua, D.W. Miano,P.K. Leley,M.Kasina,G.Mahuku, Yong jiang Zhang et al; 2010; Real-time TaqMan RT-PCR for detection of maize chlorotic mottle virus in maize seeds Manoa Stanley et al.1991 Seed Transmission of Maize Chlorotic Mottle Virus. Journal paper 9285 of the Agricultural Research Division, University of Nebraska,Lincoln. Wangai,A.W.;M.G.Redinbaugh,Z.M.Kinyua,D.W.Miano,P.K.Leley,M.Kasina,G.Mahuku , K.Scheets, and D.Jeffers. 2012. First report of Maize chlorotic mottle virus and Maize Lethal necrosis in Kenya. Plant Disease APS Journals Number 10 page 1582 7.5 Mango pollinators and their spatial differences across a forest-farm gradient in Kilimambogo, Machakos County, Kenya Magero1,4T, N Dharani1C Odhiambo2, W Kinuthia3 and M Kasina4 1 Kenyatta University, School of Environmental Studies P.O.Box 43844-00100, Nairobi 2 The Technical University of Kenya, Department of Ecology and Conservation Biology, P.O. Box 52428 00200, Nairobi 3 National Museums of Kenya, Department of Zoology, P.O.Box 40658-00100, Nairobi 4National Sericulture Research Centre, KALRO, P.O. Box 7816-01000 Thika Correspondence: Muo.Kasina@kalro.org Theme: Crop Improvement, Biotechnology and Climate Change ABSTRACT Pollination is critical for mango (Mangifera indica) fruit set and formation of quality fruits. As such, it increases tree productivity, enhances and secures farmer incomes. However, considering that pollinators are not managed in Kenya, evidence showing their status is critical to guide in decision making for pollination management. This 334 study was carried out to determine potential mango pollinators and their spatial differences across a forest-farm gradient in Kilimambogo, Machakos County in Kenya. Three sectors were selected based on their distance from the forest (0.2km, 5km and 10km). Five randomly selected plots separated by a distance of 0.2 km represented each sector. In each plot, 5 trees were randomly selected for observation of flower visitors. A total of 35 species of the flower visitors were observed representing order Coleoptera, Hymenoptera, Lepidoptera, Diptera and Hemiptera. The most abundant mango flower visitors were blow fly (Chrysomya chloropyga)(60.8%), honey bee (Apis mellifera)(3.7%), flesh fly (Sarcophaga sp) (2.5%)and housefly (Musca domestica)(1.9%). The findings clearly shows that flies are more important pollinators of mango compared to honeybees, as has been recently reported in Asia and Australia. flower visitor abundance and diversity reduced from farms closer to the forest edge towards the farmland (1409, 1225 and 653 individuals respectively) an d diversity. This can be attributed to the availability of nesting resources and/or mating sites in the natural habitat. These findings provide evidence that forest fragments and patches of natural habitats may act as reservoirs for a diversity of insects that provide crop pollination services to the adjacent farmlands. We recommend holistic pollinator management plan for farmlands in Kenya. Key words: Pollinator, Mango, spatial distribution, Chrysomya chloropyga, Apis mellifera, Musca domestica, forest-farm gradient INTRODUCTION Pollination is an ecological process fundamental for the maintenance of the viability and diversity of flowering plants and provides important ecosystems services to humans (Allen-Wardell et al., 1998; Daily et al., 1997; Kevan, 1999; Klein et al., 2007). According to Kevan and Phillips (2001) the importance of pollination in agriculture has been recognised for millennia. It has been reported that pollinator species differ in morphological characteristics and behaviours that determine their ability to pollinate (Horsburgh et al. 2011). The most effective pollinator species are those occurring in high abundance, actively moving from flower to flower (has a high visitation rate) and transferring many pollen grains on the stigmas (Rader et al. 2009). Mango is one of the most popular fruits within the tropics, and belongs to the family Anacardiaceae and genus Mangifera (Abidin, 1990). According to (Bally 2006; Aliakbarpour and Che Salmah 2010), mango flowers are visited by various insects such as wasps, ants, flies, butterflies, beetles, and bees. Mango production in Kenya is usually by smallholder farmers ranging from few trees to different size of orchards. This study therefore, seeks to determine potential mango pollinators and their spatial differences across forest-farm gradient. In Kenya, there is no prior study on Mango-pollinator relationship that has been carried out. Materials and Methods The study was carried out in Tala, sub-county located in Machakos County on the eastern slope of Kilimambogo forest. The study area stretches from latitudes 00° 45' 335 south to 10° 31' south and longitudes 360° 45'’ east to 370° 45' east. The research was conducted in two seasons (dry and wet seasons) over a period of six months from August 2015 to February 2016. Months of September and October 2016 comprised the flowering period, November-January fruit development period, and February 2016 the harvesting period. The experiment was set up in a transect through forest farm gradient stretching from the edge of the forest to about 10000m from the forest edge. Three sectors were identified to represent farms near the forest; 200m from forest edge, about 5000m from forest edge and 10000m from forest edge. In each sector five plots were randomly selected, separated by a distance of about 200m and two orchards were identified and supplemented with honey bees. A single farm constituted a plot containing at least 10 mango trees (orchard). To assess pollinator diversity and abundance, 5 plants in each plot were randomly selected. Most of the observations were carried out on a sunny day. Each plant was observed for one hour and the abundance of each visitor species recorded. Observations were done when plants were 50% blooming. Activity of the insect visiting mango flowers was recorded (whether they were collecting nectar or pollen) after which some were captured by sweep net and placed in a kilner jar containing ethyl acetate for identification at the National Museum of Kenya and pinned in the insect boxes. An exclusion experiment was set up across the three sectors where by 5 trees per farm plot were randomly selected. For each tree, 1 branch with 2 immature inflorescences were randomly selected by taking care that the panicles were of the same height, and approximately same length and exposure to sunlight. Tagged inflorescence were left open for pollination to take effect while tagged inflorescence excluded pollinators (no pollination by external forces was allowed) by using mosquito net with very fine mesh that is impermeable to pollen. The collected data was entered in excel and imported to Genstat version 17.0. Species diversity index determined using Shannon-Wiener (H) for diversity. For the treatment that is both tagged and tagged the mean was used in all the three sectors and ANOVA was used to determine if it is significant. RESULTS AND DISCUSSION A total of 3,284 insect samples, comprising 35 insect species of flower visitors were collected. The insect groups were coleopterans, hymenopterans, lepidopterans, dipterans, and hemipterans. Among the pollinating insect species found in the mango farms, the most abundant flower visiting taxa were Chrysomya chloropyga (1996), Apis mellifera (120), Musca lusoria (61), and trigona sp (24) over the two seasons. These findings were in line with those of Bally (2006) and Aliakbarpour and Che Salmah (2010), who noted that mango flowers are pollinated by various insects such as wasps, ants, flies, butterflies and beetles. Flower visitations were same despite the differences in varieties of mangoes that were flowering. This observation is in line with the findings of Anderson, (1982) who noted that majority of the pollinators choose nectar of mango flowers as their food resource. Other taxa occurred in relatively less abundance as shown in Table 1, confirming that insects play a critical role in pollination of mangoes. 336 Sung et al. (2006) reported that honey bees (Apis spp.) and flies (Musca domestica and Chrysomya megacephala) were responsible for successful pollination hence good mango production in Taiwan. Similarly, this study also observed that more than 50% of insects captured were dipterans (Table 1), and seemed to be potential pollinators of mango flowers. Table 1: Percentage of pollinators (different orders) observed on mango at Kilimambogo Order Pollinators Percentage (%) Diptera Chrysomya chloropyga 60.8 Musca lusoria 1.9 Hymenoptera Apis mellifera 3.7 trigona sp 0.7 Species diversity index shows that diversity decreased with increasing distance into the farmland from the forest edge (Table 1) indicated that sectors 1, 2 and 3 had diversities of 1.4, 1.3 and 1.2 respectively (Table 1. The index values ranges from 0 - 4.6, and for this case the values in the middle indicated that the numbers of individuals are not evenly distributed between all the species thus this measure provides a possibility of comparing species based on locality. According to Carvalheiro et al. (2010), flower visitors are negatively affected by distance to natural habitat. The study also observed the highest number of flower visitors at Sector 1 (1409), sector 2 (1225) and sector 3 (656) (Table 2). At sector 2, most of the farmers had strips and patches of land left uncultivated. Table 2: Diversity and abundance for 3 sectors Sector Distance from forest edge Abundance Diversity (H) 1 0-200m 1409 1.4 2 5000m 1225 1.3 3 10,000 656 1.2 The effect of flower visitors on mango fruits development Seven weeks after bagging and tagging (flowers left open for natural pollinator visitations) of branches, the development of fruits indicated that the tagged branches had the highest number of flowers than the bagged branches (Fig. 1). The average number of fruits in the tagged branches was 3.4±1.0, while the bagged was 1.6±0.2. The observed differences were significant (ANOVA, F=2.845, p ≤ 0.05). From the experiment set up, it was observed that mangoes are dependent on pollinators for fruit 337 development. This observation is in agreement with the findings of studies carried out by various authors (Free and Williams, 1976; Anderson, 1982; Richards, 2001; Carvalheiro et al.2010). These studies observed that pollinators are critical for successful fruit set of mango flowers. Contrary to our expectations, some panicles as shown in figure 1 were able to bear fruits which indicated that mangoes were able to set fruits despite exclusion of insect. Similar observation has been reported by Free and Williams (1976), who suggested that at least some pollination is assisted by wind or gravity. Figure 1: Mean number of fruits observed on bagged and tagged flowers The importance of natural habitat on the yield of mango During the experiment both diversity and abundance differed across the sectors depending on the distance from natural forest. The highest abundance of 42.9% was recorded at sector 1, sector 2 (37.3%) and at sector 3 (19.8%). Both abundance and diversity of flower visitors have been noted to influence the efficiency of pollination process (Kremen et al., 2002; Carvalheiro et, 2010). The study observed that plots located at 200m from natural forest had the highest mean number of mango fruits development (7.2±2.5) in tagged flowers (Fig. 3). On the contrary, plots located 5000m and 10000m from natural forests recorded very low mean number of fruits that developed (0.8±0.2 and 1.2 ±0.4) in tagged flowers. Study on bagged flowers recorded low mean fruit development (1.8±0.5), 200m; 1.1±0.3), 5000m; 2.0±0.6), 10000m that also varied minimally between plots located at 200, 5000 and 10000 metres from the forest (Fig. 2). Even though distance does not affect the yield of the mangos directly, it significantly plays a critical role in determining the number of wild pollinators that reach the farms. Natural habitats provide two valuable benefits to mango plantations; they are a source of pollinators for crop pollination, and also act as refugia for pollinators that can colonize degraded agricultural areas (Ockinger and Smith, 2007). 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Bagged Tagged M ea n n u m b er o f fr u it s 338 Figure 2: Mean number of fruits developed in both bagged and tagged at 200m, 5000m and 10,000m from natural forest CONCLUSION Pollinators play a critical role in formation of fruits in mangoes, and pollinators decline with distance from the natural habitat those farms that are near the natural habitat benefitted from the presence of wild pollinators. Most of the insects belonged to the orders of diptera and hymnoptera. The important pollinators are blowfly, honeybees flesh fly and housefly. For farms that are away from natural habitat conservation of the pollinators within farmland is very critical ACKNOWLEDGEMENT The authors would like to recognise all farmers who allowed us to use their farms for experiment We also thank KU, KALRO, NMK, and the entire team of Kenya Pollination Project for the support. The study was part of the GEF supported project “Conservation and Management of Pollinators for Sustainable Agriculture, through an Ecosystem Approach” implemented in seven countries – Brazil, Ghana, India, Kenya, Nepal, Pakistan and South Africa, coordinated by the Food and Agriculture Organization of the United Nations (FAO) with implementation support from the United Nations Environment Programme (UNEP). In Kenya the activity was carried out under KALRO. 339 REFERENCES Aliakbarpour H., Che Salmah M. R. (2010). Diurnal activity of four species of thrips (Thysanoptera: Thripidae) and efficiencies of tree nondestructive sampling techniques for thrips in mango inflorescences. J. Econ. Entomol. 103: 631–640 Horsburgh M., Semple J. C., Kevan P. G. (2011). Relative pollinator effectiveness of insect floral visitors to two sympatric species of wild aster: Symphyotrichum lanceolatum (Willd.) Nesom and S .lateriflorum (L.) Löve & Löve (Asteraceae: Astereae). RHODORA 113: 64–86. Rader R., Howlett B. G., Cunningham S. A., Westcott D. A., Newstrom-Lloyd L. E., Walker M. K., Teulon D. A. J., Edwards W. ( 2009). Alternative pollinator taxa are equally efficient but not as effective as the honey bee in a mass flowering crop. Journal of. Appiedl. Entomology. 46: 1080–1087. Anderson, D. L. H., Sedgley, J. R., Short, T., and Allwood, A. J. (1982). Insect pollination of mango in northern Australia. Australian. Journal Agriculture. 33, 541-548. Bally, I. S. E. 2006.Mangifera indica(Mango).InC. R. Elevitch (ed.), Species profiles for pacific island agroforestry. Permanent Agriculture Resources (PAR), Holualoa, Hawaii. Carvalheiro, L. G., Seymour, C. L., Veldtman, R., and Nicolson, S. W. (2010). Pollination services decline with distance from natural habitat even in biodiversity-rich areas. Journal of Applied Ecology 47(810-820). Carvalheiro, L. G., Seymour, C. L., Veldtman, R., and Nicolson, S. W. (2010). Pollination services decline with distance from natural habitat even in biodiversity-rich areas. Journal of Applied Ecology 47(810-820). Kevan, P. G., and Phillips, T. P. (2001). The economic impacts of pollinator declines: an approach to assessing the consequences. [Online]. Conservation Ecology, 5(1), 8. Kremen, C., Williams, N. M., and Thorp, R. W. (2002). Crop pollination from native bees at risk from agricultural intensification. Proceedings of the National Academy of Sciences of the United States of America, 99, 16812-16816. Ockinger, E., and Smith, H. G. (2007). Semi-natural graslands as population sources for pollinating insects in agricultural landscapes. Journal of Applied Ecology, 44, 50- 59. Allen-Wardell, G., Bernhardt, P., Bitner, R., Burquez, A., Buchmann, S., Cane, J., Cox, P.A., Dalton, V., Feinsinger, P., Ingram, M., Inouye, D., Jones, C.E., Kennedy, K., Kevan, P., Koopowitz, H., Medellin, R., Medellin-Morales, S., Nabhan, G.P., Pavlik, B., Tepedino, V., Torchio, P., Walker, S., (1998). The potential consequences of pollinator declines on the conservation of biodiversity and stability of food crop yields. Conservation. Biology. 12, 8–17. Klein, A.M., Vaissiere, B.E., Cane, J.H., Steffan-Dewenter, I., Cunningham, S.A., Kremen, C., Tscharntke, T., (2007) Importance of pollinators in changing landscapes for world crops. Proceeding. Society... (B) 274, 303–313 340 7.6 Low cost insect proof and micro-climate regulating Agronets are effective in managing infestation of American Bollworm (Helicoverpa armigera Hubner.) on field tomatoes in Nairobi and Kiambu counties, Kenya Achieng’a F.C1, M Kasina2* J Mbugi1, L Wasilwa3, F Kariuki1, P Kipyab4, M Ngouajio5, and T Martin6 1 Kenyatta University, Department of Zoological Sciences, P.O. Box 43844, 00100, Nairobi 2 National Sericulture Research Centre, KALRO, P. O. Box 7816, 01000 Thika 3Kenya Agricultural and Livestock Research Organization HQ, P.O. Box 57811-00200, Nairobi 4Kenya Plant Health Inspectorate Service HQ, P.O. Box 49592-00100, Nairobi 5Michigan State University, Department of Horticulture, Plant and Soil Sciences Bldg, 1066 Bogue Street Room A440-D, East Lansing, MI 48824, USA, current address USDA-NIFA (National Institute of Food and Agriculture), Institute of Food Production and Sustainability, Stop 2240, 1400 Independence Avenue, SW, Washington, DC 20250-2240 6Cirad UR Hortsys, Icipe - African Insect Science for Food and Health, P.O Box 30772- 00100, Nairobi, Kenya *Correspondence: cachienga@gmail.com; 0722 255 031 ABSTRACT Fruit damage by the American Bollworm (Helicoverpa armigera Hbn.) is a major challenge to small holder tomato (Solanum lycopersicum L.) farmers in Kenya. Pesticides though highly used have not effectively delivered best control system. The study was conducted on-station at Kenya Agricultural and Livestock Research Organization Kabete and Thika Practical Training Center to assess the effects of Agronets in the reduction of H. armigera damage and thus increase yields. Treatments included covering the crop with fine mesh Agronets (0.4 mm) throughout the season, fine mesh Agronets (0.4 mm) opened three times a week from 09:00-15:00h, large mesh Agronets (0.9 mm) covered throughout, large meshed Agronets (0.9 mm) opened three times a week from 09:00-15:00h, and a control (no Agronet). Each treatment was replicated five times in a completely randomized block design. The results show that there were significant treatment effects in the density of H. armigera infesting the crop as well as damaged fruits and yield. Non-protected tomatoes recorded significantly higher infestation, damage and lower yields compared with tomatoes protected using Agronets. Yields were however not significantly different even though unprotected tomatoes were fewer. It is concluded that further studies are warranted to provide best innovative way of covering the tomatoes. This is in consideration that there is recorded efficiency in reducing infestation by bollworm and therefore aspects of covering are important to consider and probably include an integrated pest and disease management approach to further reduce fruit loses. Key words: mesh size, pest, fruit damage, yield, tomatoes 341 INTRODUCTION The African bollworm, Helicoverpa armigera Hubner (Lepidoptera: Noctuidae) is one of the worst agricultural pests in Africa. It attacks a variety of food and cash crops and has been recorded as a damaging pest on 180 cultivated and wild plant species in at least 45 families. Just like in most insect pests, control of American bollworm requires careful monitoring and integration of appropriate control measures. Farmer’s approach to the management of the American bollworm relies mainly on frequent applications of pesticides with long-term persistence. However, pesticides application by small-scale farmers is generally inefficient due to the protected habitat of the larvae, poor application techniques, quick development of resistance and the fact that some pests such as the American bollworm are nocturnal (Gogo et al., 2014). On the other hand, pesticide applications are also injurious to human health, environmental safety and beneficial insects. National and international concern about this situation, the broad involvement of aid agencies in pest control programs, and recent successes of the concept of integrated pest management (IPM), formed the basis of this study. This study was carried out to determine efficiency of low cover netting on tomato to manage infestation of American bollworm. This netting technology has been tested previously in Benin and found effective on diamondback moth, Plutella xylostella (Martin et al., 2013) and in Kenya on tomatoes against H. Amigera; Gogo et al. (2014) and Juma et al. (2015). MATERIALS AND METHODS Experimental site and Study design On-station field studies were conducted at Kenya Agricultural and Livestock Research Organisation (KALRO) National Agricultural Research laboratories (NARL), Nairobi County and at Thika Practical Training Center (PTC), Muranga County in two cropping periods (under irrigation), from May 2009 to September 2009 and July 2009 to October 2009 respectively. Experimental plots 8 m2 (8 m long x 1 m wide) were used with a spacing of 50 cm. Commercially available tomato var Riogrande® seedlings were sourced from a six week old nursery covered with a fine mesh sized Agronet. The experiment was laid out in a completely randomized block design layout with five replicates. Five treatments were tested; Non-protected control (no Agronet), Agronet with fine mesh (0.4 mm) used permanently, Agronet with large mesh (0.9 mm) used permanently, agronet with fine mesh (0.4 mm) opened 3 times a week (Mondays, Wednesdays and Fridays) from 9 am to 3 pm and Agronet with large mesh (0.9 mm) opened 3 times a week (Mondays, Wednesdays and Fridays) from 9 am to 3 pm. Data collection Pest abundance: American Bollworm larvae incidence and count was recorded based on a stratified random sampling procedure whereby each plot was divided into three parts (strata) of five plants each and any plant from within each stratum sampled on a weekly basis to check larvae incidence as well as count their population up to harvest. 342 Quality and yield: Harvesting was carried out from twelve weeks after transplanting for seven consecutive weeks. The total number of fruiting plants in each plot was counted after which the harvesting process commenced. All fruits with a breaking colour were hand-picked in respective plots and bagged in well labeled bags and taken to the laboratory for weighing using an electronic balance. In the laboratory, the fruits were sorted and graded according to pest infestation as well as market grade. Each fruit category was then weighed and recorded based on the insect damage inflicted on them. Evaluation of insect pest damage on the fruits involved counting the number of fruits observed with punctures or blemishes on the skin as well as those which were unaffected thus categorisation into marketable and unmarketable grades for the respective treatment plots. Data analysis GENSTAT discovery edition IV software was used to carry out the statistical analysis. Data on the pests’ abundance and crop yield was subjected to Analysis of Variance (ANOVA) for comparisons at 5% confidence limit. Data on pest abundance however revealed strong evidence against normality. This data was thus square-root transformed (x + 0.5)0.5 before ANOVA was carried out. Means that showed significant differences were separated using Student-Newman-Keuls (SNK) multiple range test method. RESULTS Abundance of American bollworm infesting tomatoes grown under Agronets in Kenya There were no significant differences (F4,595=1.26; P=0.29) in the mean number of H. armigera recorded on tomato plants across treatments in the first cropping season (Table 1) but significance (F4,595=2.47; P=0.05) was recorded in the second cropping season (Table 1). The highest mean number was recorded on treatments with fine mesh Agronet closed permanently (0.56+0.31). This was followed by the treatments with fine mesh Agronet opened three times a week (0.12+0.07), the large mesh Agronet covered permanently (0.04+0.02) and the large mesh Agronet opened three times a week (0.04+0.02) which did not differ significantly. The control treatment had the lowest mean number recorded, (0.012+0.01). Effect of Agronets on the quality and yields of tomatoes in Kenya There were no significant differences (F4,595=0.22; P=0.93) (F4,595=1.24; P=0.30) in the mean fresh weight of non-marketable tomatoes harvested in the first cropping season (Table 2). However, in the second cropping season, significant differences were observed. The highest mean number was recorded on the control treatment (4.95+1.18). This was followed by the fine mesh Agronet treatment covered permanently (2.44+1.44) which did not differ significantly with the control. This was followed by the large mesh Agronet opened three times a week (1.62+0.60), then the large mesh Agronet covered permanently (1.35+0.55) and then the fine mesh Agronet opened three times a week (1.30+0.55), which had no significant differences between them and the fine mesh 343 Agronet treatment covered permanently (Table 2). There were no significant differences (F4,595=0.16; P=0.96) (F4,595=1.40; P=0.24) observed on the mean number of fruits harvested in the first and second cropping seasons respectively. No significant differences (F4,595=0.41; P=0.81) (F4,595=1.11; P=0.35) were also observed in the mean fresh weight of fruits harvested in the first and second cropping seasons respectively. Table 1. The mean numbers + S.E of larval H. armigera and the mean fresh fruit weight + S.E of non-marketable fruits recorded on tomato plants in the first and second cropping seasons Treatments Mean number of H. armigera larvae / plant Mean fresh fruit weight of non-marketable tomatoes /plant (g) 1st Season 2nd Season 1st Season 2nd Season FMA (P) 0.02 + 0.01 0.56 + 0.31c 58.26 + 11.58 2.44 + 1.44ab FMA (3T) 0.02 + 0.01 0.12 + 0.07b 52.14 +11.95 1.30 + 1.30a LMA (P) 0.03 + 0.01 0.04 + 0.02b 27.29 + 6.81 1.35 + 0.55a LMA (3T) 0.04 + 0.01 0.04 + 0.02b 57.62 + 19.64 1.62 + 0.60a Control (No Agronet) 0.05 + 0.02 0.012 + 0.01a 42.44 + 4.53 4.95 + 1.18b F 4,595; F 4,170 1.26 2.47 1.15 2.65 P-Value 0.29 0.05 0.33 0.04 Key: FMA (P) = Fine mesh Agronet closed permanently, FMA (3T) =Fine mesh Agronet opened three times a week, LMA (P) =Large mesh Agronet closed permanently LMA (3T) =Large mesh Agronet opened three times a week. Means within a column followed by the same letter are not significantly different at P= 0.05. Test statistic was ANOVA and means separated using SNK. DISCUSSION The H. armigera infestation was low across the seasons probably providing a reasoning why a no significant difference was reported. In the second cropping season, results on the abundance of H. armigera infesting tomatoes in Kabete showed that there were significant differences in the mean number of H. armigera recorded on tomato plants. Highest infestation was observed on the treatment with the fine mesh Agronet covered permanently followed by the rest of the treatments with no significant differences between them and the lowest on the control (no Agronet) treatment. H. armigera infestation observed inside the Agronet treatments may have been initiated from small caterpillars crawling across the Agronets as well as those hatching from pupae deposited on the ground before laying out of the treatments. This pest is known to be occasional hence the infestation levels follow expected population dynamics of this pest. Majumdar (2010) also reported that occasional pest outbreaks observed inside net houses may be as a result of insects emerging from the soil or small caterpillars 344 crawling across the net. H. armigera infestations are also thought to have been higher under Agronets treatments than in the control (no Agronet) treatment probably due to reduced action of natural enemies excluded by the covered Agronet plots. This finding on natural enemy exclusion by Agronets agrees with results from Martin et al. (2013) who reported that Agronets should be removed sometimes during the day to allow natural enemy activity as they appeared to represent a serious barrier for them. The fine mesh Agronet treatment in this study is also thought to have the highest mean number of H. armigera probably due to restricted entry of natural enemies onto the crop and exit of the pest from the crop as a result of the mesh size and covering of the crop throughout when compared to the rest of the treatments that were either opened three times a week or not covered at all (control). According to Majumdar (2010), cover pore diameter affects entry and exit of insects into the crops, thus justifying findings of this study. Berlinger et al. (2002), Martin et al. (2013) and Gogo et al. (2014) in their studies also showed that opening the nets during the day improved the protection of plants against pests probably because of the regulatory effect of beneficial insects. One of the main objectives of the use of Agronets was to protect the tomato plants from insect pest infestation and consequently damage to the fruits. Data collected on the quality and yield of harvested tomato fruits showed no statistically significant differences between treatments. The results of this study however clearly showed that use of temporary net covers for the protection of tomato crop can be an effective tool to protect plants against lepidopterans, corroborating earlier findings by Martin et al. (2013) and Licciardi et al. (2008) who worked on cabbages in Africa and Benin respectively. Generally the high marketable yields recorded in the study were attributed to use of Agronet covers for the protection of insect pests. At harvest, damage by H. armigera was the single most important source of yield loss resulting from insect damage, accounting for over 50% of the unmarketable yield. This finding is in line with those of Maerere et al. (2010), Gogo et al. (2014) and Juma (2015) who stated that H. armigera accounted for more than 50 % of unmarketable yield in open field check plots that had no treatments applied. Overall, the effects of Agronets on pest control and yield were positive. H. armigera is a significant threat to tomato production in Central Kenya. Effective control of this pest is inevitable for attaining reasonable yield levels. These findings on yield agree with results on field trials carried out by Gogo et al. (2014) and Juma (2015) who compared pest infestation on tomatoes grown in the open to those grown under Agronets. Gogo et al. (2014) concluded that Agronets are a viable strategy for improving tomato yields through reduction of the American Bolloworm population on tomatoes and can be incorporated as a component of integrated pest management strategy. Juma (2015) also concluded that Alpha-cypermethrin treated Agronets with a cover mesh diameter of 0.9 mm was effective in protecting tomato plants and tomato seedlings against major pests especially A. gossypii, B. tabaci, L. trifolii, Planococcus spp. and H. armigera. 345 CONCLUSION AND RECOMMENDATION Despite use of Agronets, pest entry into protected plots was noted thus called for additional pest management input to reduce pest pressure thus complementary pest control measures may be required alongside Agronet use. Large mesh Agronet covers opened three times a week gave better control of H. armigera when compared to the fine mesh size nets as well as those covered throughout since natural enemies were allowed to locate their hosts hence control pest populations considerably. Higher yields were generally recorded on tomatoes grown under Agronets when compared with control plots (no Agronets). Agronet covers should be used in an IPM strategy that would be effective in controlling key insect pests of tomatoes as well as increasing tomato yields. When used alone, pest pressure may drastically affect the crop and consequently result to a decline in yield. To be effective, Agronets must necessarily be installed prior to the pests’ appearance and all openings must be totally covered by the Agronets, including entrances. Complementary pest control measures are also required to reduce pest pressure. Augmentative release of natural enemies should be considered for the control of tomato pests when Agronets are used. Agronets should be removed sometimes during the day to allow natural enemy activity. The 0.9 mm mesh size net used on a temporal basis is recommended for the control of key tomato insect pests when compared to the other treatments tested in this study. ACKNOWLEDGMENTS The study was made possible through financial support by the USAID HortCRISP BioNetAgro project (Award No. EPP–A–00–09–00004)). The contents are the responsibility of HortCRISP BioNetAgro project investigators and do not necessarily reflect the views of USAID or the U.S. government. A to Z Textile Mills Ltd, Tanzania supplied the Agronets. We thank our institutions for administrative and technical support REFERENCES Berlinger, M.J., Taylor, R.A.J., Lebiush-Mordechi, S., Shalhevet, S., and Spharim, I. (2002). Efficiency of insect exclusion screens for preventing whitefly transmission of tomato yellow leaf curl virus of tomatoes in Israel. Bulletin of Entomological Research 92, 367-373. Fenemore, P. G., and Prakash A. (2006). Applied Entomology. New Age International (P) Limited, Publishers, New Delhi, India. Gogo, E. O., Saidi, M., Itulya, F. M., Martin, T. and Ngouajio, M. (2014). Eco-Friendly Nets and Floating Row Covers Reduce Pest Infestation and Improve Tomato (Solanum lycopersicon L.) Yields for Smallholder Farmers in Kenya. Journal of Agronomy 4: 1- 12. Goodwin, S., and W. Danthanarayana. (1984). Flight activity of Plutella xylostella (L.) (Lepidoptera: Yponomeutidae). Journal of Australian Entomological Society: 23: 235- 240. 346 Juma, V. H. O (2015). MSC. Agricutural Entomology Thesis, Kenyatta University. Effects of Insectici9de Treated nets in the Management of Tomato Pests and their impact on natural enemies and yield in Nairobi and Muranga Counties. Licciardi, S., Assogba K.F., Sidick I., Chandre F., Hougard J.M. and Martin T. (2008). A temporary tunnel screen as an ecofriendly method for small-scale farmers to protect cabbage crops in Benin. International Journal of Tropical Insect Science 27:152- 158. Maerere, A.P., Sibuga, K.P. and Mwajombe, K.K. (2010). Baseline survey report of tomato production in Mvomero district-Morogoro region, Tanzania. Sokoine University of Agriculture Faculty of Agriculture, Morogoro, pp. 1-31. Majumdar, A. (2010). Online Exclusive: Results from the 2009 Insect Monitoring Pilot Project in Alabama. American Vegetable Grower (R.Gordon, ed.). Miester Media, MI. {Online}http://www.growingproduce.com/americanvegetable grower/?storyid=4204 Martin, T., Palix, R., Kamal, A., Deletre E., Bonafos, R., Simon, S. and Ngouajio M. (2013). A repellent treated netting as a new technology for protecting vegetable crops. Journal of Economic Entomology 106: 1699 – 1706. 7.7 Appropriate pollinator management scheme for kenyan farmlands Guantai1,2 MM, J Mueke1 and M Kasina2 1 Department of Zoological Sciences, Kenyatta University, P.O. Box 43844,00100, Nairobi 2 National Sericulture Research Centre-KALRO, P.O. Box 7816 code 01000 Thika Correspondence: Mary.Mwari@kalro.org or Muo.Kasina@kalro.org ABSTRACT A one year (2014-2015) study was carried out in three regions representing low, medium and high agriculture intensification to determine presence and distribution of pollinators and their nests in farmlands bordering nearby natural habitats of Mt Kilimambogo forest, Kakamega forest and Mt Kenya forest, respectively. This is in response to scientific evidence in Kenya showing increasing pollination deficit of various pollinator-dependent crops, which is responsible for declining yields both in quantity and quality. The results of this study show that pollinator nests are more concentrated in farms farther from the forest as the forest provided nests for bees found around the forest. In addition, the pollinator density reduces with increasing distance from the forest. In all the regions, hedgerows were a permanent feature, under different level of heterogeneity. It was found that some plants flowered all year round, attracting and providing various pollinators food resources. Some of these plants had pithy stems capable of providing bee nests. This study provides an opportunity to implement catchment approach pollinator management scheme that include various aspects of 347 understanding, utilizing and protecting pollinators at farm level. The paper provides detailed analysis of the scheme and proposes best implementation approach to ensure farmers benefit from this essential ecosystem service. Keywords: agro-ecology, bee forage, bee nests, ecosystem service INTRODUCTION Pollination is an essential ecosystem service for increased crop production. About 35% of global crop production is attributed to the work of pollinators (Klein et al., 2007). Animal mediated pollination is carried out by insects, birds, butterflies, ants, among others; of which 90% is by bees (Michener, 2000). Out of the 2000 species of bees recorded (Kremen et al., 2007), about 3,000 species have been found in sub-saharan Africa (Gikungu, 2006). In Kenya, several studies have indicated that bees are the most important pollinators of most crops (Gikungu, 2006; Kasina et al., 2009a; Oronje et al., 2011). There have been concerns of pollinator decline in various parts of the world especially in Europe and North America (Kearns et al., 1998). In Kenya, though not yet documented, some research done at Kakamega has shown that the number of bees have decreased (Kasina et al., 2009a). Anthropogenic factors such as clearance of land for agriculture resulting in habitats loss, pesticides use and little knowledge on bees resulting to their loss (Tuell and Isaacs, 2010) are the main reason for this decline. Loss of biodiversity has led to a reduction in pollination services, which are needed to ensure sustainable food production (Morandin and Kremen, 2013). One of the ways of conserving the pollinators is by increasing the diversity of floral plants in farmlands so that bees have sufficient foraging and hiding sites (Kirk and Isaacs, 2012; Mwangi et al., 2012). Hedgerows have been cited as important in providing both forage and nesting for bees (Tuell and Isaacs, 2010; Gonigle et al., 2016). In Kenya no study has yet be carried out on the effectiveness of hedgerows in providing nesting sites for bees. This implies that bee presence in the farmlands is not assured and ultimately yields are reduced. Therefore this study sought to determine which bees visited hedgerows for forage, nesting and the best combinations of plants on the hedgerow that provided floral resources for bees throughout the year. 2 Materials and Methods The study was carried out in three sites; Mt Kenya, Kilimambogo and Kakamega forest along a transect from the forest towards the farmlands. In each site four sectors were chosen separated by a 5km distance for data collection purposes. Data on number of nests and plants visited was documented on monthly basis for a year. Data analysis was done using GENSTAT statistical software. Skewed data was subjected through square root transformation to normalize it. Analysis of variance (ANOVA) was used to compare means across the sectors and significance was sought at 95% level of confidence limit. Significant means were separated using the Fishers protected LSD (least significant difference). 348 RESULTS AND DISCUSSIONS Kakamega site recorded more nests (41.5%) compared with Mt. Kenya and Kilimambogo sites (33.3% and 25.1%) respectively. A significant (P=0.014) difference in the number of nests per site was recorded (Table 1). Nearer the forest, fewer bee nests were found on the hedgerows compared to the sectors far away indicating that the forest was providing nesting sites for bees found nearby (Fig 1). This corroborates with earlier studies in Kakamega (Hagen 2008; Kasina et al., 2009b) where more bees were found inside forest as there was abundance of forage and nesting sites for them compared to the farmland. Similar studies indicated that, it was impossible for bees to live without forests as this is their main source of nesting and forage (Gikungu, 2006). Table 1: Number of bee nests recoded at various sites in Kenya, October 2014 Fig 1: Number of nest found at various distances from forest About 80% of the nests found in Mt. Kenya, Kilimambogo and Kakamega sites were made of wood and the rest were found on the ground. This conforms to earlier studies by Michener (2000) who noted that species of Apidae, Megachilidae and Colettidae use abandoned nests located in wood cavities. Presence of few ground nesters is an indication of absence of a certain preference for nesting by bees. For example, excessive mulching with wood chips has been found to affect ground-nesting bees, which need 0 100 200 300 400 500 600 700 0-5km from forest 10-20Km from forest N u m b e r o f n e st s Distance of the nest from the forest Sites Number of nests (Percentage) Mean Kilimambogo forest 204 (25.1) 50.75a Mt. Kenya forest 272 (33.3) 68.00a Kakamega forest 339 (41.5) 84.75b P value 0.014 349 bare soil or a thin layer of natural leaf litter (Frankie et al., 2009). Hedgerows harboured 80% of all the nests recorded while the rest were found on the house rafters and within the crop. At the hedgerow, only trees provided sites for large nests. However, most of the trees are pruned and any dead twig used for fuel wood (Kasina et al., 2009b) thus denying these large bees areas to nest on. Thus nests were found on short dry twigs that cannot be harvested for fuel wood due to size limitation. The twigs had pithy stems which were easily tunneled by smaller bees for nesting purposes. Such twigs are quite few, confirming the lower number of large nests on the hedgerows. Practices such as the removal of dry wood around farms are known to destroy nesting sites for solitary bees (Kremen et al., 2007). Among the observed pollinator floral host plants, only 10 of them were recorded flowering every month of the year (Table 2). Plants in the family Asteraceae were the most common of the plants listed followed by Lamiaceae (Table 2). This confirms earlier studies in Kenya by Gikungu (2006) and Karanja et al., (2010) who reported plants in the family Asteraceae as important bee plants that provide resources for bees throughout the year. Table 2: Some of the host plants that flowered most frequently in all sectors, September 2013 to August 2014 Scientific name Common name 1 Achyranthes aspera Devil’s horsewhip 2 Aspilia mosambicensis Aspilia 3 Bidens pilosa Black jack 4 Commelina diffusa Climbing day flower 5 Lantana trifolia Threeleaf verbena 6 Ocimum suave Mtule basil 7 Plectranthus barbatus Indian coleus 8 Sesbania sesban Egyptian pea 9 Solanum incanum Sodom apple 10 Tagetes minuta Mexican marigold 350 RECOMMENDATIONS Considering the dependence of farmers on hedgerows to provide fuel wood integrated bee forage and nest management need to be devised, which include capacity building of farmers to enhance their skills on how to conserve and utilize bees for crop production. For example, farmers would keenly include artificial nests based on bee-preferred wood material in the hedgerows and strategic places in the farms. REFERENCES Frankie, G.W., Thorp, R.W., Hernandez, J., Rizzardi, M., Ertter, B., Pawelek, J.C., Witt, S.L., Schindler, M., Coville, R. and Wojcik, V.A. (2009). Native bees are a rich natural resource in urban California gardens. California Agriculture 63(3):113-120. Gikungu, M.W. (2006). Bee diversity and some aspects of their ecological interactions with plants in a successional tropical community. PhD Dissertation, University of Bonn. Gonigle, L.K., Williams, N.M., Lonsdorf, E. and Kremen, C. (2016). A tool for selecting plants when restoring habitat for pollinators. Conservation letters. Doi: 10.1111/conl. 12261. Hagen, M. (2008). Composition, spatial heterogeneity and seasonal turnover of plant- pollinator communities in an agriculture-forest mosaic. Dissertation, University of Bielefeld, Germany. Karanja, R.H.N., Njoroge, G.N., Gikungu, M.W. and Newton, L.E. (2010). Bee interactions with wild flora around organic and conventional coffee farms in Kiambu district, Central Kenya. Journal of Pollination Ecology 2(2): 7-12 Kasina, M., Mburu, J., Kraemer, M. and Holm-Müller, K. (2009a). Economic benefit of crop pollination by bees: A case of Kakamega small-holder farming in Western Kenya. Journal of Economic Entomology 102(2): 467-473. Kasina, M., Kraemer, M., Martius, C. and Wittmann, D. (2009b). Diversity and activity density of bees visiting crop flowers in Kakamega, Western Kenya. Journal of Apicultural Research and Bee World 48(2): 134-139 Kearns, C., Inouye, D. and Waser, N. (1998). Endangered mutualisms: the conservation of plant–pollinator interactions. Annual Review of Ecological Systems 29: 83–112. Kirk, A.K and Isaacs, R. (2012). Predicting Flower Phenology and Viability of High bush Blueberry. Hortscience 47(9): 1291–1296. Klein, A., Vaissiere, B.E., Cane, J.H., Dewenter, I., Cunningman, S.A., Kreman, C and Tscharntke, T. (2007). Importance of pollinators in changing landscapes for world crops. Journal of Biological Sciences 274: 303-313 Kremen, C., Williams, N.M., Aizen, M.A., Gemmill-Herren, B., LeBuhn, G., Minckley, R., Packer, L., Potts, S.G., Roulston, T., Steffan-Dewenter, I., Vazquez, D.P., Winfree, R., Adams, L., Crone, E.E., Greenleaf, S.S., Keitt, T.H., Klein, A.M., Regetz, J. and Ricketts, R.H. (2007). Pollination and other ecosystem services produced by mobile organisms: a conceptual framework for the effects of land-use change. Ecology Letters 10: 299-314. 351 Michener, C.D. (2000). The Bees of the World, xiv + [1] +913 pp. Baltimore: Johns Hopkins University Press. Morandin, L.A. and Kremen, C. (2013). Hedgerow Restoration Promotes Pollinator Populations and Exports Native Bees To Adjacent Fields. Ecological Applications 23(4): 829–839 Mwangi, D., Kasina, M., Nderitu, J., Hagen, M., Gikungu, M. and Kraemer, M. (2012). Diversity and abundance of native bees foraging on hedgerow plants in the Kakamega farmlands, western Kenya. Journal of Apicultural Research 51(4): 298- 305. Oronje, M.L., Hagen, M., Kasina, M. and Kraemer, M. (2011). Pollinator diversity Behaviour and Limitation on yield of Karela (Momordica carantia L. Cucurbitaceae) in Western Kenya. African Journal of Agricultural Research 7(11): 1629-1638 7.8 Avocado (Persea americana) yield as influenced by pollinators in Kandara, Murang’a County, Kenya Mulwa J 1,2, R Gathu2 and M Kasina1 1Kenya Agricultural and Livestock Research Organization (KALRO), P.O. Box 7816- 01000 Thika, Kenya 2Department of Agricultural Science and Technology, Kenyatta University, P.O. Box 43844-00100 Nairobi, Kenya Correspondence: Muo.Kasina@kalro.org/josemulwa@gmail.com ABSTRACT Avocado (Persea americana Mill) is an important crop that provides source of income and nutrition for many Kenyans. However, its production is affected by many factors. Pollination has not been considered in the past as a key factor of production. This study was carried out to determine the contribution of pollination to the productivity of this crop in terms of fruit set and fruit yield. It was carried out in farmer fields at Kandara, Murang’a County. The treatments included unlimited access of flowers by pollinators, and preventing access by pollinators. A total of 44 trees were randomly selected in 4 farms in season 1 and repeated in preceding seasons. A minimum distance of about 10 m and 200 m was maintained from tree to tree and farm to farm, respectively. On each tree, two randomly selected terminal branches with inflorescences of the same age and size were identified, one bagged to deny pollinators visits and another left open for unlimited access by pollinators. Preliminary results show highly significant difference (P= 0.002) in fruit yield across the treatments, translating to 70% pollination deficit, which confirms the world recorded pollination deficit of avocado. A 305% increase in fruit set was recorded when flowers were exposed to pollinators. This translated to a 352 333% increase in fruit yield, considering fruit abortion during the growth period. The study confirms that pollination is a major factor of avocado production. The findings affirm the importance of pollination management in avocado and calls for urgent utilization of pollinators for increased productivity and quality of fruits. Key words: bees, pollination, avocado pollinators INTRODUCTION Avocado (Persea americana Mill) is an important crop that provides source of income and nutrition for many Kenyans. However, its production is affected by pollination provision (Perez-Balam et al., 2012), which is a precursor to fertilization of ovules, leading to fruit set. In other parts of the world, avocado is commercially provided with honeybees (Apis mellifera) for pollination provision (Ish-Am, 2005). In its nature, the crop is also known to be visited by stingless bees, wasps and flies (Gazit and Degani, 2002; Can-Alonzo et al., 2005; Afik et al., 2006). Lack of sufficient pollinators has been reported to results in no or only a few fruits in Israel, California and South Africa (Ish- Am, 2005). Ish-Am (Ibid.) thus noted that pollination is a limiting factor for avocado productivity. A recent economic analysis in Kenya by Kasina and Gemmill-Herren (2014) and globally by FAO (2009) showed that pollination is important for fruits and result to about 25% yield loss if not provided. Avocado yield-dependence on pollen vectors is much higher than this. Where farmers do not manage pollination, yield losses may be immense. This study was carried out therefore, to determine the contribution of pollination to the yields of avocado in terms of fruit set and fruit yield. MATERIALS AND METHODS This study was carried out in farmer fields at Kandara, Murang’a County, which leads in terms of small-scale avocado growing for export market in Kenya. Treatments included unlimited access of flowers by pollinators, and preventing access of flowers by pollinators. A total of 44 trees were randomly selected in 4 farms in season 1 and repeated in preceding seasons. A minimum distance of about 10 m and 200 m was maintained from tree to tree and farm to farm, respectively. On each tree, two randomly selected terminal branches with inflorescences of the same age and size were identified, one bagged to deny pollinators visits and another left open for unlimited access by pollinators to represent 2 treatments replicated 44 times. Fruits from the 2 different treatments were counted for fruit set differences and later harvested at the end of the season to calculate avocado pollination deficit. RESULTS AND DISCUSSION A 305% increase in fruit set was recorded when flowers were exposed to pollinators (Figure 1). The difference was statistically significant (p=0.002) confirming that pollen vectoring in avocado is crucial for fruit set as reported elsewhere (e.g. Ish-Am, 2005). 353 Figure 1: Mean avocado fruits set per tree terminal branch at the start of fruiting A more than 3-fold increase in fruit yield was recorded when the crop was provided with pollination compared without pollination provision (Figure 2). The difference in yields translates to a 70% pollination deficit, which confirms the world recorded pollination deficit of avocado. This shows that insufficient pollination could be one of the major limiting factors in avocado production in Kenya considering that farmers are not known to manage pollination (Kasina et al., 2009). Figure 2: Mean avocado fruits yield per tree terminal branch subjected to different treatments The study confirms that pollination is a major factor of avocado production. The findings affirm the importance of pollination management in avocado and calls for urgent utilization of pollinators for increased productivity and quality of these fruits. 0 5 10 15 20 25 Bagged Unbagged N o o f fr u it s p e r te rm in al b ra ch Treatments Avocado fruits set 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 Bagged Unbagged M e an N o o f fr u it s p e r te rm in al b ra n ch Treatments Avocado Yield 354 Indeed, pollination management is a production factor and its lack of management would lead to poor seed set, fruit set and fruit quality (Sabbahi et al., 2006). CONCLUSION This study found that avocado trees exposed to unlimited access by pollinators resulted to 305% increased fruit set and 333% yield increase compared to those denied pollinator visits. The management and protection of pollinators are therefore of vital importance in avocado production. ACKNOWLEDGEMENT This study was made possible by the Government of Kenya through the HRF (HCD) Award (Fruit Fly Project, PI: Dr Muo Kasina). We thank Kandara farmers for their willingness to partner. We thank KALRO management for technical and administrative support offered during this study. REFERENCES Alcaraz, M. Librada, and Hormaza, J.I. (2009). Avocado pollination and fruit set-a perspective from Spain. California Avocado Society Yearbook 92: 113-135. Degani, C., Lahav, E. and El-Batsri, R. (2003). Caging single avocado trees beehive does not guarantee exclusive formation of selfed progeny. HortScience 38 (7), 1433– 1434. Afik, O., Drag, A., Kerem, Z., and Sharif, S. (2006). Analysis of avocado (Persea Americana) nectar properties and their perception by honeybees (Apis mellifera). Journal of Chemical Ecology 32: 1949-1963. Gallai, N. and Vaissière B.E. (2009). Guidelines for the economic valuation of pollination services at a national scale. Rome, FAO Garner, L.C. and C.J. Lovatt. (2008). The relationship between flower and fruit abscission and alternate bearing of ‘Hass’ avocado. J. Amer. Soc. Hort. Sci. 133:3-10. Garner, L.C., V.E.T.M. Ashworth, M.T. Clegg and C.J. Lovatt. (2008). The impact of outcrossing on yields of ‘Hass’ avocado. J. Amer. Soc. Hort. Sci. 133:648-652. Gazit, S. and C. Degani. (2002). Reproductive biology, p. 101–133. In: A.W. Whiley, B. Schaffer, and B.N. Wolstenholme (eds.). The avocado. CAB International, Wallingford, U.K. Ish-Am G. (2005). Reproductive biology of the avocado-a review. New Zealand and Australian Avocado Growers’ Conference, CD Publication. Tauranga, New Zealand. Ish-Am, G. and E. Lahav. (2011). Evidence for a major role of honeybees (Apis mellifera) rather than wind during avocado (Persea americana) pollination. J. Hort. Sci. and Biotech. 86:589-594. Kasina, M. and B Gemmill-Herren. (2014). Vulnerability of crop production to pollinator decline in Kenya. The 8th Egerton University International Conference and Expo, 2014. P 362-365. 355 Sabbahi, R, D de Oliveira, and J. Marceau. (2006). Does the honeybee (Hymenoptera: Apidae) reduce the blooming period of Canola? Journal of Agronomy and Crop Science. Vol. 192, Issue 3: 233-237. Can-Alonso, C., Quezada-Euàn, J.J.G., Xiu-Ancona, P., Moo-Valle, H., Valdovinos- Nunez, G.R. and Medina-Peralta, S. (2005). Pollination of ‘criollo’ avocados (Persea americana) and the behavior of associated bees in subtropical Mexico, Journal of Apicultural Research. 44, 3-8. Perez-Balam, J., Quezada-Euan, J.J.G., Alfaro-Bates, R., Medina, S., McKendrick, L., Soro, A. and Paxton, R.J. (2012). The contribution of honey bees, flies and wasps to avocado (Persea americana) pollination in southern Mexico. J Poll Ecol 8:42–47. 356 8.Parallel Session 7: Technology diffusion impact and sustainable Agribusiness value chains 8.1 Effect of phosphorous on yield and growth of soybean (Glycine max L) varieties in central highlands of Kenya Murage, F.M.1*, Mugwe, J.N.2, Ngetich, K.F.1, Mucheru-Muna, M.M.3 and Mugendi, D.N.1 1University of Embu, P.O. Box 6- 60100, Embu Kenya 2 Kenyatta University, P.O. Box 43844-00100, Nairobi, Kenya Department of Agricultural Resource Management, School of Agriculture and Enterprise Development, 3 Kenyatta University, P.O. Box 43844-00100, Nairobi, Kenya, Department of Environmental Sciences, School of Environmental Studies *Corresponding Author: Email: mucirafrancis@yahoo.com ,Tel: 0702685528 ABSTRACT Soybean (Glycine max (L.) Merrill) is one of the most important legume crops being introduced in the smallholder farming systems of the Central Highlands of Kenya to improve income and household nutrition of farmers. However, low phosphorus fixation is one of the major causes of low soybean yields. Central Highlands of Kenya has humic Nitisols with high acidity and high inherent Phosphorous fixing. Information on how P affects soybeans in this region is scarce yet critical. The objective of the trial was to evaluate effects of Phosphorous (P) on growth of two soybean varieties (Gazelle and SB 19) and yield. The source of P was Triple Super Phosphate (TSP) which provided 40 kg P ha-1 which is the recommended rate. The trial was carried out at Muthambi, Sub County Tharaka Nithi County for two cropping seasons; Short Rains 2015(SR 2015) and Long Rains 2016 (LR 2016). The experimental design was a randomized complete block design (RCBD) with four treatments replicated eight times. The treatments were; Gazelle with 40 kg P ha-1,Gazelle with 0 kg P ha-1, SB 19 with 40 kg P ha-1and SB 19 with 0 kg P ha-1 .Data was subjected to analysis of variance (ANOVA) in SAS 9.2 and means separated using Least Significant Difference (LSD) (p=0.05). Results showed that there was significant (P< 0.0001) difference in plant height and number of pods between the two varieties during SR 2015 season. Result also showed that use of P at 40 Kg/Ha led to increase of Gazelle yields by 34% and that of SB 19 by 7% compared to the control during SR 2015. The study recommends use of Triple super phosphate in central highlands of Kenya in soybean production Key words: Soybean Gazelle, SB 19, Phosphorous 357 INTRODUCTION Soybean [Glycine max (L.) Merrill] is the world’s leading source of protein and oil. It has the highest protein content of all food crops and is second to groundnut in terms of oil content among grain leguminous crop (Ikeogu and Nwofia, 2013). It also contributes to the improvement of soil fertility through biological nitrogen fixation (Sanginga et al., 2002). The spread of soybean crop from its native land of origin has been mainly due to its adaptability and predominant use as a food crop for human nutrition, source of protein for animals, medicinal plant and lately as an industrial crop (Ikeogu and Nwofia, 2013). Nevertheless, the crop is relative new under traditional cropping systems of Sub-Saharan Africa (Sanginga et al., 2002). In the central Highlands of Kenya, soybean production is increasingly being practiced by smallholder farmers, but the yields are still lower than the reported yields for the Sub-Saharan Africa region. Some of the highlighted constraints in the current attempts for the successful incorporation of soybean crop in to the cropping system of these zones are soil related constraints such as low pH, nutrient deficiencies (phosphorus, nitrogen), and toxic levels of some metals like aluminium and iron (Rubaihayo et al., 2000), and agronomy related constraints such as lack of improved varieties (Ikeogu and Nwofia, 2013). Soybean, like all other nitrogen fixing leguminous crops, requires P for its proper growth and N fixation. Soybean’s effective contribution to soil fertility improvement can be inhibited due to P deficiency (Giller and Cadish 1995). Legumes are especially sensible to low P availability because the biological nitrogen fixation requires high levels of P. The P deficiency can limit the nodules formation while the P fertilization can overcome the deficiency (Carsky et al., 2001). Soils that fix P can severely affect the yields of crops. Phosphorous fixation in the soils increases with the time of contact between soluble phosphates and soil particles. Most of legume crops develop poorly under low soil fertility conditions; especially where the acid conditions of the soils makes free oxides of iron (Fe) and aluminum (Al) to react with soil P forming insoluble compounds, thus not allowing its availability in the soil (Bationo et al., 2011). Recent studies conducted revealed positive response and lack of response of soybean varieties to moderate P application of 30 kg P2O5 ha-1 and inoculation with rhizobia was also noted (N2 Africa, 2011). Compared to the control treatment (0kg ha-1 P and without application of inoculants), the grain yield in soils with moderate fertility increased from 0.8 t ha-1 to 1.5 t ha- 1 with inoculants application and then to 2 t ha-1 when inoculants and P where applied. However, there was no clear response of P and inoculants application in acidic soils similar to the Central Highlands of Kenya. In this region, apart from the soils being acidic, information about the response of soybean crop to phosphorus fertilizers is lacking. Thereby, this study intended to evaluate the response of soybean crop in terms of yield and growth to phosphorus fertilizer. MATERIAL AND METHODS Site Description The experiment was set at Muthambi Sub County; Tharaka Nithi County. It was carried out among eight members of Thagania group during short rain (SR) 2015 season and 358 Long Rain (LR) 2016 season. According to Jaetzold et al. (2006), The area is in upper midlands 2 and 3 (UM2–UM3) with an altitude of approximately 1,500 m above sea level, annual mean temperature of about 200C and annual rainfall varying from 1,200 to 1,400 mm. The rainfall is bimodal, falling in two seasons, the long rains (LR) lasting from March through June and short rains (SR) from October through December. The soils are Humic Nitisols (Jaetzold et al. 2006), which are deep, well weathered with moderate to high inherent fertility. Experimental design The experiments were set up in eight farmers’ fields. The experimental design was a randomized complete block design (RCBD) with four treatments replicated eight times. The trials were implemented for two cropping seasons; short rains 2015 and long rains 2016. The plots measured 4.0 M by 4.5 M and test crop was Soybean, Gazelle and SB 19 varieties. The treatments were: Gazelle with 40 kg P ha-1, Gazelle with0 kg P ha-1, SB 19 with 40 kg P ha-1, SB 19 with0 kg P ha-1. Field Management Land was ploughed manually using a hand hoe. Triple super phosphate (TSP) was the source of P and was applied per furrow and well mixed with the soil during planting. Soybean was sown on 14th and 15thOctober, 2015 by planting 3 seeds per hole. The spacing was at 50 cm × 10 cm inter- and intra-row, respectively. Two weeks after emergence, the seedlings were thinned to 2 plants per hill. First and second weeding and spraying was done during the growing season. Data Collection At physiological maturity stage, soybeans were harvested from a net plot of 10.5 m2. Plants were uprooted and the roots cut using a panga and fallen leaves were collected and weighed together in the field. Subsequently, the pods were threshed and fresh weight of the grain was determined. Thereafter, the grain samples were sun dried and yields were determined and adjusted to 12.5 % moisture content. DATA ANALYSIS Data generated were subjected to analysis of variance (ANOVA) using Statistical Analysis Software (SAS) version 9.2. Least Significance Difference (LSD) at 95% of significance level was used to separate means. 359 RESULTS Table 1: The farmer variety assessment tool of SB 19 and Gazelle during SR 2015 Variety-(Rate from 1=least preferred, to 10 most preferred) SB 19 GAZELLE Number of pods 9 7 Size of pods 8 8 Number of seeds per pod 7 7 Filling of pods 8 7 Early maturity 10 8 Grain yield 8 6 Vigor of plant 7 9 Height of plant 9 7 From the table above farmers rated SB 19 variety better than Gazelle in terms of Number of pods per plant. They rate gazelle and SB 19 equally in terms of the size of the pods. In terms of the number of seeds per pod they rate both varieties equally. They rate SB 19 as better than gazelle in terms of filling of the grains. SB 19 matured earlier than Gazelle. Gazelle had more vigor than SB 19. SB 19 was taller than Gazelle. Table 2: plant Height, No of pods, yields, Stover yield and weight of 100 seeds during SR 2015 Variety Weight of 100 seeds(g) No of pods Plant height(cm) Yields(T/Ha) Stover yield(T/Ha) Gazelle 19.21 a 19 b 43.1 b 1.67 a 2.16 a SB 19 13.40 b 34 a 61.3 a 1.61 a 1.75 a p- value <0.0001 <0.0001 <0.0001 0.8559 0.0618 LSD (0.05) 1.29 5.53 5.57 0.65 0.43 Numbers followed by the same letter are not significantly different 360 From the table above there was a significant difference in weight of 100 seeds between the two varieties. There was a significant difference in number of pods and plant height of the two varieties. Table 3: plant Height, No of pods, yields, and weight of 100 seeds during SR 2015 data Treatment Plant height(cm) Number of pods Weight 100 seeds(gms) Yields t Ha-1 Gazelle with 40 kg P ha-1 45.7 21 19.79 2 Gazelle with 0 kg P ha-1 40.49 17 18.64 1.33 SB 19 with 40 kg P ha-1 62.33 34 13.65 1.67 SB 19 with 0 kg P ha-1 60.2 34 13.15 1.55 From the table above SB 19 with 40 kg P ha-1 had the highest height with 60.2 cm. Gazelle with 40 kg P ha-1 had the highest weight of 100 seeds. Gazelle with 40 kg P ha-1 had the highest yield. Table 4: No of nodules, no of pods, plant height and Yield during LR 2016 Variety No of Nodules No of pods Plant height Yields(T/Ha) Gazelle 5.5 a 19 a 39.55 b 0.804 a SB 19 6.1 a 18 a 46.34 a 0.6265 a p- value 0.7934 0.6588 0.0129 0.2155 LSD (0.05) 3.843 4.0996 5.2647 0.2855 Numbers followed by the same letter are not significantly different There was significant difference in plant height of the two varieties with SB 19 being taller than Gazelle. 361 Table 5: No of nodules, no of pods, plant height and Yield during LR 2016 Treatment Plant height(cm) Number of pods Nodules Yields t Ha-1 Gazelle with 40 kg P ha- 1 39.59 19.7 5.4 0.8563 Gazelle with 0 kg P ha-1 39.51 18.4 5.6 0.7518 SB 19 with 40 kg P ha-1 47.7 20.1 7.9 0.736 SB 19 with 0 kg P ha-1 44.98 16.2 4.1 0.5171 SB 19 with 40 kg P ha-1 had the highest plant height and number of pods. Gazelle with 40 kg P ha-1 had the highest yield. Figure 1: Graph on yields SB 19 and Gazelle with fertilizer and no fertilizer variety Fertilizer No fertilizer Gazelle 2 Mg/ha 1.33 Mg/ha SB 19 1.67 Mg/ha 1.55 Mg/ha 0 0.5 1 1.5 2 2.5 Fertilizer No fertilizer Y ie ld in M g/ h a Variety Gazelle SB 19 362 DISCUSSION There was a significant different in plant height of the two varieties in both seasons. SB 19 being taller than gazelle with a mean plant height of 61.3cm (Table 2). Application of TSP increased nutrient supply which enhanced vegetative growth, affecting plant height and yields (Umoetok et al., 2007). The present findings were in agreement with the result of (Bothe et al. 2000) who reported that the application of phosphorus at 75 kg ha-1 enhanced the plant height at highest value. (Maurya and Rathi 2000) also found the positive effect of P on the plant height of soybean. P fertilizer might have supplied P which enhanced N uptake and microorganism’s development, therefore mineralization. There was significant difference between the numbers of pods per plant of the two varieties during SR 2015 season and SB 19 having the highest number of pods (Table 2). Application of fertilizer led to increase of the number of pods per plant in both varieties. These findings were in agreement with (Reddy and Giri1989) who reported that phosphorus application at 20 kg P ha-1 increased the pod yield of soybean. Similar results were also found in soybean by (Jana et al. 1990) and (Singh and Bajpai1990). There was significance difference between the weights of 100 seeds of the two soybean varieties. Gazelle had the highest with 19.21 grams. Application of fertilizer led to increase of the weight of 100 seeds. There was no significance difference between the yields of the two soybean varieties. Gazelle had slightly higher yields than SB 19 (Table 2 and Table 4). There was an increase of yields of the two varieties in the two seasons. These results are in agreement with the findings of (Maurya and Rathi 2000) who reported the increase in grain yield of soybean with the application of P .There was no significant (0.3811) interaction between treatment and variety on yields of soybean. Application of TSP increased nutrient supply which enhanced vegetative growth, affecting plant height and yields (Umoetok et al., 2007). P fertilizer might have supplied P which enhanced N uptake and microorganism’s development, therefore mineralization. CONCLUSION AND RECOMMENDATION Gazelle had a higher Stover yield than SB 19. Gazelle was better in terms of yields than SB 19. In terms of plant height and number of pods per plant SB 19 was better than Gazelle. In conclusion Gazelle outweighed SB 19 in terms of production. Farmers need to be applying Triple super phosphate (TSP) at rate of 40 kg P ha-1 for increased yields regardless of the variety in question. Acknowledgments The authors acknowledge AGRA-SHP (Alliance for a Green Revolution in Africa, Soil Health Program) for providing funding to conduct the trial under SoCo project. We appreciate the SoCo project team especially the technicians and field coordinator for their support during data collection. 363 REFERENCES Abera, T., Feyissa, D., and Husuf, H. (2005). Effects of inorganic and organic fertilizers on grain yield of maize-climbing beans intercropping and soil fertility in Western Oromiya, Ethiopia. Paper presented in a conference on International Agricultural research for development. Stuttgart-Hohenheim. October 11-13, 2005. Bakayoko, S., Soro, D., Nindjin, C., Dao, D., Tschannen, A., Girardin, O., and Assa, A. (2009) Effects of cattle and poultry manures on organic matter content and adsorption complex of a sandy soil under cassava cultivation (Manihotesculenta, Crantz). Afric. J. Environ. Sci. Technol., 3(8): 190-197. Bationo, A., Hartemink, A., Lungu, O., Naimi, M., Okoth, P., Smaling, E., and Thiombiano, L., (2006). African soils: Their productivity and profitability of fertilizer use. Background papers.Africa Fertilizer summit.AbujaNigeria.Chiezey UF, Odunze AC (2009) Soybean response to application of poultry manure and phosphorus fertilizer in the Sub-humid Savanna of Nigeria. J. Ecol. Nat. Environ., 1(2): 025-031. Bationo, A., Waswa, B., Okeyo, J., Maina, F., Kihara, J., and Mokwunye, U., (2011). Fighting poverty in SubSahara Africa: The multiple roles of legumes in integrated soil fertility management. Springer Dordrecht Heidelberg, London, New York Research. Pp 246. Bothe, D.T., Sabale, R.N. and Raundal, P.U. (2000). Effect of phosphorus, plant population and P-Solubilizer on soybean-fenugreek cropping system. Journal of Maharashtra Agricultural Universities 25(3), 310-311 Carsky, R. J., Singh, B. B., and Oyewole, R. (2001). Contribution of early–season cowpea to late season maize in the savanna zone of West Africa. Biol. Agric. Hort., 18, 303- 315. Chimdi, A., Gebrekidan, V., Kibret, K., and Tadesse, A. (2012). Effects of Liming on Acidity-Related Chemical Properties of Soils of Different Land Use Systems in Western Oromia, Ethiopia. World J. Agric. Sci., 8(6): 560-567. Crawford, Jr, T.W., Singh, U., and Breman, H. (2008).Solving problems related to soil acidity in Central Africa’s Geat Lakes Region. International Center for Soil Fertility and Agricultural Development (IFDC) - USA. Giller, K.E., and Cadisch, G. (1995). Future benefits from biological nitrogen fixation: an ecological approach to agriculture. Plant Soil, 174, 255-277. http://dx.doi.org/10.1007/BF00032251. Hassan, A., Mohamad, A., Abdu, A., Idrus R.M., and Besar, N.A. (2010). Soil properties under Orthosiphonstamineus (Benth) intercropped with Duriozibethinus (Murr) and treated with various organic fertilizers. Paper presented on 19th World Congress of Soil Science, Soil Solutions for a Changing World 1 – 6 August 2010, Brisbane, Australia. Heisey, P.W., and Mwangi, W. (1996). Fertilizer Use and Maize Production in Sub- Saharan Africa.CIMMYT Economics Working Paper 96-01. Mexico, D.F.: CIMMYT. 364 Ikeogu, U. N., and Nwofia G. E. (2013) Yield parameters and stabilty of soybean [Glycine max. (L.) merril] as influenced by phosphorus fertilizer rates in two ultisols. Journal of Plant Breeding and Crop Science. 5 (4), 54– 63. DOI: 10.5897/JPBCS12.014 Imas, P., and Magen, H. (2007). Role of Potassium Nutrition in Balanced Fertilization for Soybean Yield and Quality - Global Perspective. In: Imas P and Vyas AK (eds) Proceedings of regional seminar on recent Advances in potassium nutrition management for soybean based cropping system. September 28-29, 2007. National Research Centre for Soybean, Indore, India, p. 126. Jaetzold, R., Schmidt, H., Hornet, Z.B., and Shisanya, C.A. (2006).Farm management handbook of Kenya.Natural conditions and farm information.2nd Edition.Vol.11/ C. Eastern Province.Ministry of agriculture/GTZ, Nairobi, Kenya.Jaetzold R, Schmidt H (1983) Farm managementhandbook of Kenya. Natural conditions and farm Jana, P.K., Ghatak, S., Barik, A., Biswas, B.C., Souda, G., Mukherjee, A.K., 1990.Response of summer groundnut to phosphorus and potassium. Indian Journal of Agronomy 16, 60-63. Jibrin, J.M., Chude, V.O., Horst W.J., and Amapu, I.Y. (2002). Effect of Cover Crops, Lime and Rock Phosphate on Maize (Zea mays L.) in an Acidic Soil of Northern Guinea Savanna of Nigeria. J. Agric. Rural Dev. Trop. Subtrop., 103(2): 169-176. Kamara, A.Y., Kwari, J., Ekeleme, F., Omoigui, L., and Abaidoo, R. (2008). Effect of phosphorus application and soybean cultivar on grain and dry matter yield of subsequent maize in the tropical savannas of north-eastern Nigeria.Afric. J. Biotech., 7 (15): 2593-2599. Kanyanjua, S.M., Ireri, L., Wambua, S., and Nandwa, S.M. (2002). Acidic soils in Kenya: Constraints and remedial options. KARI Technical Note No. 11. Nairobi, Kenya. Kihanda, F.M., Warren, G.P., and Micheni, A.N.A. (2007).Effects of manure application on crop yield and soil chemical properties in a long-term field trial in semi-arid Kenya.Bationo A (eds.), Advances in Integrated Soil Fertility Management in Sub- Saharan Africa: Challenges and Opportunities, pp. 471-485. Kihanda, F.M., Warren, G.P., and Atwal, S.S. (2004).The Influence of Goat Manure Application on Crop Yield and Soil Nitrate Variations in Semi-Arid Eastern Kenya. In Bationo A (eds) Managing nutrient cycles to sustain soil fertility in Sub-Saharan Africa. Academy Science Publishers (ASP). Nairobi. Kenya. Kimani, S.K., Macharia, J.M., Gachengo, C., Palm, C.A., and Delve, R.J. (2004). Maize production on the central highlands of Kenya using cattle manures combined with modest amounts of mineral fertilizer. Uganda J. Agric. Sci., 9(1): 480-490. Kisinyo, P.O., Gudu, S.O., Othieno, C.O., Okalebo, J.R., Opala, P.A., Maghanga. J.K., Agalo, D.W., Ng'etich, W.K., Kisinyo, J.A, Osiyo, R.J., Nekesa, A.O., Makatiani, E.T., Odee, D.W., and Ogola, B.O. (2012). Effects of lime, phosphorus and rhizobia on Sesbaniasesban performance in a Western Kenyan acid soil.Afric. J. Agric. Res., 7(18): 2800-2809. 365 Latif, M.A., Mehuys, G.R., Mackenzie, A.F., Alli, I., Faris, M.A. (1992). Effects of legumes on soil physical quality in maize crop. Plant and Soil, 140: 15-23. Liang W, Wu X, Zhang S, Xing Y, Wang R (2011). Effect of organic amendments on soil water storage in the aeolian sandy land of northeast China. Proceedings of the Electrical and Control E Engineering (ICECE), International Conference on 16th – 18th Sept. 2011. pp. 1538-1540. Mahasi, J.M., Vanlauwe, B., Mursoy, R.C., Mbehero, P., and Mukalama, J. (2010). Increasing productivity of soybean in western Kenya through evaluation and farmer participatory variety selection. In: LA Wasilwa, J Ouda, IW Kariuki, NS Mangale and MN Lukuyu (eds). Transforming Agriculture for Improved Livelihoods through Agricultural Product Value Chain.Proceedings of the 12th KARI Biennial Scientific Conference. November 8-12, 2010. Nairobi -Kenya. Mairura, F.S., Mugendi, D.N., Mwanje, J.I., Ramisch, J.J., Mbugua, P.K. (2007). Assessment of farmers´ perceptions of soil quality indicators within smallholder farms in the central highlands of Kenya. In: Bationo A., Waswa BS, Kihara J, Kimetu J (Eds.), Advances in integrated soil fertility management in Sub-Saharan Africa: Challenges and opportunities. Springer, Dordrecht, NL. pp. 1035-1046. Makokha, S., Kimani, S., Mwangi, W., Verkuijl, H., Musembi, F. (2001). Determinants of Fertilizer and Manure Use for Maize Production in Kiambu District, Kenya. D.F.: International Maize and Wheat Improvement Center (CIMMYT) and Kenya Agricultural Research Institute (KARI). Maurya, B.M., and Rathi, K.S., 2000. Growth and development of soybean as influenced by intercropping with pigeon pea and phosphorus level. Gujarat Agricultural University Research Journal 26(1), 1-5 Mugendi, E., Gitonga, N., Cheruiyot, R., and Maingi, J. (2010). Biological Nitrogen Fixation by promiscuous Soybean (Glycine max L. Merril) in the Central Highlands of Kenya: Response to Inorganic fertilizer soil Amendments. World J. Agric. Sci., 6(4): 381-387. Mugwe, J., Mugendi, D., Kungu, J., Mucheru-Muna, M. (2007). Effect of plant biomass, manure and inorganic fertilizer on maize yield in the Central highlands of Kenya African Crop Sci. J., 15 (3): 111-126. N2 Africa, Podcaster n°10. (2011). Putting nitrogen fixation to work for smallholder farmers in Africa, October 2001. 7pp. Nieuwenhuis, R., and Nieuwelink, J. (2002). Cultivation of soybean and other legumes.Agrodok-Series No. 10. Agromisa Foundation, Wageningen, The Netherland. Njeru, P.N.M. (2009). The effect of push-pull management and different fertilization strategies on maize crop yield in central Kenya. MSc Thesis. Uppsala University. Sweden. Nwachukwu, O.I and Ikeadigh, M.C. (2012).Water Use Efficiency and Nutrient Uptake of Maize as Affected By Organic and Inorganic Fertilizer. PAT, 8(1): 199-208 366 Okalebo, J.R, Gathua, K.W., and Woomer, P.L. (2002). Laboratory Methods of Soil and Plant Analysis: A working manual. 2nd Edition: TSBF-CIAT. SSSEA. KARI. Sacred Africa. Okwuagwu, M.I., Alleh, M.E., Osemwota, I.O. (2003). The effects of organic and inorganic manure on soil properties and yield of okra in Nigeria.Afric. Crop Sci. Confer. Proc., 6: 390-393. Onwonga, R.N., Lelei, J.J., and Mochoge, B.E. (2010). Mineral nitrogen and microbial biomass dynamics under different acidic soil management practices for maize production. J. Agric. Sci., 2(1): 16-30. Rahman, M.A., Meisner, C.A., Duxbury, J.M., Lauren, J., and Hossain, A.B.S. (2002). Yield response and change in soil nutrient availability by application of lime, fertilizer and micronutrients in an acidic soil in a rice-wheat cropping system. Paper presented in 17th WCSS, 14th -21st August 2002. Thailand. Rasoulzadeh, A., and Yaghoubi, A. (2010). Effect of cattle manure on soil physical properties on a sandy clay loam soil in North-West Iran. J. Food Agric. Environ., 8(2): 976 - 979. Reddy, P.R.N., and Giri, G., 1989.Response of soybean varieties to plant density and P. Indian Journal of Agronomy 33(4), 515-516. Risse, L.M., Cabrera, M.L., Franzluebbers, A.J., Gaskin, J.W., Gilley, J.E., Killorn, R., Radcliffe, D.E., Tollner, W.E., and Zhang, H. (2006). Land Application of Manure for Beneficial Reuse. Biological Systems Engineering: papers and Publications. Paper 65. Sacchidanand, B., Sawaker, N.J., Ghesraya, R.S., Shinde, D.A., and Sinha, S.B. (1980). Response of soybean to S and P level and inoculation with Rhizobium on nodulation, leghaemoglobin content and nitrogen uptake in fodder cowpea legume Research 12(1), 17-30. Salahin, N., Islam, M.S., Begum, R.A., Alam, M.K., and Hossain, K.M.F. (2011).Effect of tillage and integrated nutrient management on soil physical properties and yield under tomato-mungbean-t.aman cropping pattern. Int. J. Sustain. Crop Prod., 6(1): 58-62. Sanchez, P.A., and Jama, B.A. (2002). Soil fertility replenishment takes off in East and Southern Africa, pp. 23–45. In: Vanlauwe B, Diels J, Sanginga N and Merckx R (eds) Integrated Plant Nutrient Management in Sub-Saharan Africa. CAB International, Wallingford, UK. Sanginga, N. (2003). Role of biological nitrogen fixation in legume based cropping systems: a case study of West Africa farming systems. Plant and Soil, 252: 25-39. Sanginga, N., and Woomer, P.L. (2009). Integrated Soil Fertility Management in Africa: Principles, Practices and development process.Tropical Soil Biology and Fertility Institute of the International Centre for Tropical Agriculture. Nairobi. 263 pp. Sanginga, N., Okogun, J. A., Vanlauwe, B., and Dashiell, K. (2002). The contribution of nitrogen by promiscuous soybeans to maize based cropping in the moist savanna of Nigeria. Plant Soil 241: 223-231. 367 Shen, Q.R., and Shen, Z.G. (2001). Effects of pig manure and wheat straw on growth of mung bean seedlings grown in aluminium toxicity soil.Biores. Tech., 76: 235-240. Shirani, H., Hajabbasi, M.A., Afyuni, M., and Hemmat, A. (2002). Effects of farmyard manure and tillage systems on soil physical properties and corn yield in central Iran. J. Soil Till. Res., 68(2): 101-108. Singh, V.K., and Bajpai, R.R., (1990). Effect of P on the growth and yield of rainfed soybean. Indian Journal of Agronomy 35(3), 310-311. Tisdale, S.L., Nelson, W.L., Beaton, J.D., and Havlin, J.L. (1985). Soil fertility.5th Edition. Macmillan Publishing Company. USA. Umoetok, S.B.A., Uko, A.E., Archibong, B.E., Ukeh, D.A., and Udo, I.A. (2007). Effect of application of inorganic fertilizer and poultry manure on insect pests and yield of soybean (Glycine max L.) in the rain forest zone of Nigeria. J. Food Agric. Environ., 5(2): 149-152. Vanlauwe, B., Bationo, A., Chianu, J., Giller, K.E., Merckx, R., Mokwunye, U., Ohiokpehai, O., Pypers, P., Tabo, R., Shepherd, K.D., Smaling, E.M.A., Woomer, P.L., and Sanginga, N. (2010). Integrated soil fertility management.Operational definition and consequences for implementation and dissemination. Outlook Agric., 39(1): 17-24. 8.2 A comparative analysis of the determinants of gross margins from African Indigenous Vegetables in semi-arid and humid rural areas of Kenya Jamleck Oroko Osiemo1,2*, David Jakinda Otieno1, Rose Adhiambo Nyikal1, Komi Fiaboe2, Hippolyte Affognon2,3 1 Department of Agricultural Economics, University of Nairobi 2 International Center for Insect Physiology and Ecology 3International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Bamako, Mali *Corresponding author: osiemojamleck@gmail.com ABSTRACT This study assesses the economic returns accrued from production of African Indigenous Vegetables (AIVs) in rural semi-arid and humid areas of Kenya. Multi-stage sampling method was used to select 567 farmers from Busia, Machakos and Nyamira. Data collection was done using a semi-structured questionnaire through Open Data Kit (ODK). Analysis was done through descriptive statistics, analysis of variance (ANOVA), gross margin analysis and a multiple regression. The very humid Nyamira registered the highest gross margin per acre per annum (Ksh. 437, 231), followed by humid Busia (Ksh. 171, 651). Semi-arid Machakos registered negative gross margins per acre (Ksh. -55, 726). ANOVA revealed that gross margins were statistically different across the regions at 5%. In Busia, the gross margins were significantly influenced by number of AIVs grown for commercial purposes, household head’s years of schooling, 368 AIVs diversity, contract farming and market participation. In Nyamira, gross margins were significantly influenced by the distance to a water source, gender of household head, on-farm crop diversity, contract farming and AIVs diversity. In Machakos, household size, area under AIVs production, distance to nearest tarmac road, AIVs diversity, and primary occupation significantly influenced gross margins. Diversity of AIVs was the only factor that influenced gross margins in all the three regions. This study concluded that the region where AIVs are produced influence gross margins, hence the need for region-specific policy interventions. Key words: African Indigenous Vegetables, gross margins, OLS, ODK, Kenya. INTRODUCTION AIVs are vegetables that have had their original habitat in Africa (Gockowski et al., 2003). The vegetables have a wide range of advantages over exotic vegetables. Some of the advantages are; easy availability, high nutrition value and high resistance to pests and disease (Ayanwale et al., 2011; Onim and Mwaniki, 2008). Over 210 AIVs species have been identified in Kenya (Maundu, 1996). These species serve different purposes in various areas in the country. For instance, Kimiywe et al. (2006) observed that consumers attach medicinal value to AIVs in Nairobi. In Uasin Gishu, Mwamburi (2003) noted that AIVs are consumed due to the belief that they can cure many diseases such as stomach disorders, malaria, and ear and eye problems. The vegetables also contribute directly to the economy since they form part of the Horticultural Sector which earns about 36% of the Agricultural Gross Domestic Product (GDP) (GoK, 2012). Production is dominated by small scale farmers, mainly women in rural areas for subsistence purposes (Abukutsa et al., 2005; Abukutsa, 2003). However, due to increased awareness about the benefits of the AIVs, there has been increased demand for the vegetables (Virchow, 2008). Consequently, production, and marketing has spread to urban areas, like Nairobi and its surrounding (Irungu et al., 2007). This has led to AIVs shifting from being grown for subsistence to being cash crops (Ojiewo et al., 2013). Development of markets for AIVs provides an opportunity that can be utilized to reduce poverty and ensure nutrition security among the poor. Various studies that have looked on agricultural commercialization particularly (Beatrice and Anna, 2014; Rao and Qaim, 2011; Pingali and Mark, 1995) agree that commercialization, which is marked by more input and output market participation (Nepal and Thapa, 2009) improves the well-being of the farmers. Evidence from other countries such as Bangladesh has shown that vegetable production, mostly indigenous vegetables has improved livelihoods in the country (Wannasai and Shrestha, 2008; Rahim et al., 2007). This is not the case in Kenya since areas well known for the production of AIVs specifically Western and Eastern Kenya register the highest malnutrition and poverty levels (KIPPRA, 2013). It has not been empirically proven if the scenario persists due to low revenues from AIVs, or high production costs associated with AIVs production, hence this study. 369 MATERIALS AND METHODS This study was carried out in Nyamira, Busia and Machakos. The areas were purposively selected based on various criteria, namely; history of AIVs production, AIVs diversity, proximity to Nairobi and agro ecological conditions, where Nyamira and Busia represent humid regions and Machakos represent the semi-arid area. Multi-stage sampling was used in this study due to its appropriateness when the population is scattered over a vast geographic area, and precision in hypothesis testing (Allen et al., 2002). This approach has been used in studies like (Gotor and Irungu, 2010; Wanyama et al., 2010; Kibet et al., 2009) where it was not possible to get a sampling frame. A minimum sample of 384 arrived at using the Cochran approach (Cochran, 1977), was later adjusted to 537 as suggested in Israel (1992) to cater for farmers not contacted and non-response. Nevertheless, 567 respondents were interviewed following budget flexibility. Data analysis involved assessment of gross margins, analysis of variance and a multiple regression. Gross margins were arrived at after deducting variable costs as explained in Oseni and Aledwale, (2013). Labor costs were calculated using the market price for labor (Fox et al., 2005) and man hours spent. The hours were aggregated according to Braun (1991) and appropriate labor equivalences according to Abdulahi (1990) were used in cases where women and children provided labor. Factors influencing the gross margins were analyzed using multiple regression since it allows for discrimination between the influences of the exogenous variables on gross margin. The data used in this study fulfilled the requirements for OLS (Gujarati, 2003). 3. Results and discussions The overall gross margin was Ksh. 174, 353. Nyamira had the highest gross margin per acre (437, 231) compared to Busia (171, 651) and Machakos (-55, 726) (see table 4). The mean gross margin for Machakos was statistically different from those of Busia and Machakos at 5%. The high total earnings per acre as well as high gross margins in Nyamira may be attributed to high yields in the region as a result of relatively high input utilization as seen from the high expenditure on input in Nyamira. Labor costs formed the largest portion of the production costs in all the three regions mainly because vegetable production is labor intensive (Lumpkin and Weinberger, 2005), and valuation of own resources like family labor in this study. Labor productivities were statistically different at 5% across the three areas with Busia having the highest labor productivity (131) compared to Nyamira (73) and Machakos (-34). The main cause for higher labor productivity in Busia might be attributed to availability of other agricultural activities options in the county. The negative labor productivity in semi-arid Machakos is mainly associated with low yields due to low input utilization in addition to lack of alternative income sources. Table 4 shows the factors influencing gross margin generally in the three areas. For the very humid Nyamira, Distance to water source and Number of crops grown negatively influenced gross margins whereas Diversity of AIVs during the short season, 370 Gender of household head and Contract farming had a positive influence. Longer distances to water sources impair practices such as irrigation hence increasing production risks. Similar findings on the negative correlation between gross margins from vegetables and distance to water sources have been made by Mwaura et al. (2013) and Kundu et al. (2010). Growing many crops on the farm translate to more competition for resources by the crops hence the negative influence on gross margins. Notwithstanding, a wider diversity of AIVs grown had a positive correlation with gross margins as the farmer is able to spread both production as well as marketing risks. This means that specialization in production of AIVs would result to an increase in the gross margins accrued by the farmers. Contract farming also increases gross margins as it has been observed in other studies such as Rao and Qaim (2011) and Miyata et al. (2009) since it reduces transaction costs and marketing risks, and facilitates access to technical support by farmers (Mwaura et al., 2013; Ngugi et al., 2007). Being a female farmers as was found to reduce economic returns accrued from AIVs as was also observed by Mwaura et al. (2013) due to the facts that; male farmers are more likely to participate in markets (Bellemare and Barrett, 2006), and control the household resources and even dictate how the resources are allocated (Jecknoniah et al., 2013) making them more likely to use better and more inputs. In the humid Busia, gross margins were positively influenced by, Diversity of AIVs during the short season, Number of varieties grown for commercial purposes, Group membership, Credit access, Contract farming and Selling of AIVs, and negatively by Education years of the household and salaried employment. Education years of the household head had a negative influence on gross margins. The finding tarries with what Mwaura et al. (2013) found that learned people will tend to pursue other activities other than production of AIVs. Therefore, training on the potential the vegetables have to generate income might be a viable way of making AIVs production attractive even to the learned people. The positive influnce of selling AIVs (market participation) on gross margins may be attributed to the fact that market orientation promotes input utilization which in turn increase yields. This finding agrees with Bellemare and Barrett (2006) who observed that market participation increases farmer welfare. Group membership on the other hand increases farmers’ bargaining power such that farmers can sell to channels with more returns (Panda and Sreekumar, 2012) and also improve access to credit and information, and reduces transaction and transport costs (Fischer and Qaim, 2012; Ngugi et al., 2007), factors which increase the returns accrued from production of AIVs. The influence of contract farming, number of varieties grown for commercial purposes and diversity of AIVs is as in the explanation above for Nyamira. Income diversification (salaried income) through involvement in off farm activities influence gross margins from AIVs positively since farmers who do not depend on agriculture are likely to earn higher incomes than their counterparts solely depending on agriculture (Janvry and Sadoulet, 2001). Higher income among farmers is associated 371 with higher market participation (Rao and Qaim, 2011; Boughton et al., 2010) and also relatively higher input utilization hence high gross margins. For semi-arid Machakos, Household size, Area under AIVs, Log distance to tarmac, Number of crops grown negatively influenced gross margins whereas Diversity of AIVs grown during the dry season and Primary occupation influenced gross margin positively. The negative influence of household size on gross margin may be due to the fact that larger households reduce the quantity of vegetables sold in markets as was observed by Otieno et al. (2009). The negative correlation of gross margins with area under AIVs agrees with Haggblade et al. (2010) that large farms tend to yield low returns per hectare of land per year compared to small farms. The inverse relationship between farm size and returns may be attributed to the productivity of the land, amount of input requirement and market failures (Barret et al., 2010; Lamb, 2003). In addition, production of AIVs is labor intensive and as such an increase in the area would require hired labor where family labor may not be adequate. Consequently transaction costs on labor supervision will increase (Poulton et al., 2010), resulting to less gross margin and low input productivity. The inverse productivity relationship of AIVs with land dictate a proportionate increase in inputs required for the production of the vegetables for farmers to gain proportionately from the land increase (Boughton et al., 2010). Long distance to tarmac road results to low gross margins due to impaired market access and reduced group membership (Fischer and Qaim, 2010). Distance to tarmac road may also influence the availability of inputs, and thus reduce/increase input costs. Davis et al. (2010) observed that farmers who were far away from tarmac roads had a low participation rate in farmer field schools. These farmers were found to have not only low productivity but also low incomes. The effect of diversity of AIVs and the number of crops grown is as in the explanation for Nyamira. Diversity of AIVs grown during the short season was the only significant variable having the same influence in all the three regions. The importance of having a large AIVs diversity on gross margin is as been discussed earlier. 372 TABLES Table 1: Analysis of AIVs gross margin Variable Busia (n=190) Nyamira ( n=177) Machakos (n=200) Overall (n=567) Average total earnings/acre 310, 493 a 980,174 b 30,213 a 420, 683 Average labor cost/acre 106, 776 a 439, 548 b 76, 310 a 199, 910 Average fertilizer cost/acre 1, 254 a 10, 125 b 172 a 3, 641 Average manure cost/acre 8, 224 a 23, 093b 5, 375 a 11, 860 Ave. Planting material /acre 18, 798 ab 26, 880 b 3, 063 a 15, 770 Average pesticide cost/acre 622 a 36, 100 b 641 a 11, 704 Ave. Transport cost/acre 3, 148 ab 6, 198 b 325 a 3, 104 Ave. production cost/acre 138, 842 a 542,943 b 85, 939 a 246, 329 Average gross margin/acre 171, 651 a 437,231 b -55,726 a 174, 353 Ave. gross margin/unit labor 131 c 73 b -34 a 55 Source: Survey data, 2014 Note: The gross margins are in Kenyan shillings (Kshs 95 were equivalent to 1USD$ at the time of survey). The letters account for statistical difference in the means. Same letter across any two or all the three regions means that there is no significant statistical difference in the mean of the variable at 5% significance level. 373 Table 2: OLS regression results for each of the study regions Busia (n =190) Nyamira (n= 177) Machakos (n=200) Variable β P value Β P value β P value Household size 0.0492 0.303 0.1109 0.169 -0.2516 0.004** Distance to water source -0.0576 0.800 -1.2062 0.000*** -0.0709 0.402 Area under AIVs -0.8821 0.444 2.7956 0.332 -1.0968 0.000*** Irrigation use 0.4458 0.162 0.3608 0.254 0.3843 0.509 Fertilizer use 0.3869 0.281 0.1227 0.691 -0.4475 0.392 Pesticide use -0.0948 0.766 0.3257 0.555 0.4338 0.262 Gender -0.1094 0.743 0.8854 0.048** -0.2463 0.623 Log distance to tarmac -0.0764 0.709 -0.0189 0.815 -0.3355 0.022** Number of commercial variety 0.2435 0.006** 0.1031 0.539 0.4540 0.278 Number of crops grown -0.0375 0.710 -0.1926 0.088* -0.2772 0.085* Education years -0.0737 0.029** -0.0184 0.667 -0.0013 0.981 Group membership 0.5764 0.068* 0.0530 0.934 -0.0143 0.977 Diversity of AIVs short season 0.3546 0.000*** 1.3710 0.000*** 1.1774 0.033** Contract 1.2541 0.024*** 2.3993 0.001*** -0.3396 0.891 Selling of AIVs 1.2054 0.000** 0.4992 0.191 0.3199 0.431 Credit access 0.5451 0.083* 0.0369 0.924 0.3182 0.416 Land AIVs squared 0.5652 0.481 -2.3704 0.640 0.0898 0.044** OCU= Salaried employment -0.0144 0.969 0.2615 0.559 0.1852 0.737 OCU=Casual worker 1.2896 0.130 0.3737 0.450 2.1763 0.000*** Retail business 0.1947 0.597 0.0541 0.899 0.1787 0.733 Constant -0.6280 0.216 -2.3321 0.005*** -0.0714 0.947 No. Observations 190 No. Observations 177 No. Observations 200 F(20, 169) 13.18 F(20, 156) 6.53 F(20, 179) 4.35 Prob>F 0.0000 Prob>F 0.0000 Prob>F 0.0000 R2 0.3857 R2 0.4278 R2 0.3270 Root MSE 1.7189 Root MSE 1.8985 Root MSE 2.3741 374 4. Conclusions This study has shown that the income generation ability of AIVs depends on the region they are grown. The regions are characterized by different environments which influence the kind of production practices used hence yields, and proximity to markets which influence farmer’s decision whether to participate in output and/or input markets. Therefore, area specific interventions for promotion of AIVs production and marketing should be pursued. For the humid Busia, the institutional factors for instance farmers’ organization into groups, credit access and contract farming play a major role in enhancing economic returns from AIVs. This therefore calls for the county government and other organizations concerned with the promotion of AIVs to facilitate, and ease access to credit. Measures for instance, input subsidy, and promotion of table banking may provide viable options. Farmers should also be encouraged to participate in AIVs markets in order to improve their incomes through interventions such as construction of proper selling and collection points, and linking the farmers to markets for instance through contract farming. For the very humid Nyamira, intensification practices such as irrigation are recommended and provision of piped water to reduce distance to water sources. For the semi-arid Machakos, increase in land under AIVs has to be accompanied with a similar increase in utilization of inputs in order to gain proportionately from the land increase. Therefore strategies like input provision by the county governemnt and other actors should be sought for so as to increase returns from the vegetables. Birth control may also help in checking household sizes to enhance market participation. Acknowledgements The authors acknowledge the Deutsche Gesellschaft für International Zusammenarbeit (GIZ) for funding this study and Dr. Beatrice Muriithi of ICIPE for her technical support. References Abdulahi, A. M., 1990. Pastoral Production Systems in Africa: A study of nomadic household economy in central Somalia. Farming Systems in Resource Economics in the Tropics, 8. Abukutsa-Onyango, M. O., 2003. The role of African indigenous vegetables in poverty alleviation in Kenya. Proceedings of the First PROTA International Workshop, 23–25 September 2002, Nairobi, Kenya, pp.269–270 Abukutsa-Onyango, M., 2005. Seed production and support systems for African leafy vegetables in three communities in western Kenya. African Journal of Food, Agriculture, Nutrition and Development, 7(3). Allen, M., Kilpatrick, D., Armstrong, M., Briggs, R., Course, G., and Pérez, N., 2002. Multistage cluster sampling design and optimal sample sizes for estimation of fish discards from commercial trawlers. Fisheries Research, 55(1), 11-24. Ayanwale, A. B., Oyedele, D. J., Adebooye, O. C., Adeyemo, V. A., Tenywa, J. S., Taulya, G., and Santos, L., 2011. A socio-economic analysis of the marketing 375 chain for under-utilized indigenous vegetables in southwestern Nigeria. In 10th African Crop Science Conference Proceedings, Maputo, Mozambique, 10-13 October 2011. (pp. 515- 519). African Crop Science Society. Barrett, C. B., Bellemare, M. F., and Hou, J. Y., 2010. Reconsidering conventional explanations of the inverse productivity–size relationship. World Development, 38(1), pp.88-97. Boughton, D., Mather, D., Barrett, C. B., Benfica, R., Abdula, D., Tschirley, D., and Cunguara, B., 2007. Market participation by rural households in a low-income country: An asset-based approach applied to Mozambique. Faith and economics, 50(1), pp.64-101. Bellemare, M. F., and Barrett, C. B., 2006. An ordered Tobit model of market participation: Evidence from Kenya and Ethiopia. American Journal of Agricultural Economics, 88(2), pp.324-337. Cochran, W. G., 1977. Sampling Techniques, third edition, Wiley, New York. Davis, K., Nkonya, E., Kato, E., Mekonnen, D. A., Odendo, M., Miiro, R., and Nkuba, J., 2012. Impact of farmer field schools on agricultural productivity and poverty in East Africa. World Development, 40(2), pp.402-413. De Janvry, A., and Sadoulet, E., 2001. Income strategies among rural households in Mexico: The role of off-farm activities. World development, 29(3), pp.467-480 Fischer, E. and Qaim, M., 2012. Linking Smallholders to Markets: Determinants and Impacts of Farmer Collective Action in Kenya. World Development, 40(6), pp. 1255– 1268. Fox, P., Rockström, J., and Barron, J., 2005. Risk analysis and economic viability of water harvesting for supplemental irrigation in semi-arid Burkina Faso and Kenya. Agricultural Systems, 83(3), pp.231-250. Gockowski, J., Mbazo’o, J., Mbah, G. and Fouda Moulende, T., 2003. African traditional leafy vegetables and the urban and peri-urban poor. Food Policy, 28(3), pp221–235 Gotor, E., and Irungu, C., 2010. The impact of Bioversity International's African leafy vegetables programme in Kenya. Impact assessment and project appraisal, 28(1), pp.41-55.Government of Kenya, 2012. National horticulture policy. Government printers, Nairobi. Gujarati, D.N. 2003. Basic Econometrics 4th Edition McGraw Hill Gujarati, DN, (2003). Basic Econometrics. Haggblade, S., and Hazell, P. B., 2010. Successes in African agriculture: Lessons for the future. Intl Food Policy Res Inst. Irungu, C., Mburu, J., Maundu, P., Grum, M. and Hoeschle-Zeledon, I., 2007. Analysis of Markets for African Leafy vegetables within Nairobi and its Environs and Implications for On-Farm Conservation of Biodiversity, Rome, Italy: Global Facilitation Unit for Underutilized Species, pp.54. Israel, G. D., 1992. Determining sample size. University of Florida Cooperative Extension Service, Institute of Food and Agriculture Sciences, EDIS. 376 Jeckoniah, J., Mdoe, N., and Nombo, C., 2013. Mapping of Gender Roles and Relations along Onion Value Chain in Northern Tanzania. International Journal of Asian Social Science, 3(2), pp.523-541. Kibet, L. K., Mutai, B. K., Ouma, D. E., Ouma, S. A., and Owuor, G., 2009. Determinants of household saving: Case study of smallholder farmers, entrepreneurs and teachers in rural areas of Kenya. Journal of Development and Agricultural Economics, 1(7), pp.137-143. Kimiywe, J., Waudo, J., Mbithe, D. and Maundu, P., 2006. Nutraceutical benefits of traditional foods: A food-based strategy for alleviation of hunger and malnutrition. In The Role of African Universities in the Attainment of the Millennium Development Goals, Kenyatta University, Nairobi, Kenya, Proceedings of MDG Conference held in Nairobi, Kenya, 14–18 November 2005, pp.2–11 Lamb, R. L., 2003. Inverse productivity: Land quality, labor markets, and measurement error. Journal of Development Economics, 71(1), pp.71-95. Maundu, P., 1996. Utilization and conservation status of wild food plants in Kenya. The Biodiversity of African Plants, Proceedings of the 15th AETFAT Congress, 22–27 August 1994, Wageningen, Kluwer Academic Publishers, Dordrecht, The Netherlands, pp.678–683. Miyata, S., Minot, N., and Hu, D., 2009. Impact of contract farming on income: linking small farmers, packers, and supermarkets in China. World Development, 37(11), pp.1781-1790 Muriithi, B. W., and Matz, J. A., 2014. Smallholder participation in the commercialization of vegetables: evidence from Kenyan Panel Data. Available at SSRN 2405462. Mwamburi, A.L., 2003. Nutritional value and utilization of African Indigenous vegetables in Uasin Gishu District. In 3rd workshop proceedings on Sustainable horticultural production in the tropics; 26th- 29th November Maseno university Kenya. Nepal, R., and Thapa, G. B., 2009. Determinants of agricultural commercialization and mechanization in the hinterland of a city in Nepal. Applied Geography, 29(3), pp.377- 389. Ngugi, I. K., Gitau, R. and Nyoro, J. K., 2007. Access to High Value Markets by Smallholder Farmers of African Indigenous Vegetables in Kenya, IIED, London Ojiewo, C., Tenkouano, A., Hughes, J. D. A., and Keatinge, J. D. H., 2013. Diversifying diets: using indigenous vegetables to improve profitability, nutrition and health in Africa. Diversifying Food and Diets: Using Agricultural Biodiversity to Improve Nutrition and Health. Onim M. and P. Mwaniki, 2008. Cataloguing and Evaluation of Available Community/Farmers-Based Seed Enterprises on African Indigenous Vegetables (AIVs) in Four ECA Countries. Entebbe, Uganda: Association for Strengthening Agricultural Research in Eastern and Central Africa. 377 Otieno, D. J., Omiti, J., Nyanamba, T., and McCullough, E., 2009. Market participation by vegetable farmers in Kenya: A comparison of rural and peri-urban areas. African Journal of Agricultural Research, 4(5), pp. 451-460. Pingali, P. L., and Rosegrant, M. W., 1995. Agricultural commercialization and diversification: processes and policies. Food policy, 20(3), pp.171-185. Poulton, C., Dorward, A., and Kydd, J., 2010. The future of small farms: New directions for services, institutions, and intermediation. World Development, 38(10), pp.1413- 1428. Rahim, M.A., Anwar M., Naher N., and Islam F., 2007. Indigenous vegetables play a great role to overcome poverty level in flood and hunger prone (Monga) areas of Bangladesh. Rao, E. J., and Qaim, M., 2011. Supermarkets, farm household income, and poverty: insights from Kenya. World Development, 39(5), pp.784-796. Virchow, D., 2008. Indigenous vegetables in East Africa: sorted out, forgotten, revitalized and successful. New Crops and Uses: Their role in a rapidly changing world, 79. Von Braun, J., 1995. Agricultural commercialization: impacts on income and nutrition and implications for policy. Food policy, 20(3), pp.187-202. Wannasai, N., and Shrestha, R. P., 2008. Role of land tenure security and farm household characteristics on land use change in the Prasae Watershed, Thailand. Land Use Policy, 25(2), pp.214-22 Wanyama, M., Mose, L. O., Odendo, M., Okuro, J. O., Owuor, G., and Mohammed, L., 2010. Determinants of income diversification strategies amongst rural households in maize based farming systems of Kenya. African Journal of Food Science, 4(12), pp.754-763 8.3 Determinants of Intensity of Participation in Avocado Producer and Marketing Groups in Murang’a County, Kenya Catherine Nyambune Maindi, Daniel Kyalo Willy, Lucy Wangare Ngare Department of Agribusiness Management and Trade, Kenyatta University, P.O. Box 43844-0100, Nairobi, Kenya. Correspondence: Email: catherinemaindi@yahoo.com, willy.kyalo@ku.ac.ke , wangare.lucy@ku.ac.ke, ABSTRACT This study analysed the determinants of intensity of participation in avocado producer and marketing groups (APMGs) in Murang’a County, Kenya. Intensity of participation would offer insights on member commitment and free riding to the APMGs. Member commitment in group marketing is critical to planning and implementing production and marketing functions in order to achieve successive collective marketing. Otherwise 378 the groups may become dysfunctional or even collapse. The number of group meetings attended, quantity of individual sales made and amount of annual contributions made were used as proxies for member commitment. A total of 110 members were sampled proportionately to the group size. The data was collected using a structured questionnaire then analysed using Double Hurdle model in STATA 13. The results revealed that the ideal group size should constitute of 14 to 16 members for optimum participation. Farm size, membership density, group age, trust index, group size, homogeneity index, decision making index, fines existence and provision of free inputs were also key determinants of intensity of participation in the APMGs. Policy implications drawn by this study emphasizes on the importance of the smaller group size, decision making index, trust and the role of social norms and values in enhancing member commitment and participation in collective initiatives. The notable policies will serve to facilitate success in collective marketing and longevity of the groups. Keywords: Collective action, Avocado Producer and Marketing Groups, intensity of participation, free riding, Double-Hurdle model INTRODUCTION Collective actions through producer and marketing groups is one of the interventions adopted to enable the smallholders address high transaction costs and other market imperfections and contend with modernization of agri-product markets (Shiferaw et al., 2009). The smallholders in Kenya are the major actors constituting up to 80% in agricultural sector. They form an integral component of backward and forward linkages to other economic sectors besides their contribution to rural employment and food supply which is critical to transform the rural economy (GoK, 2010). Despite this, smallholders face challenges like high transaction costs, coordination failure, weak institutions, lack of/asymmetric information, reduced government investments coupled with the recent globalization of agri- commodity markets (Poulton, 2014; Wanyama, 2014). Consequently the smallholders are exposed to market risks, diseconomies of scale and lack of competitive advantages (Shiferaw et al., 2009; Poulton, 2014). It is against this backdrop that producer and marketing groups have gained popularity to empower farmers to consolidate their resources to achieve scale economies, competitive advantages, and meet demands of continuous supply and larger volumes consistent with stringent quality and standards required (Poulton, 2014; Wanyama, 2014). Research shows that marketing through farmer groups increased members’ total annual income up to 27% and 16% from bananas and French beans respectively (Fischer and Quaim, 2014; Korir et al., 2015). Additionally, farmer groups enhanced marketing performance in Tanzania (Barham and Chitemi, 2009), better prices in Costa Rica and high profits as a result of quality standards and consistent bulky export marketing of coffee in Brazil (Wollni and Zeller, 2006). Despite this, empirical evidence shows that the producer and marketing groups are either dysfunctional or are not sustained long enough to achieve members’ interests but end up disintegrating (Fischer and Quaim, 2014; Barham and Chitemi, 2009).This may be attributed to free riding. For successful collective action to be realised in agricultural marketing, better planning and 379 implementation of production and marketing functions, enforcing contractual agreements and conflict resolutions are critical. To achieve this, member commitment and contributions in various group activities is critical to achieve the perceived common goal and hence viability of the groups in the long run. Therefore, the objective of this study was to analyse the determinants of intensity of participation in collective action using an example of avocado producer and marketing groups (APMGs) in Murang’a County, Kenya. Several studies analysed the intensity participation and free riding in collective action. One study analysed extent of free riding in terms of time and money contributed which assumed a categorical variable (Vorlaufer et al., 2012), quantity of collective sales (Mukundi et al., 2013), quantity of collective sales made and the number of meetings attended (Fischer & Quaim, 2014). However, still exists a dearth of empirical evidence on determinants of intensity of participation in producer and marketing groups. This study therefore addressed the gap by assessing how individual characteristics, group characteristics and policy and institutional factors influence the frequency of group meetings attendance, amount of annual contributions besides the quantity of sales made collectively. Policy and practices are further discussed in the last section. Material and Methods Study area: The study was conducted in Murang’a County, Kenya. Murang’a County government pilots collective action in smallholder agricultural marketing. However, avocado fruits have received renewed attention owing to the area’s potential to produce avocadoes with high oil and wax content. In addition to that, there is ready domestic and export markets (Kakuzi, Olivando and Vegpro companies). The export companies contracts APMGs which are legally registered through Avocado Growers Association of Kenya (AGAK) to sell their avocadoes collectively at designated collection centers. Data collection and analysis: The data was collected using a pre tested questionnaire from the 110 proportionately sampled group members. The double hurdle model was used for the analysis. The model assumes that households make two sequential decisions which are modeled using different variables. First, decision to participate, in this case is whether a member was involved or not in group meetings, annual contributions and group sales. Secondly, decision on intensity of participation which took a value denoting the number of meetings attended quantity of group sales in proportion to the total marketable surplus and amount of group annual contributions. The parameters were then estimated using Stata version 13. RESULTS AND DISCUSSION 3.1 Determinants of intensity of participation in APMG activities Table 1 shows the results for the determinants of participation APMG activities. The farm size had a positive influence on intensity of participation in annual contributions, quantity of sales and number of meetings. Large farm sizes have high incentives because of bulky production which outweighs costs of participation in group as a result 380 of scale economies (Mukundi et al., 2013, Fischer & Quaim, 2014)). A unit increase in trust index increases the amount of annual contributions made by 65.7% points. Trust serves to enhance individual commitment, coordination, and cooperation among group members. Moreover, trust among the actors reduces the risk factors hence they are bound by mutual obligation and reciprocity (Vanni,2014). In regard to group age, older groups may have stronger ties associated with shared norms, trust and values unlike newer groups with greater uncertainty of reciprocity internal cooperation (Vorlaufer et al., 2012, Korir et al., 2015 and Markelova and Meinzen, 2009).The effect of membership density was positive for the amount of annual contributions and share of group sales. Membership densities are an avenue through which social capital is accumulated. Social capital facilitates to minimize opportunistic behaviour, dissemination and sharing of information and collective decision making ((Markelova and Meinzen, 2009). Fine existence had significant (P<0.05) and positive effect on the share of annual contributions made and number of meetings attended. Fines enhance accountability and compliance within the group’s crafted rules. On the other hand, each additional unit of free input, increases share of annual contributions made to the group by 86.3%. Free inputs act as inducements to members to intensify participation in group activities (Markelova and Meinzen, 2009). An increase in homogeneity index by one unit increased the number of meetings attended by 58%. This is in view that, groups with members of similar social, cultural, political or economic characteristics are more stable and cohesive (Markelova and Meinzen, 2009). In regard to group size, each additional group member decreases the individual quantity of collective sales and number of meetings attended by 2.1% and 0.2% respectively. This upholds the findings of Fischer (2014), Markelova and Meinzen (2009) and Korir (2015) that unlike larger groups, smaller groups are more cohesive which enhances coordination and easier monitoring of other members hence reduced incentives of free riding. 381 Table 14: Double Hurdle results: Factors influencing intensity of participation in group activities Amount of annual contributions Quantity of group sales Number of meetings attended Coeff. Std. Error p- value Marginal Effects Coeff. Std. Error p- value Marginal Effects Coeff. Std. Error p- value Margina l Effects Farm size 0.478* 0.022 0.037 0.015 0.110** 0.411 0.008 0.759 -0.245 0.024 0.297 -0.047 Trust index 0.836** 0.367 0.001 0.657 0.296 0.529 0.577 0.965 0.386 0.370 0.299 0.033 Group size -0.004 0.009 0.662 -0.001 -0.235*** 0.005 0.000 -0.021 -0.003** 0.001 0.002 -0.002 Gender 0.092 0.078 0.243 0.031 -0.686 0.119 0.567 -0.005 0.250 0.082 0.759 0.030 Group age 0.015** 0.004 0.001 0.002 0.199** 0.005 0.001 0.018 0.004 0.044 0.916 0.002 Off farm income 0.022 0.057 0.695 0.283 -0.093 0.104 0.372 -0.957 0.030 0.607 0.620 0.062 Homogeneity index 0.003 0.002 0.136 0.072 0.247 0.352 0.486 0.227 0.549* 0.216 0.012 0.580 Membership density 0.069* 0.027 0.013 0.005 0.148** 0.049 0.003 0.118 0.016 0.028 0.565 0.002 Free input 0.202* 0.096 0.037 0.863 0.059 0.127 0.646 0.104 0.078 0.984 0.426 0.036 Decision making index 0.734** 0.208 0.001 0.023 1.803*** 0.351 0.000 0.184 0.862*** 0.219 0.000 0.826 Fine existence 0.191** 0.661 0.005 0.058 0.001 0.113 0.994 0.089 0.187** 0.069 0.007 0.009 Timely payment -0.125 0.101 0.220 -0.325 -0.119 0.123 0.336 -0.055 -0.246 0.138 0.076 -0.519 Constant -0.648* 0.323 0.047 -0.007 0.313 0.572 0.586 0.821 0.296 0.320 0. 356 0.930 Model Summary No. of observations 137 137 137 LR χ2 (11) 170.14 164.82 140.35 Prob > chi2 0.0000 0.0000 0.0000 Pseudo R2 0.6581 0.6820 0.6250 Log likelihood -44.189 -38.428 -67.208 *, ** and *** denotes significance levels at 5%, 1% and 0.1% respectively. 382 Decision making index was strongly associated with the intensity of participation in the three measured group activities. Decision making among the group members fosters internal democracy and among the actors, decision making index facilitates in creating and reinforcing “exploitation-free” contractual agreements between group members and export agents. 3.2 Group Size and Intensity of Participation in APMG Activities To estimate the relationship between group size and quantity of sales made through the group, bivariate non-parametric regression was used in Distributive Analysis Stata Package (DASP). This method was used to estimate the local derivative of the group size with respect to the quantity of group sales without having to specify the functional form linking them. Regression with the non-parametric curves module was performed with Nadaraya-Watson approach. This approach allows observations to enter and exit the model smoothly by selecting the bandwidth, which controls the degree of smoothing (Araar A. et al., 2013). Figure 1 shows the relationship between group size and the individual sales made through the group. Figure 1: Relationship between group size and group sales The findings reveal that the ideal group size should constitute of 14 members for optimum intensity of participation in the group sales. Below 14 members, despite the stronger ties, the smaller group size implies lower economies of scale. On the other hand, although groups of more than 14 members enjoy economies of scale, they tend to have weaker social cohesion. This promotes the incentives to free ride due to inability to monitor the behavior of peers. 8 0 0 2 0 4 0 6 0 G ro u p S a le s ( K G ) 25% 50% 75% 95% 15 20 25 30 35 400 105 Group Size (Number of Members) (Nadaraya-Watson Estimation Approach | Bandwidth = 2.79 ) 383 Figure 2: Relationship between group size and group meeting attendance Figure 2 shows the relationship between group size and number of meetings attended. Adding an extra group member results to increase in number of meetings attended. However, this increase is up to the point where the group size is about 16 members. Additional group members enhance the members’ voice and solidarity due to enhanced interpersonal interactions which strengthens their social ties (Korir, 2015). Beyond 16 members, for each additional member results to a decline in the number of meetings attended. This may be explained by the weaker social cohesiveness and difficulties in monitoring other members’ behavior which increases the monitoring costs which also makes larger groups, in this case, of more than 16 members not ideal. 5. Conclusion and Policy Implications The results revealed that a group size of 14 to 16 members is ideal for optimum participation. Membership to other groups and networks, democratic participation in group functioning, similarly in members’ identity, trust, imposing fines to non- compliant members and provision of free inputs emerge as key determinants of intensity of participation in the avocado producer and marketing groups. Policy implications drawn from the study enhance strategies to intensify participation in collective action for successful collective marketing and sustainability of the groups in the long run. First the group size should constitute 14 to 16 members. Strategies to foster democratic participation of members in group functioning, transparency and accountability systems are vital to assure members on governance and financial management aspects. Furthermore, the existing extension policies in collaboration with the respective actors need to identify and integrate the strategies that recognize the role of the social norms and values in their training and dissemination of knowledge. Finally, the contracting export agents should also honor their contractual agreements to heighten trust from the groups. 0 2 0 4 0 6 0 8 0 M e e tin g s A tt e n d e d ( N u m b e r) 25% 50% 75% 95% 405 10 15 20 25 30 350 Group Size (Number of Members) (Nadaraya-Watson Estimation Approach | Bandwidth = 2.84 ) 384 References Barham, J. and Chitemi, (2009). ‘Collective Action Initiatives to Improve Marketing Performance: Lessons from Farmer Groups in Tanzania’, Food Policy, Vol. 34, pp. 53–59. Fischer, E. and Qaim, M. (2014). ‘Smallholder Farmers and Collective Action: What Determines the Intensity of Participation?’ Journal of Agricultural Economics Vol.65, No. 3,683-702. Government of Kenya, GoK. (2010). Agricultural Sector Development Strategy, 2010- 2020. Korir C.H., Lagat K.J., Mutai C.M., and Ali O. W., (2015), ‘Influence of Social Capital on Producer Groups’ Performance and Market accesss Amongst Smallholder French Beans Farmers in Kirinyaga County’. Journal of Economics and Suatainable Development Vol.6 No.2 Markelova, H., Meinzen-Dick, R., Hellin, J., and Dohrn, S. (2009). ‘Collective Action for Smallholder Market Access’. Food Policy, 34(1), 1–7. Mukundi, E., Mathenge, E., and Ngigi, M. (2013). ‘Sweet Potato Marketing among Smallholder Farmers: The Role of Collective Action’. Invited paper presented at the 4th International Conference of the African Association of Agricultural Economists, September 22-25, 2013.Hammamet, Tunisia. Poulton (2014), ‘International Food Policy Research Institute.The Role of Cooperatives in Modernizing Supply Chains. Sustainable Solutions for Hunger and Poverty’. Shiferaw B., Obare G., Muricho G., and Silim S. (2009). ‘Leveraging institutions for collective action to improve markets for smallholder producers in less-favored areas’. Afjare Vol.3 No. 1 March. Vanni, F. (2014). ‘Agriculture and Public Goods. The role of Collective Action’. XV, 150 p.26. ISBN: 978-94-07-7456-8. Vorlaufer M., Wollni, M., and Mithofer, D.(2012). ‘Determinants of Collective Marketing Performance: Evidence from Kenya’s Coffee Cooperatives’. Georg-August University Göttingen. Wanyama (2014), ‘International Food Policy Research Institute, Sustaining Solution for Ending Hunger and Poverty “Collective Action among African Smallholders Trends and Lessons for Future Development Strategies’. Wollni M., and Zeller M. (2006), ‘Do Farmers Benefit from Participating in Specialty Markets and Cooperatives? The Case of Coffee Marketing in Costa Rica’. 385 9. Parallel Session 8: Crop Management, Biotechnology and Climate Change 9.1 Evaluation of the effect of (A. indica, A. sativum and L. camara ) leaf extracts on root-knot nematode management on tomatoes in Mwea-East, Kenya. Mavuze Maxwell Mbalanya1, Birgen Jornah1 and Kariuki George2 1Department of plant sciences, Kenyatta University, 2Department of Agricultural sciences Kenyatta University. Abstract Tomato (Solanum lycopersicum L.) is an annual sub-tropical vegetable grown in Kenya for both domestic and export market. Its production in Kenya has suffered a major blow due to the high disease incidences caused by nematodes. Root-knot nematodes are known to cause greater damage to vegetables. The current methods of nematode control involve use of chemical nematicides but the major concerns have been the arising problem of pollution as a result of the use of chemicals and development of nematicide resistant strains arising from excessive application of chemical nematicides. This makes it important to research on new and effective methods of nematode management that are less risky to the environment. The objective of this study was to evaluate the effect of the plant extracts of A. indica, A. sativum and L. camara on root-knot nematodes affecting tomatoes in Mwea-East, Kenya. Field experiments were done to determine the efficacy of the extracts on the control of Meloidogyne spp. Treatments included A. indica, A. sativum and L. camara extracts, a commercial nematicide, vydate and untreated control. The experimental design was a completely randomized block design involving five treatments. Data on growth and Meloidogyne spp. infestation was collected. All the extracts significantly (p≤ 0.05) caused more growth as indicated by the higher shoot height, root length and dry weight when compared to the untreated plants. In the experiments the shoot height and root length showed significant difference (p≤ 0.05) with L. camara treatment recording the highest values. Galling index, egg mass index, final J2 in the soil and the reproductive factor differed significantly among the treatments (p≤ 0.05). The treatment with L. camara gave the highest yield and greatly reduced nematode damage. This results show that plant extracts are an alternative to chemical nematicides that cause pollution to the environment. Keywords: Tomato; Plant extracts; Root-knot nematode; Growth. INTRODUCTION Tomato (Solanum lycopersicum L.) is an annual sub-tropical vegetable. Its origin is in the Southern America Andes. The crop was introduced in East Africa by the colonialists in early 1900 (Wamache, 2005). It was first used as food in Mexico and these later spread to other parts of the world following the colonization of the Americans by the Spanish. Tomatoes grow well in warm conditions and its varieties are now found growing in many regions of the world sometimes in greenhouses in regions that have very cool climates. Uses of tomatoes may range from its use as an ingredient in dishes, as a salad, a drink or as a sauce. There are considerations of 386 tomato to be a vegetable for culinary purpose even though it is classified as a fruit botanically (Smith, 1994). Many plant parasitic nematodes attack tomato plants growing in different regions. Tomato plants are known to be the most favorable hosts for root-knot nematodes (Dropkin, 1980, Dnyaneshwar et al., 2010). The root-knot nematode groups known to attack tomatoes include the 4 major species of Meloidogyne (M. hapla, M. arenaria, M. incognita and M. javanica) and their known races in both indoor and outdoor cultivations (Onkendi et al., 2014). The use of synthetic nematicides has been the common practice in the management of plant parasitic nematodes. The indiscriminate use of these pesticides in nematode control has resulted in environmental pollution, phytotoxicity and nematode resistance (Conway, 1995; Yudelman et al., 1998). Some synthetic pesticides are also very expensive and some farmers are not able to buy them (Koon-Hui et al., 2010). The study was therefore done to evaluate on the possibility of plant extracts being used as alternatives for the management of root-knot nematodes in tomatoes. The plant extracts are considered to be more ecologically friendly and do not cause pollution to the environment. MATERIALS AND METHODS Study area The study was carried out in Mwea East, Kirinyaga County, Kenya. Samples collection and processing Sampling was done using purposeful sampling method in farms where tomatoes were grown. Samples of diseased tomato plants and soils surrounding the roots were collected from the farmers’ fields in Mwea East. A composite sample was obtained by mixing the sub-samples collected from the same farm. For further processing, the composite samples were then taken to the laboratory at Kenyatta University. Plant extracts preparation The mature leaves of L. camara were collected from Kenyatta University while the leaves of A. indica were collected from Garissa- North Eastern region of Kenya due to their high presence. The bulbs of A. sativum were bought from the local market in Mwea town. Some of the collected plant materials were brought to the lab for identification by a plant taxonomist. The remainders were thoroughly washed in sterile distilled water to remove soil and other debris. The leaves of A. indica and those of L. camara were then left to air dry for five days under a shade. A sterile electric blender was then used to ground the dried leaves to powder. A. sativum bulbs were washed in distilled water that had been sterilized and then taken in an oven for drying. Temperatures were set at 400C and left for 48 hours .The dried bulbs were then ground in to powder. A 100 milliliter of sterilized distilled water was placed in different one litre conical flasks. A 20 g powder of the plant extracts was then added to the stoppered conical flasks and left for 24 hours. After a 24 hour soaking, filtration was done using cheese cloth and the extracts stored in the refrigerator until use. The extracts were considered standard solutions “S” (100% concentration) (Mukhopadhyay and Tiwari, 2001 and Taye et al., 2012). 387 Field and experimental plot preparation The field experiment was conducted at Vick Agricare Limited Research farm in Mwea East, Kirinyaga County. The initial population of Meloidogyne spp. in the field was assessed by sampling soils before application of treatments. Tomato seedlings for use in these experiments were raised in a nursery. Experimental plots measured 16 m2. Each plot contained 3 planting rows, each row with 7 plants. There were a total of 21 plants per plot. In each planting hole, 5 g of DAP fertilizer was applied and mixed thoroughly with soil during transplanting. After transplanting, 30ml of each plant extract was applied at each plant station. Vydate was used as the positive control and was applied at a rate of 8 litres/ha while untreated plots served as the negative controls. A completely randomized block design with three replicates was used for the plots. Harvesting commenced two months after transplanting. After each harvest, the fruits were weighed using an electric balance scale. In order to determine the total yield, the total number of fruits per plant was also recorded. Ninety (90) days after transplanting, the experiments were terminated and data on tomato growth and Meloidogyne spp. reproductive parameters recorded. Measurement of plant growth parameters During termination of the field experiment, shoot height was measured from the base of the stem (soil line) to the newest apical node. Roots were then separated from shoots. Shoot fresh weight per plant was determined using a weighing balance. All stems and root systems were then oven dried at 60oC until constant weight was obtained. Root and shoot dry weights were determined separately. Data on fruit weight per block and number of fruits per plant were obtained. Determination of the galling index The galling index was determined using the scale of 0 to10 as described by Bridge and Page (1980). On this scale 0 = No galls on roots, 1 = few small galls observed upon close examination, 2 = small galls which are clearly visible while the main root is clean, 3 = numerous small galls observed, 4 = numerous small galls and a few big ones seen, 5 = 25% of the root system galled, 6 = 50% of the root system severely galled, 7 = 75% of the root system severely galled, 8 = no healthy root, but plant still green in colour, 9 = root system completely galled and plant dying, and 10 = plants and roots dead. Determination of the egg mass index Egg mass index was done by counting the egg masses in root system. Phloxine B solution (0.15 g/l) was used for staining the egg masses on the roots for period of about 15 minutes (Holbrook et al., 1983. Egg masses were counted under a microscope and scored using a scale of 1-9 where: 1 = no egg masses, 2 = 1-5, 3 = 6- 10, 4 = 11-20, 5 = 21-30, 6 = 31-50, 7 = 51-70, 8 = 71-100 and 9 = >100 egg masses per root system (Sharma et al., 1994). Determination of the population of root-knot juveniles in soil The nematode juveniles were extracted from different soil samples collected from the different experimental units using the method suggested by Coyne et al., (2007). 388 Two hundred (200) cm3 of fine soil was taken and placed on a two layer of serviettes placed on plastic containers with perforations which were then placed on plastic plates on a bench (Nchore et al., 2011). Water was then added to the extraction plates until the soil became dump. The set up was then left for 48 hours for the nematode migration to take place. After the period, the contents of the tray containing the nematodes were sieved using a 38 μm mesh into a beaker and allowed to settle. After an hour, the excess water was drained by use of a dropper to make the suspension to 20 ml for each set up. This nematode suspension was used for counting. Counting of the juveniles was done using a compound microscope under low power objective lens (x10). The nematode population was then estimated based on the totals obtained. Reproduction factor The nematode reproduction factor (RF) was determined and expressed as Rf = Pf/Pi, where Rf = reproduction factor, Pf = final population of nematode at harvest and Pi = initial J2s inocula. DATA ANALYSIS Data collected was subjected to Minitab statistical software program. Variation among the means was established using Analysis of variance (ANOVA). Tukey’s HSD was used to separate means where significant differences occurred. Mean values significantly different at P ≤ 0.05. RESULTS Effects of A. indica, A. sativum and L. camara extracts on tomato growth parameters The results showed that the number of plant leaves and the fruits in all treatments did not differ significantly (P ≤ 0.05, Table I). L. camara and Vydate treatments recorded a higher (P ≤ 0.05) stem height when compared to the other treatments. The untreated control and A. sativum had the shortest plants. Shoot dry weight, Root length and root dry weight were significantly higher (P ≤ 0.05) in treatments with L. camara and vydate. The lowest values were recorded by the untreated plants. Effects of plant extracts of A. sativum, L. camara and A. indica on galling index and egg mass index on tomatoes The results from the experiment showed that, galling index and egg mass index were significantly lower in the treated plants than in the untreated control (P ≤ 0.05, Table II). The lowest galling and egg mass indices were recorded in plants under vydate and L. camara treatments. No significant differences (P ≤ 0.05) were observed in the galling index of plants under A. sativum and A. indica treatments (Table II). Effect of the extracts on the population of juveniles in the soil, reproductive factor and the average fruit yield The number of juveniles, reproductive factor and the average yields were determined. A significant difference (P ≤ 0.05) was observed in the treatments for the average yield in the experiment (Table III). Plants treated with L. camara recorded a 389 higher yield (P ≤ 0.05) than the other treatments. The lowest yield was with the untreated plants. The number of juveniles in the treatment with L. camara was not significantly different from that with A. sativum (P ≤ 0.05) however it was lower (P ≤ 0.05) than the untreated control. The value of the reproductive factor was significantly (P ≤ 0.05) lowest in L. camara and Vydate, but significantly highest in the untreated control (Table III). The untreated plants gave a significantly lower (P ≤ 0.05) average yield but different significantly (P ≤ 0.05) in the other treatments with L. camara treated plants recording the highest yield (Table III) Tables Table I: Mean values for growth parameters of tomatoes grown in the field naturally infested with Meloidogyne spp. and treated with extracts from A. indica, A. sativum and L. camara. Growth parameters of tomatoes treated with different plant extracts and the positive and negative controls Treatme nt No. of leaves Stem Ht. Root length Shoot dry wt. Root dry weight Fruit number A. indica 45.73±4.29 a 80.87±5.38 b 8.6±0.19c 19.93±0.53 c 7.36±0.28 d 29.80±3.37 a A. sativum 42.27±3.28 a 76.53±5.0bc 9.73±0.28b 19.77±0.45 c 8.14±0.07c 30.73±4.91 a UC 56.73±3.35 a 75.47±5.43 bc 8.23±0.33c d 19.31±0.50 d 7.33±0.35 d 26.07±2.63 a L. camara 56.73±8.28 a 95.60±5.28a 12.73±0.55 a 28.80±1.75 a 9.82±0.22a 40.33±4.94 a Vydate 30.47±4.9a 92.20±3.42a 10.27±0.30 b 24.13±1.22 b 8.99±0.51a b 32.27±4.64 a F-value 1.66 3.46 25.46 15.79 6.84 1.58 P-value 0.169 0.012* 0.0001* 0.0001* 0.0001* 0.190 Mean measurement of five replicates (lengths in cm, weights in grams) Mean values denoted by the same letters are not significantly different at P ≤ 0.05 UC- untreated control. 390 Table II: Mean galling index and egg mass index on tomatoes grown in the field naturally infested with Meloidogyne spp. and treated with extracts of A. indica, A. sativum, L. camara , Vydate and untreated control Treatment Nematode parameters taken on tomatoes under different treatments with plant extracts and positive and negative control Galling index Egg mass index A. indica 2.53 ± 0.77b 3.4 ± 0.47b A. sativum 2.73 ± 0.49b 2.4 ± 0.40c Untreated control 8.07 ±0.52a 5.53± 0.31a L. camara 2.33 ± 0.83c 2.27 ±0.30c Vydate 1.46 ± 0.74c 1.87 ±0.41c F-value 14.90 14.90 P-value 0.0001* 0.0001* Mean value denoted by the same letters are not significantly different at P ≤ 0.05 Table III: Mean number of Juveniles, reproductive factor and average yields of tomato fruits in the field under treatment with extracts of A. indica, A. sativum, L. camara, Vydate and untreated control Treatment Number of Juveniles Reproductive factor Average Yields (Kg) per block A. indica 166.7 ± 18.6b 0.64 ± 0.07b 10.67 ± 1.67ab A. sativum 140.3 ± 19.4bc 0.54 ± 0.07bc 10.83 ± 2.17ab Untreated control 360 ±35.5a 1.39 ± 0.14a 7.0 ± 0.58b L. camara 120.67± 8.67c 0.46 ± 0.03bc 13.0 ± 1.53a Vydate 52.7 ± 16.6d 0.20 ± 0.06d 8.33 ± 0.33b F-value 28.54 28.54 4.30 P-value 0.0001* 0.0001* 0.018 Mean measurement of the parameters in three blocks Mean values denoted by different letters are significantly different at P ≤ 0.05 DISCUSSION Effect of the plant extracts of A. indica, A. sativum and L. camara on plant growth parameters The results demonstrated that the plant extracts enhanced plant growth as shown by the increased plant root length, shoot height, number of leaves, dry root and shoot weight when compared to the untreated control. Plants treated with L. camara extracts showed greater enhanced growth and increase in height than the A. indica and A. sativum. This results support those mentioned by Wondimeneh et al., (2013) who observed that application of plant extracts led to an increased plant height when compared to the untreated plants. 391 The increase in infestation as reflected by the number of galls might be the reason for the reduction of plant height in the untreated control. Distortion of the plant roots and nematode colonization of vascular bundles could reduce supply of nutrients to plants. The galls that are formed on the roots interfere with the functioning of the roots hence hindering important functions like water and nutrients uptake (Sikora et al., 2005). The botanicals provide a suitable environment for the roots to absorb and utilize the soil nutrients that results in reduced damage caused by nematodes (Abubakar et al., 2004). Shorter plant height was recorded from untreated inoculated control tomato plants. The stunting action of Meloidogyne spp. may result in the reduced plant height of the untreated plants. Jinfa et al., (2006) has also reported on the reduced plant height and shortness of plants infected with root-knot nematodes. Fruit number and weight of tomato Tomato yield per plot increased significantly in the treatments with the botanical extracts when compared to the untreated control. This result supports those reported by Wondimeneh et al., (2013) who demonstrated that the application of botanical extracts including neem and L. camara leaf extracts significantly improved the yield of tomato in the field with L. camara giving the best results. The difference in toxicity levels of the different extracts from the plants may be due to their variations in chemical compositions and also their differences in the concentrations of the toxic compounds. This may be the cause of the differences observed in performance on yield and nematode infestation. Maqbool and Firoza, (1996) reported that different plant leaves and seeds extracts act differently on Helicotylenchus spp. infested tomatoes. Effects of the extracts on nematode parameters (population of juveniles in the soil, egg mass index and galling index) The extracts significantly reduced the population of juveniles in the soil, the galling index and the egg mass index on roots when compared to the untreated plants. The nematicidal effect of the plant extracts could be the cause of the reduced number of motile larvae that penetrates the roots of the plants. This effect of the extracts is believed to act directly on infective J2 larvae. Agbenin et al., (2005) reported that A. sativum extract resulted in reduced soil larval populations and also a decrease in the egg mass and galling indices on plant root. The egg mass index of L. camara and A. sativum was not significantly different from that of chemical nematicide (Vydate). This shows that both were effective in the control of root-knot nematodes infecting tomatoes. This was also reported by Wafaa et al., (2013) who noted that castor seed and A. sativum aqueous extracts led to a reduction in the egg mass index and the galling index in the roots of tomato. The reduced population densities of the nematodes accompanied by increased growth of the tomatoes shows that the tested plant extracts have nematicidal potentials. The extracts may be toxic to the eggs or the juveniles resulting in reduction in nematode numbers in the soil. 392 Abdi, (1996) attributed the reduction in the galling index to poor penetration by the J2s and the retardation in their activities like feeding and reproduction. This could be further attributed to plant extracts. Nematode population at harvest The nematode count at the start of the field experiment was 260 juveniles /200 cc soil. However, the final population density in the soil at the time of harvest showed that the root-knot nematodes had significantly (p ≤ 0.05) reduced in the treatments with plant extracts over untreated control. From the results the number of J2s in the soil in L. camara and vydate treatments did not differ significantly. These results are similar to those of Wondimeneh et al., (2013) who reported that plant extracts from mexican marigold, Lantana and bitter leaf leaves at 5% concentration lowered the final nematode density significantly over the untreated control and was comparable with the synthetic nematicide. Farms with garlic or intercropping garlic with other crops have shown potential in decreasing the reniform nematode population and M. incognita. This is attributed to the allelopathic substances contained in garlic or that garlic did not offer the essential elements important in nematode development (Ameen, 1996). The extracts are believed to contain certain chemicals that possess ovicidal and larvicidal properties that inhibit the multiplication of the nematodes. This could be the cause of the reduction in the final population densities of nematodes in the soil (Agbenin et al., 2005). Qamar et al., (2005) reported that the aerial parts of L. camara contain chemical substances such as lantanone, oleanolic acid, lantanoside and camaric acid that posses nematicidal properties that work against M. incognita. Ahmad et al., (2010) has also reported that various concentrations of L. camara leaf extracts have deleterious effect that act against nematodes. CONCLUSIONS This study showed that among the different plant extracts tested L. camara leaf extract at 5% concentration worked well against the Meloidogyne spp. It was able to increase the tomato yield and showed greater potential for use in management of root-knot nematode in tomato plants. The results of the present study show that extracts from different plant species have varying degrees of antinematicidal effects. The presence of natural phytochemicals in different plants that have antinematicidal effects is a promising option for control of root-knot nematodes. ACKNOWLEDGEMENT I would like to express my sincere gratitude to Dr. Jonah Birgen and Dr. George Muhia Kariuki of Kenyatta University. I thank them for the material support and guidance they offered me. I also thank the laboratory staff of Kenyatta University for their assistance in apparatus and laboratory skills. 393 REFERENCES Ahmad, S., Akhter, M., Zia-Ul-Haq, M., Mehjabeen, A. S. (2010). Antifungal and nematicidal activity of selected legumes of Pakistan. Pakistan Journal of Botany Ameen, H. H. (1996). Influence of garlic Allium sativum on populations of Rotylenchulus reniformis and Meloidogyne incognita infecting cowpea and tomato. Al-Azhar J. Agric. Res. 23:77–85. Bridge, J. and Page, S. L. J. (1980). Estimation of root knot nematode infestation levels on roots using aerating chart. Tropical Pest Management 26:296-298. Conway, G. (1995). The depletion of natural resources: the impact of food. A 2020 vision for food, agriculture and the environment. Washington, DC: International Food Policy Research Insti-tute; p. 125. Coyne, D. L., Nicol, J. M., and Cladius-Cole, B. (2007). Practical Plant Nematology: A Field and Laboratory Guide. Cotonou, Benin, SP-IPM secretariat, International Institute of Tropical Agriculture (IITA), Pp 34-35. Dropkin, V. H. (1980). Introduction to plant Nematology, John Willy and sons (New York.). Food and Agricultural Organization (2006). FAO production year book, Basic Data Unit, Statistics division, FAO, Rome, Italy 55: 125-127. George, R.A.T. (1999). Vegetable seed production 2nd edition. Solanaceae, Lycopersicon,CAB International Wallingford UK (p 213-226). Green, C. D. (1971). Mating and host finding behavior of plant nematodes. In: Plant Parasitic Nematodes, Academic Press, New York, pp. 247-266. Holbrook, C. C., Knauft D. A. and Dickson, D. W. (1983). A technique for screening peanut for resistance to Meloidogyne arenaria. Plant Disease. 1983;67: 957–958. Jinfa, Z., Waddell, C., Sengupta, G. C., Potenza, C. and Cantrell, R. G. (2006). Crop Science. 46(46):1581-86. Koon, W., Krueger, R., Mcsorley, R. and Dover, K. E. (2010). Marigolds (Tagetes spp.) for nematode management. Research gate.net/publication/ 237446794. Nchore, S. B., Waceke, J. W., and Kariuki, G. M. (2011). Use of agro-industrial waste and organic amendments in managing root-knot nematodes in black nightshade in selected parts of Kenya. African Crop Science Conference Proceedings, 10: 223 – 229. Okendi, E. M., Kariuki, G. M., Marais, M. and Moleleki, L. N. (2014). The threat of root-knot nematodes (Meloidogyne spp.) in Africa. Journal of plant pathology.63(4):727-737. Sharma, S. B., Mohiuddin, M., Jain, K. C. and Remanandan, P. (1994). Reaction of pigeon pea cultivars and germ plasm accessions to the root-knot nematode, M. javanica. Journal of Nematology, 26: 644-652. Taye, W., Sakhuja, P. K. and Tefera, T. (2012). Evaluation of plant extracts on infestation of root-knot nematode on tomato (Lycopersicon esculentum mill). Journal of Agriculture. Res. Dev 2(3):86-91. Tiwari, A. K. and Mukhopadhyay, A. N. (2001). Testing of different formulations of Gliocladium virens against chickenpea wilt complex. Indian phytopathol. 54:67- 71. 394 Wondimeneh, T., Sakhuja, P. K. and Tadele, T. (2013). Evaluation of plant extracts on infestation of root-knot nematode on tomato (Lycopersicon esculentum Mill). Journal of Agricultural Research and Development Vol. 2(3). pp. 086-091. Yudelman, M., Ratta, A. and Nygaard, D. (1998). Pest management and food production: looking to the future, food, agriculture and the environment. Discussion Paper 25. International Food Policy Research Institute; Washington, DC; pg. 50. 9.2 Assessment of sweet melon pollinators grown under field conditions at KALRO-Kabete, Kenya. Ong’ute H W1, E D Kokwaro1 and M Kasina2* 1Kenyatta University, Department of Zoological Sciences, P.O. Box 43844-00800, GPO, Nairobi 2National Sericulture Research Center-KALRO P.O. Box 7816 code-01000 Thika *Correspondence: Muo.kasina@kalro.org ABSTRACT Sweet melon, (Cucumumis melo L.)is a succulent and extra-sweet fruit demanded by the high-end consumers in Kenya. The fruit is valued mainly because of its high concentration in vitamins and anti-oxidant properties. Sweet melons are produced in the warm areas under field and greenhouse conditions in Makueni and Machakos counties, Kenya. Without pollinator visits, the crops suffers in production. The study was motivated by the limited information about pollination provision in sweet melons grown in Kenya. The purpose of the study was to identify the pollinators of the sweet melons grown in the field. An experiment was set at KALRO-Kabete field plots and observations were made on the flower visitors. The flower handling time amongst three insect visitors, : honeybees, butterflies and hoverflies was recorded using a stopwatch from 8.00am to 4.00 pm. The results of the study showed that the honeybees recorded the highest mean number of visits between 0900hrs and 1600hrs. The honeybees also spent a significantly longer time foraging on the sweet melon flowers as compared to the butterflies and hoverflies, p <0.001. The findings suggest that honeybees contribute significantly to pollination for sweet melons. As such, farmers are encouraged to keep honey bees to assure fruit set and high quality fruits. Key words; Honeybees, pollinators, pollination, sweet melons INTRODUCTION Sweet melon, Cucumis melo is vegetable crop economic value that is cultivated in different parts of the world. Traditionally, Cucumis melo is perceived to be a dessert plant that originated from Africa and was first domesticated in Egypt as a source of food (Goldman, 2002). The crop belongs to the family Cucurbitaceae, and it is cultivated in the warm areas in the tropics. The fruits are a dietary choice for some people because of its Vitamin C, Vitamin E, carotenoids that are known natural antioxidants. Consumers are attracted to the sweetness of the fruits and also the health benefits that come with consumption. In the world, high production of sweet 395 melons is done in the United States, China, Turkey, Mexico and Brazil(Sebastian, Schaefer, Telford& Rennera, 2010).In Kenya production of sweet melons is slowly gaining popularity in the warm areas of Makueni,Machakos, and Garissa (Horticultural Crops Directorate, 2006).Production of sweet melons in the field is challenged by pests and diseases (Goldman, 2002). For example, the common pests that constrain production range from cutworms, whiteflies, Aphids, mites and fruit flies. Common diseases of the crop include powdery mildew, blight, and fruit rot. Faced with these challenges, farmers rely on the use of pesticides to control the pests and diseases (Ascher and Pickering 2011). Essentially, pollination determines fruit set in sweet melons and other cucurbits (Roubik, 1995;Abrol 2011).Researchers in the United States recently covered the impact of systemic pesticides such as neonicotinoids on honey bees. Systemic pesticides have been linked to colony collapse disorder (Ascher and Pickering 2011). Where pollination services are not provided farmers risk losing the entire crop because the unpollinated pistillate flowers end up aborting(Kasina et al., 2009). Information about the best pollinating agents for sweet melons in Kenya is still scant even though farmers cultivate the crop on. Because of pollination needs of sweet melons, the present study was conducted with the aim of assessing and identifying the most effective pollinators of sweet melons grown in the field. MATERIALS AND METHODS The study was conducted between September and December 2014 and from July to Dec 2014 on an experimental plot at KALRO-Kabete. The site is located Northwest of Nairobi Nairobi 36º 41’ E and 01º 15’S along Waiyaki Way 7 kilometers from the Central Business District.The experiment was laid out using a Randomized Block design. Planting of sweet melon galia variety seeds was done in a plot size of 10 x10 m with a spacing of 1x1 m. Three seeds were sown leading to the germination and development of 2 or 3 plants that were later thinned out to one. Insect visitors on the sweet melon flowers were observed during the day and were collected using a sweep net. Samples of the collected insect visitors were taken to the entomology laboratory for identification. Sampling of the sweet melon flowers was done randomly , and 1 plant was selected in every row of sweet melon plants. Observations of the foraging behavior of insect visitors commenced 2-3 days after the sweet melon flowering. The foraging insects were observed between 0900 and 1600. The number of flowers that were visited was recorded and the time spent on the flowers was taken using a stopwatch. Pollinator exclusion experiment was carried out using the bagging method in which some selected female flower birds were bagged in order to prevent visitation by bees and other pollinating agents. In addition to recording the number and durations of visits, direct observations of foraging behavior on sweet melon flowers was done in the experimental field. Pollen and nectar harvesting by the worker bees was registered based on the foraging activity of each visitor. On maturity the weight of the fruits, size and number of developed seeds were collected and recorded. Data was statistically analyzed using Genstat software version14. RESULTS AND DISCUSSION A total of 3 insect species were observed, collected and recorded visiting the sweet melon flowers between 09.00 and 1600. The honey bees recorded the highest mean 396 number of visits to the sweet melon flowers 24.03±5.8as shown in table 1. The butterflies and hoverflies had the lowest mean number of visits during the sweet melon bloom (0.83±0.0). The flower handling time recorded from the three insect visitors showed that honey bees had the highest mean flower handling time 10.06±0.3. The butterflies flowed with a mean flower handling time of 5.0±1.0. The hoverflies had the lowest mean flower handling time of 3.6±0.3 in the two seasons of investigations in the field. A significant difference was observed in the sweet melon flower handling time between the honey bees, butterflies and hoverflies P <.001 as shown in figure 1. Table i: The mean number of visits ± SEM by honey bees, hoverflies and butterflies on sweet melon staminate and pistillate flowers season 1 and season 2 at KALRO Kabete. Mean number of visits ( ± SEM) Honey bees Butterflies Hoverflies Season 1 17.875±2.3 0.5±0.1 0.25±0.1 Season 2 6.167±1.0 0.33±0.2 0.33±0.2 Totals 24.03±5.8 0.83±0.0 0.58±0.0 Table ii: The mean flower handling time ( ± SEM) by honeybees, hoverflies and butterflies on open staminate and pistillate flowers in season 1 and season 2 at KALRO Kabete. Mean flower handling time ( ± SEM) Honey bee Butterfly Hoverfly pvalue Season1 5.16±0.1 3.0±0.0 2.0±0.0 0.022 Season 2 4.9±0.2 2.0±1.0 1.6±0.3 <.001 Totals 10.06±0.3 5.0±1.0 3.6±0.3 In other past studies on sweet melons and other crops(Squash,Pumpkin,castor bean ), results have shown how the foraging behavior of honeybees, their duration of visits and pollen deposition correlate with their pollination success of different crops (Thomson, J. Goodel l,K 2006; Artz, Hsu,& Nault, 2011; Cane, Sampson& Miller, 2011; Rizzardo et al. 2012).The results showed that the honey bees recorded the highest number of visits on the sweet melon flowers. Their highest mean number of visits on the flowers during the bloom suggests that they have a high chance of pollinating the pistillate flowers.The foraging behavior of honeybees observed in the field was more beneficial than that of the butterflies and hoverflies. Worker bees were observed introducing their heads between the stigma or the anther. The honey bees took a longer time foraging on the flowers than the hoverflies and the butterflies. Being the commonest flower visitors on the sweet melons in the field shows the success of honey bees in pollinating the plants. Honeybees foraging behavior and harvesting of pollen from one flower to another makes them effective 397 in pollinating the female flowers. The extended visits by honey bees on the sweet melon flowers were due to the pollen reward they get from the flowers. The increased visitation duration by the honey bees increased their success in pollinating the sweet melons. Manetas and Petropoulou (2000), record that honey bees prolonged visit on flowers is because of availability of nectar, and this increases the probability of pollen being deposited on the female flowers and successful pollination. In a recent study in Brazil by Ribeiro et al. (2015), the value of honeybees on Cucumis melo was determined in an experiment that compared the number of honeybees visiting flowers in plots with different hives. Their findings suggest that the number of honeybee visits on the sweet melon flowers correlates with quantity and quality fruit production. For instance, where there was no hive or 1 hive the effect of honeybee visitation was not strong as compared in the plots where the hives were 2 or 3 thus having a stronger visitation effect. CONCLUSIONS AND RECOMMENDATIONS Sweet melons grown in the field benefit from pollination by honeybees that prefer their flowers. The high number of visits made on the sweet melon flowers increases the chances of pollination. The effectiveness of honeybees in pollinating sweet melons has been shown with fruit set, fruit size and weight. Honeybees are therefore an invaluable asset in sweet melons and other vegetable crops production. Further investigations are needed to determine the impact of other factors such as diseases on their foraging activity and effectiveness in pollination of crops. REFERENCES Abrol D. P. (2011). Pollination Biology: Biodiversity Conservation and Agricultural Production. Springer Science & Business Media. Artz, D. R., C. L. Hsu, and B. N. Nault. (2011). Influence of honey bee, Apis mellifera, hives and field size on foraging activity of native bee species in pumpkin felds. Cane, J.H., Sampson, B.J., & Miller, S.A. (2011). Pollination value of male bees: the specialist bee Peponapis pruinosa(Apidae) at summer squash (Cucurbita pepo). Environmental Entomology 40:614-620. Maynard,L. (2007). Cucurbit Crop Growth and Development. Retrieved from https://www.agry.purdue.edu/CCA/2007/2007/Proceedings/Liz%20Mayn ard-CCA%20proceedings%201_KLS.pdf Roubik,D.W (1995). Pollination of Cultivated Plants in the Tropics, Issue 118. Food & Agriculture Org. Ribeiro et al. (2015). Honey bees ( Apis mellifera ) visiting flowers of yellow melon ( Cucumis melo ) using different number of hives. Cienc. Rural vol.45 no.10 Santa Maria Oct. 2015 Rizzardo R.A.G. et al. (2012). Apis mellifera pollination improves agronomic productivity of anemophilous castor bean (Ricinus communis). Anais da Academia Brasileira de Ciências, v.84, p.605-608, 2012 398 Shawki MA, Titera D, Kazda J, Kohoutkova J and Taborsky V, (2006). Toxicity to honeybees of water guttation and dew collected from winter rape treated with Nurelle D. Plant Protection Science 42:9–14 Serrano, A. R.; Guerra-Sanz. J. M. (2006). Quality fruit improvement in sweet pepper culture by bumblebee pollination. Scientia Horticulturae, 110 (2): 160-166. Thomson, J. Goodel l,K. (2006). Influence of Bee Specis (Hymenoptera;Apiformes) with contrasting Behaviors on Pollen Movement in a Mustard Brassica rapa (Brassicaceae) and the Muskmelon Cucumis melo (Cucurbitaceae). 399 9.3 Assessment of genetic diversity of Kenyan Dolichos bean (Lablab purpureus) (L) using simple sequence repeat (SSR) markers *Grace N. Kamotho1, Miriam G. Kinyua2, Reuben M. Muasya3, Simon T. Gichuki4, Bramwel W. Wanjala4, Esther N. Kimani4 1Karatina University, School of Agriculture and Biotechnology, P.O. Box 1957- 10101, Karatina, Kenya E.Mail: gracekamotho@yahoo.com 2University of Eldoret, School of Agriculture and Biotechnology, P. O. BOX 1125, Eldoret, Kenya E.Mail:mgkinyua@africaonline.co.ke 3South Eastern Kenya University, P.O. Box, 170-90200, Kitui, Kenya. E.Mail:masambia@gmail.com 4Kenya Agriculture and Livestock Research Organization, Biotechnology Institute, P.O. Box 57811-00200, Nairobi, Kenya E.Mail:esther.kimani@gmail, bramwelwanjala@yahoo.com *Correspondence: gracekamotho@yahoo.com ABSTRACT Dolichos bean or Lablab is a multipurpose drought tolerant crop in the family Leguminosae. Its leaves and immature pods are used as vegetables, dry beans as grains and the whole plant is used as fodder. Lablab varietal development in Kenya has been inhibited by lack of information on its genetic diversity. The aim of this study was to assess the genetic diversity of the Kenyan Lablab using simple sequence repeat (SSR) markers with a view to designing a varietal improvement programme. Ten simple sequence repeat primers were assayed on ninety six (96) Lablab genotypes collected from various parts of Kenya. The genotypes were grouped into 15 populations based on place of collection. DNA isolation was done using a modification of the cetyltrimethylammonium bromide (CTAB) method and subjected to polymerase chain reaction (PCR). Expected heterozygosity ranged from 0.23 to 0.46 and on average was 0.38. The average polymorphic information content (PIC) was 0.63. Analysis of molecular variance (AMOVA) showed that 15% genetic variation was among populations and 85% was within populations. Highest Nei’s genetic distance was between Western and Mwingi populations while lowest genetic distance was between Embu and Meru populations. Cluster analysis separated the 15 populations into two main groups. Narrow genetic variation of genotypes may result during long cultivation history of species as an adaptation to local agro-climatic conditions. Farmers continuously select good seed for planting based on desirable agro-morphological traits such as yield, disease resistance, drought tolerance and earliness in maturity. Molecular markers are scattered throughout the genome and their association with various agronomic traits is influenced by selection pressure. This study indicates that the Kenyan Lablab has a narrow genetic base and highly related. The need to design breeding programs with the aim of broadening the genetic base is of paramount importance. Key words: Dolichos bean, genetic diversity 400 INTRODUCTION Dolichos bean or Lablab bean (Lablab pupureus (L.) Sweet) belongs to the family Leguminosae. Its multipurpose uses make it an important species globally (Schippers, 2000). It has primarily been an ornamental annual vine in the United States of America for generations (Stevens, 2012). In India and China, the complete plant is edible (Kala et al., 2010). Young leaves are eaten raw in salads and older leaves are cooked like spinach. Flowers are eaten raw or steamed. The large starchy root tubers can be boiled and baked. The immature and dried seeds can be boiled or fried and both the leaf and seed of Lablab are rich in proteins (Kala et al., 2010). Lablab is useful as a cover crop and forage (FAO, 2012). Its dense green cover during the dry season protects the soil and decreases erosion by wind or rain. As green manure, it provides organic matter and minerals. It also fixes nitrogen into the soil thereby improving crop yields. Due to its drought tolerance, Lablab grows in a diverse range of environmental conditions (Valenzuela and Smith, 2002). In Kenya, Lablab contributes highly to food security among the small scale farmers in rural areas where it is grown. Despite its importance, limited research has been conducted on the species as a food crop. The lack of information on genetic diversity of Lablab has immensely hindered its improvement in Kenya. Estimation of genetic diversity in a crop species is prerequisite for its improvement. The use of germplasm with distinct DNA profiles helps to generate breeding populations with broad genetic base (Singh et al., 2012). In this study SSR markers were used to assess the genetic diversity of Lablab grown in Kenya. MATERIALS AND METHODS Plant materials A total of 96 Lablab purpureus accessions were collected from the gene bank of Kenya, Rift Valley, Eastern, Coast, Western, Nyanza and Central provinces of Kenya. One (1) gramme of leaf tissue was harvested when seedlings developed the first two fully grown leaves. The leaf tissue was placed in eppendorf self-standing tubes each containing two ceramic beads. The tubes with beads and leaf tissue were placed in the geno-grinder machine (Benchtop homogenizer, Fast prep*-24) which was set to run for one minute at 4 up-down movements per second (4M/S). A modified of the cetyltrimethylammonium bromide (CTAB) method (Kimani et al., 2012) was carried out as described below. Nine hundred micro-liters (900µl) of extraction buffer (2% CTAB, 100mM Tris- Hydrochloric acid (HCl) pH 8.0,1.4M Sodiumchloride (NaCl), 50mM Ethylenediaminetetraacetic acid (EDTA), 2% Polyvinylpyrolidone (PVP) 10µ l of 2% β-mercaptoethanol) was added to the leaf tissue and ground to form a slurry. The slurry was incubated at 650C for 15 minutes in a water bath with constant shaking. It was then centrifuged at 13000rpm for five minutes. Six hundred microliters (600µl) of the supernatant were transferred into a fresh eppendorf tube and an equal volume of chloroform: isoamyl alcohol (24:1) was added. The eppendorf tubes were shaken well before separating the contents in a centrifuge at 13000rpm for 5 minutes. Five hundred microliters (500µl) of the aqueous phase was transferred into a fresh tube and an equal volume of chloroform: isoamyl alcohol (24:1) added. The tubes were shaken well and then centrifuged for 5 minutes at 13000rpm. Four hundred microliters (400µl) of the aqueous phase was transferred to a fresh tube and an equal volume of ice-cold 401 isopropanol added, and mixed by inverting several times to precipitate the DNA. The tubes were centrifuged at 13000rpm for 5 minutes. The supernatant was decanted, leaving the DNA pellet at the bottom of the tube. The pellets were washed using 500µ l of 70% ethanol, and spun for 1minute before they were air dried for 1hour. The dried pellets were re-suspended in 50µl of sterile distilled water. RNA was removed by adding two microliters (2µl) of pancreatic ribonuclease A (RNase A) (10mg/ml) and incubating the samples for 1 hour at 37oC. The samples were stored at minus 20oC. The quantity and quality of genomic DNA was examined by comparing the template DNA isolated from samples with a DNA ladder (gene ruler) of one kilo base (1 kb) in a 0.8% agarose gel using 1x TBE buffer and viewed in a gel box (G: Box, Syngene). The diluted DNA samples were subjected to polymerase chain reaction (PCR) in a thermocycler machine (Techne TC 412, Applied Biosystems Veriti systems) amplification using simple sequence repeat (SSR) markers specific for Lablab purpureus (Table I) RESULTS AND DISCUSSION Markers’ effectiveness in detecting allele availability and polymorphism In this study, a total of 43 alleles were detected and all were polymorphic (Table II). The number of alleles at a marker locus is related to the genetic diversity that can be revealed by a particular marker. The more alleles at a locus, the higher the degree of diversity that can be revealed and the more efficiently closely related genotypes can be distinguished (Nagy et al., 2012). SSR markers are locus-specific and generally amplify one locus (Gupta and Varshney, 2000). According to Shibairo et al., (2015), heterozygosity is considered low if it is less than 0.4, moderate (0.4 - 0.7) and high when greater than 0.7. The expected heterozygosity at each polymorphic locus ranged from 0.23 (LabT6) to 0.46 (LabT1) and on average was 0.38 (Table II) indicating a low heterozygosity or a narrow genetic base of the Lablab accessions studied. Polymorphic information content (PIC) provides an estimate of discriminatory power of a locus by taking into account not only the number of alleles expressed, but also the relative frequency of those alleles (Nagy et al., 2012). The highest PIC value of 0.67 was observed in SSR primers LabT3, LabT7 and LabT33 while the lowest PIC value of 0.58 was observed in primer LabT6 Table II). Polymorphic information content values range from 0 (which is an indicative of monomorphism) to 1.0 (very high discriminative power with many alleles in equal frequencies). The higher the PIC value, the more informative is the SSR marker (Nagy et al., 2012). Genetic distance between populations of Lablab bean Genetic distance is the difference between two entities that can be described by allelic variation or the extent of gene difference between populations or species (Nei, 1987). In this study, pairwise comparison of Nei’s unbiased genetic distance among the 15 populations ranged from a low of 0.092 between Embu and Meru populations to a high of 1.081 between Mwingi and Western populations (Table III). This implies that the Embu/Meru populations are closely related and this could be attributed to the nearness of these geographical regions hence there could be higher chances of local communities sharing the Lablab bean accessions as seed stock. The identified 402 genetically distinct populations, for instance Mwingi and Western, could be potentially important sources of germplasm for further improvement programme in the Lablab genotypes in Kenya. Hybridizing selected members from the two populations could probably result to genotypes with high heterosis. Genetic differentiation of populations The extent of genetic differentiation was estimated between and within populations using F statistics of Wright, (1951). Three F coefficients are generally used in genetic diversity studies. These are: (FIT) that estimates correlation of genes within individuals over all populations; (FST) that estimates correlation of genes of different individuals in the same population and, (FIS) that estimates correlation of genes within individuals within populations (Muhamad et al., 2003). (FST) equals zero when subpopulations are identical in allele frequencies and one when they are fixed for different alleles. (FST) is a measure of genetic differentiation over sub-populations and is always positive (Muhamad et al., 2003). In this study the FST was generally low ranging from 0.188 to 0.399 with a mean of 0.270 (Table IV), implying that Lablab sub-populations used in this study did not have much difference in allele frequencies. The NM is a measure of gene flow and ranged from 0.377 to 1.081 with a mean of 0.733 (Table IV). This indicates that there was high gene flow among the Lablab populations. This could have come about due to movement of seeds from one geographical region to another. This is also supported by the AMOVA (Table V) which denotes that most (85%) of the molecular variation in Lablab bean accessions was partitioned within populations, with lesser amounts (15%) partitioned among populations. A survey conducted in Lablab growing areas of Kenya in 2006 indicated that a substantial percentage (20.4%) of Lablab bean growers obtained seed from neighbours and markets (Kamotho et al., 2010; Kinyua and kiplagat, 2012). Seed exchange for planting is common especially in crops that do not have structured certified seed production systems. This movement of seeds across regions is likely to have an impact on the gene flow in Lablab bean. Additionally, although Lablab is predominantly a self-pollinating crop which shows little inbreeding depression, significant levels (6-10%) of natural cross pollination occurs (Kukade and Tidke, 2014) which could also have contributed to gene flow. Mean allelic analysis across the fifteen populations An evaluation of diversity based on expected heterozygosity demonstrated that the highest population diversity existed among the Lablab bean accessions from Nairobi followed by Genebank-Eastern, Meru, Makueni, Nyeri, Murang’a, Thika, Riftvalley, Embu, Lamu, Machakos, Genebank-Coast, Mwingi, Nakuru and finally Western in that order (Table VI). Among the Lablab bean accessions analyzed for genetic diversity, accessions collected from Nairobi and Eastern region of Kenya but conserved at the genebank of Kenya showed the highest Shannon diversity index of I = 0.96 and I = 0.90, respectively, while accessions collected from Western and Nakuru exhibited the lowest index of 0.464 and 0.576 respectively (Table VI). The analysis of allelic patterns across the 15 Lablab bean populations revealed that accessions from Nairobi had the largest number of different alleles (Na) of 2.9. Since Nairobi is the capital city of Kenya, the high gene diversity found there could be attributed to the 403 fact that it is a business hub region where traders of Lablab bean from all over the country converge. Principal Coordinate and Cluster Analysis The principal coordinate analysis was undertaken in order to confirm the clustering pattern obtained from unweighted pair group method with arithmetic averages (UPGMA) cluster analysis and exploit the resolving power of ordination. The 15 Lablab populations generally segregated with a high degree of overlap among them (Fig.1). This indicates similarity among the populations. A dendrogram constructed (Fig. 2) on the basis of a genetic distance matrix and by unweighted paired group method with arithmetic averages (UPGMA resolved the 15 Lablab bean populations into two distinct clusters. According to the resultant dendrogram, fourteen populations were clustered into one cluster (B) while population number 15 (Western) which comprised of accessions from Bungoma and Kisumu separated on its own cluster (A) (Fig. 2). Narrow genetic variation of genotypes may result during the long cultivation history of species as an adaptation to the local agro-climatic conditions (Seehalak et al., 2006). Farmers continuously select good seed for planting based on desirable agro- morphological traits such as yield, disease resistance, drought tolerance and earliness in maturity. Molecular markers are scattered throughout the genome and their association with various agronomic traits is influenced by the cultivator under selection pressure induced by domestication. Therefore, the need to design breeding programs with the aim of broadening the genetic base of Lablab bean is of paramount importance. This could be achieved by introgression of genotypes from the wild and also from different regions in Africa, Australia, and Asia. 404 TABLES AND FIGURES Table I: List of ten SSR Primer Pairs (specific for Lablab purpureus) Used in the Study Prim er name Primer Sequence Expect ed produ ct size (bp) Tmo C Repeat motif Forward 5’-3’ Reverse 5’-3’ Lab T1 ACCAGAATGGTTTCTCA AGTTCCT GGTGAACCTTCCTACACC ATGACT 273 56.1 (TA)7 Lab T2 GTGCGCGTCACTTATTA GTTCTTA CAATATCTTCACGTAACC ACGGTA 224 54.6 (TATAT C)7 Lab T3 CAGATCGATTGGTAGCT GGATTTC CCTCCTTACAGAAAGGGT AGCCTAGT 194 57.8 (TG)7 Lab T6 TCAATCGTTGTTGGAAG AGGGTAT GTCTCCTTCAACTGTGTC CACTGA 187 57.5 (TGG)6 Lab T7 CAGCAGTGTTGCCTCAT ACAGAAC TGTACTTAGCCAAGATCA GGCACA 123 57.5 (ATG)6 Lab T14 GGCATGGTGAAGATTGA AGAAGAG AGAAGCAGAGGACAGGT GAATTGT 255 57.8 (GA)8 Lab T24 GATCAGCTCCAGACTGC TGACG TAACCCTCCATTCATTGT CCATTC 202 58.5 (TC)7 Lab T25 GGGTTGAAGCTCACACA AATTCTT CCAATGATGGTTGTATGA GTAGCAC 126 57.4 (TGGT)5 Lab T28 CTTTCTCCATGCAGACC AAACTTC CCTGTAAATAACTGTCCT GGGAAGC 204 57.9 (ATG)6 Lab T33 CTAACCATGGCCTTGAG TGGTACT AATGAGTGAATGCAGCA GTAG 345 57.3 (CTTTTC )5 Table II: Characteristics of the 10 Lablab SSR Markers Indicating Major Allele Frequency, Number of Alleles, Expected Heterozygosity and Polymorphism Information content (PIC) Marker Major Allele Frquency Allele Number Expected Heterozygosity PIC LabT1 0.3177 4.0000 0.4583 0.6530 LabT2 0.5000 5.0000 0.2292 0.6400 LabT3 0.3958 5.0000 0.4688 0.6715 LabT6 0.5313 4.0000 0.2083 0.5763 LabT7 0.3594 5.0000 0.4271 0.6701 LabT14 0.3698 4.0000 0.3021 0.6416 LabT24 0.4427 4.0000 0.4688 0.6133 LabT25 0.4115 4.0000 0.4271 0.6371 LabT28 0.4115 4.0000 0.4896 0.5838 LabT33 0.3698 4.0000 0.3229 0.6735 Mean 0.4109 4.3000 0.3802 0.6360 405 Table III: Lablab Pairwise Population Matrix of Nei’s Genetic Distance Key: EMB: Embu, GENC: Genebank Coast, GENE: Genebank Eastern, MAC: Machakos, MAK: Makueni, MER: Meru, MUR: Murang’a, MWI: Mwingi, NAI: Nairobi, NAK: Nakuru, NYE: Nyeri, RIF: Rift Valley, TH: Thika, WEST: Western EMBU GE_C GE_E LAMU MAC MAK MERU MUR MWI NAI NAK NYER I RIFTV THIKA WEST 0.000 EMBU 0.357 0.000 GEN_C 0.348 0.414 0.000 GEN_E 0.437 0.498 0.293 0.000 LAMU 0.614 0.712 0.272 0.317 0.000 MACH 0.435 0.341 0.223 0.271 0.370 0.000 MAK 0.092 0.343 0.312 0.352 0.549 0.346 0.000 MERU 0.402 0.459 0.138 0.342 0.235 0.200 0.371 0.000 MUR 0.536 0.678 0.247 0.298 0.572 0.351 0.557 0.332 0.000 MWI 0.373 0.160 0.326 0.434 0.577 0.169 0.280 0.351 0.506 0.000 NAI 0.348 0.521 0.478 0.527 0.863 0.520 0.485 0.430 0.318 0.426 0.000 NAK 0.384 0.532 0.193 0.386 0.413 0.194 0.225 0.121 0.382 0.255 0.464 0.000 NYERI 0.275 0.422 0.344 0.229 0.656 0.395 0.227 0.513 0.234 0.294 0.339 0.360 0.000 RIFTV 0.649 0.577 0.250 0.382 0.357 0.446 0.638 0.297 0.654 0.468 0.682 0.390 0.676 0.000 THIKA 0.483 0.542 0.446 0.496 0.627 0.358 0.336 0.553 1.081 0.409 0.966 0.487 0.651 0.588 0.000 WEST 406 Table IV: F-Statistics and Estimates Of Differentiation of all Lablab Bean Populations for Each Locus Locus FIT FST FIS NM LabT1 0.244 0.188 0.069 1.081 LabT2 0.401 0.399 0.004 0.377 LabT3 0.019 0.248 -0.305 0.757 LabT6 0.296 0.381 -0.136 0.407 LabT7 0.106 0.204 -0.123 0.978 LabT14 0.332 0.274 0.079 0.661 LabT24 0.139 0.215 -0.096 0.914 LabT25 0.257 0.309 -0.074 0.560 LabT28 0.142 0.213 -0.090 0.926 LabT33 0.274 0.273 0.002 0.667 Mean 0.221 0.270 -0.067 0.733 SE 0.037 0.023 0.036 0.076 Table V: Analysis of Molecular Variance (AMOVA) for 15 Populations of Lablab Bean and Partitioning of the Total Diversity into Population Components Source of variation df TSS MSS Estimated Variance Percent molecular variance (%) P PhiPT (ΦPT Among Populations 14 258.445 18.460 1.567 15% 0.010 0.153 Within Populations 81 701.274 8.658 8.658 85% Total 95 959.719 10.225 100% PhiPT (ΦPT) = the estimate of population genetic differentiation based on permutation across the full data set, df = degree of freedom, TSS = total sum of squares, MSS = Mean sum of squares, Table VI: Mean Number of Different loci (Na), Number of Effective loci (Ne) Expected Heterozygosity (He), Shannnon index (I) Across the 15 Lablab Bean Populations. Mean EM GC GE LM MC MK ME MU MW NA NA NY RV TH WE Na 2.30 2.30 2.90 2.60 2.60 2.50 2.60 2.70 2.30 2.90 1.90 2.20 2.40 2.80 1.90 Ne 2.05 1.83 2.31 1.97 1.92 2.17 2.26 2.16 1.71 2.51 1.64 2.10 2.01 2.15 1.51 He 0.48 0.44 0.56 0.47 0.46 0.51 0.53 0.51 0.40 0.58 0.36 0.51 0.48 0.51 0.30 I 0.73 0.67 0.90 0.75 0.74 0.81 0.85 0.83 0.63 0.96 0.58 0.75 0.75 0.84 0.46 Key: EM: Embu, GC: Genebank Coast, GE: Genebank Eastern, MC: Machakos, MK: Makueni, ME: Meru, MU: Murang’a, MW: Mwingi, NA: Nairobi, NK: Nakuru, NY: Nyeri, RV: Rift Valley, TH: Thika, WE: Western 407 Fig.1: PCoA: Scatter plot showing the clustering pattern of 15 Lablab bean populations represented by different colours and symbols. KEY: The 15 Lablab populations are Embu, Genebank-Coast, Genebank-Eastern, Lamu, Machakos, Makueni, Meru, Murang’a, Mwingi. 0.20 0.15 0.10 0.05 0.00 Fig.2: Genetic relationship among 15 populations of Lablab purpureus using dendrogram based on Darwin’s genetic identity distance. C o o rd . 2 Coord. 1 Principal Coordinates EMBU GEN_COAST GEN_EASTERN LAMU MACHAKOS MAKUENI MERU MURANGA MWINGI NAIROBI NAKURU NYERI RIFTVALLEY THIKA WESTERN Embu Meru Gen_Coast Makueni Nairobi Gen_Eastern Muranga Nyeri Lamu Mwingi Rift Valley Nakuru Machakos Thika Western B2 A B A B 408 CONCLUSION Lablab germplasm assayed by SSR markers is of narrow genetic base. Lablab bean populations from Mwingi and Western were found to be distantly related and therefore selected genotypes of desirable agronomic traits from the two populations could be hybridized to produce genotypes with probably high heterosis. Results of this study are expected to benefit Lablab bean breeding efforts in Kenya as well as aid in conservation of Lablab germplasm. ACKNOWLEDGEMENT The authors express appreciation to Kirk House Trust, (UK), Kenya Agriculture and Livestock Research Organization (Biotechnology Institute-Nairobi) and Karatina University for supporting the field and laboratory activities of this study. REFERENCES FAO. (2012). Grassland species index. Lablabpurpureus. http://www.fao.Org/ag/AGP /AGPC /doc/Gbase/DAT A/Pf000047.HTM (Accessed on 10th May, 2015). Gupta, P.K. and Varshney, R.K. (2000). The development and use of microsatellite markers for genetic analysis and plant breeding with emphasis on bread wheat, Euphtyica 113: 163-185. Kala, K.B., Soris, T.P., Mohan V.R., and Vadivel, V. (2010). Nutrient and chemical evaluation of raw seeds of five varieties of lablab purpureus (L.) Sweet. Advances in Bioresearch, Vol.1 (1) June 2010:44-53. Kamotho, G.N., Kinyua, M.G., Muasya, R.M., Kimani, E.N., and Orwa, D.O. (2010). Abaseline survey on production, Utilization and Marketing Constraints of Lablab bean: Impact on Lablab Bean Improvement in Kenya, International Journal of Professional Pactice, Vol.1, pp. 21-29. Kimani, E.N., Wachira, F.N., and Kinyua, M.G. (2012). Molecular diversity of Kenyan lablab bean (Lablab purpureus(L.) Sweet) accessions using amplified fragment length polymorphism markers. Am J Plant Sci.3:313-321. Kinyua, M. G., and Kiplagat, O.L. (2012). Lablab (Lablab purpureus L. Sweet) bean Improvement using mutation and biotechnological techniques. Dansten Agencies, Nairobi, Kenya. Kukade S.A. and Tidke J A., (2014). Reproductive biology of Lablab lablab L. (Fabaceae). Indian Journal of plant Sciences, Vol.3 (2) April-June, pp 22-25. Muhammad, A., Bakhsh, A., Haqqani, A.M., and Bashir, M. (2003). Genotype- environment interaction for grain yield in chickpea (Cicer arietinum L.). Pakistan J Botn 35(2):181-186. Nagy, S., Poczai, P., Cernak, I., Gorgi, A., Hegedus, G. and Taller, J. (2012). PICcalc: An online progrmme to calculate polymorphic information content for molecular genetic studies.Bioche. Genet.50: 670-672. DOI 10.1007/s10528- 012-9509-1. Nei M., (1987). Molecular Evolutionary Genetics. Columbia University Press, New York. Schippers, R.R. (2000). African indigenous vegetables: An overview of the cultivated species, p 95. Chatham, UK: Central Avenue, Chatham Maritime. 409 Seehalak, W.P.,Sommanas, W., and Srinives, P. (2009). Microsatellites markers for mungbean developed from sequence database, Molecular Ecology Resources 9: 862 – 864. Shibairo, S., Mutitu, E., Obukosia, S., Ngugi, E., Magomere, T. (2015). The population structure of wild sorghum species in agro-ecological zones of Western Kenya. Erepository.uonbi.ac.ke/handle/11295/86434. (Accessed`on 24th June, 2015). Singh, S.K., Lavanyan, R.G., Bhat, K.V., Babu, G.S., Arya, L., Verma, M., Hussain, Z., Roy, S., Rathi, R.S., and Misra, A.K. (2012). Microsatellite markers revealed genetic diversity in mungbean mutant lines. Indian Journal of hill Farming, 25(1): 38-43. Stevens, J.M. (2012). Bean, hyacinth - Dolichos lablab (L). or Lablab purpureus (L.) Sweet. Institute of Food and Agricultural Sciences (IFAS) Univ. of Florida Extension. http://edis.ifas.ufl.edu/mu019. (Accessed on 10th August, 2015). Valenzuela, H., and Smith J. (2002). Sustainable agriculture green manure crops. SA- GM-7. Cooperative Extension Service, College of Tropical Agric. And Human Resources,Univ. of Hawaii at Manoa. http://www. ctahr. hawaii. edu/ oc/ freepubs/ pdf/ Green Manure Crops /lablab.pdf (accessed 14 July, 2015). Wright, S. (1951).The genetical structure of populations. Ann. Eu- gen.15:323–354. 9.4 Response of yield and yield components of maize as affected by tied ridges, fertilizers and cropping system in the semi- arid areas of Machakos County, Kenya Mwende, N.1, 1*, Danga, B. 0 1, Mugwe J1 and Kwena Kizito2* 1Department of Agricultural Resource Management, Kenyatta University, P.O. Box 43844- 00100, Nairobi Kenya. 2 Kenya Agricultural and Livestock Research Organization, P.O. Box 14733-00800, Nairobi *Correspondence: ngie.mwende@ku.ac.ke, 0711624969 ABSTRACT Lack of enough soil moisture and depletion of nutrients in the soil as a result of continuous cropping and cultivation of soils without adequate use of external inputs is one of the major crop production limitations in the semi-arid areas of Kenya. An experiment was set up in Machakos County Kenya, in 2015 and 2016 to investigate the response of yield and yield components of maize to tied ridges, fertilizers and cropping systems. On average, tied ridges combined with 5t/ha of FYM and monocropping gave the highest grain yield (1.11t/ha) followed by conventional combined with FYM 5t/ha and monocropping (0.89t/ha); while the least was from conventional combined with FYM 0t/ha and intercropping (0.26t/ha). For the biomass fresh weight, the highest was given by tied ridges combined with FYM 5t/ha and monocropping (3.87t/ha); followed by tied ridges with FYM 5t/ha and intercropping (2.65t/ha) and the least was from conventional tillage with FYM 0t/ha and intercropping (1.40t/ha). The ears dry weight ranged from 1.36t/ha to 0.34t/ha with the highest from tied ridges combined with FYM 5t/ha and 410 monocropping (1.36t/ha).The least was 0.34t/ha which was produced from conventional tillage, FYM 0t/ha and intercropping. The highest Stover dry weight was given by tied ridges, FYM 5t/ha and monocropping (0.93t/ha) while the least was from conventional, FYM 0t/ha and intercropping (0.44t/ha).Generally, the effect of tied ridges, fertilizers and cropping system had a highly significant effect (p value = <0.001) on grain yield, biomass fresh weight, ears dry weight, stover and grain weight. Key words: Tied ridges, fertilizers, cropping system, Growth, yield and yield components INTRODUCTION Food shortage as a result of declining yields and inadequate extreme fluctuations in the availability of water for plant growth in Sub- Saharan Africa is a common phenomenon among the smallholder farmers as reported by Baron et al.,(2005). According to Powlson et al., (2011), soil erosion is the major contributor to nutrient loss mostly where agronomic inputs are low, vegetation cover is scarce, soils are not resilient and also in places where heavy rain fall sometimes occur.Stoite et al., (2009) indicate that, soil erosion directly has an adverse impact on the productivity of the land as a result of loss of soil, water and nutrients. Reports by Shisanga et al., (2009) show that, soil impoverishment in Machakos County is also associated with poor farming practices and high costs of inorganic fertilizers. In addition, nutrient loss and water deficit mostly at the root zone is an important key factor that affects crop production (Bossio et al., 2010). According to Ngigi et al., (2006), increased soil moisture storage at the root zone (situ rain water conservation) reduces run- off and soil loss. According to literature, there is an important relationship between soil and water conservation and therefore it’s important to address all of the concurrently. Soils in Machakos County are characterized by low water holding capacity and low organic matter. The average maize yield is less than 0.5t/ha (only 1/3 of the potential) as reported by NEMA, (2013). Therefore, combining tied ridges, fertilizers and cropping system may be a solution to effective sustainable crop production. This study was carried out to determine the effect of tied ridges, fertilizers and cropping system on the yield and yield components of maize in Machakos County. MATERIALS AND METHODS An experiment was conducted in KARLO Katumani in Machakos County, Kenya in short rainy season 2015 and long rainy season 2016. The treatments were: water conservation at two levels (Conventional and tied ridges), fertilizers at three levels (0t/ha FYM, 5t/ha FYM, and 20kgN/ha) and cropping system at two levels (monocropping and intercropping). The experiment was laid out in a randomized complete block design (RCBD) with plot sizes of 5.4m x 3.6m replicated four times. The tied ridges were imposed during land preparation. The spacing of the ridges was 90cm tied at 2.5m interval. The ridges were 20cm high and ties (cross ties) 20 cm high. Maize variety planted was KVD1 (a short duration seed with maturity between 85 to 100 days recommended for dry areas) and cow pea var. K.80. The maize seeds were sown 3 seeds per hole then thinned to 2 seeds at a spacing of 90cm x 30cm. The cowpea was sown 2 seeds per hill at 30cm between maize rows for the 411 intercrop plots.Intial soil characterization before the start of the experiment was done and also the FYM used was analyzed. Triple super phosphate (TSP) at a rate of 60kg/ha was put in the planting holes in all the plots. Where FYM was a treatment, it was mixed with TSP. Calcium ammonium nitrate (CAN) was applied as a top dress at a rate of 20kg/ has in plots where it was a treatment. Other agronomic procedures in maize production were appropriately followed after planting. At maturity, maize was harvested from a net plot measuring 3.4m x 1.6m.The fresh plant biomass weight, ears fresh weight and stover fresh weight was taken using an electronic weighing balance; after which the ears and stovers were sampled, sundried and their dry weight taken and expressed on a 12.5% water content basis. DATA ANALYSIS The collected data was subjected to analysis of variance using Gens tart 15th Edition and the means were separated using Tukey’s HSD at p < 0.05. RESULTS For the plant biomass fresh weight, the means ranged between 3.86t/ha to 1.16t/ha with tied ridges combined with 5t/ha FYM and monocropping producing the highest mean (3.86t/ha). This was followed by tied ridges with 5t/ha FYM and intercropping (2.65t/ha). Conventional tillage with 0t/ha FYM and intercropping recorded the lowest mean (1.16t/ha). In terms of the total number of ears /ha, the means ranged from 17002 ears/ ha to 11028 ears /ha with conventional tillage combined with 5t/ha FYM and monocropping registering the highest mean (17002 ears /ha). The second highest mean was recorded from tied ridges with 5t/ha FYM and monocropping (1654 ears/ha). The lowest mean was produced by conventional tillage with 5t/ha FYM and intercropping (11028ears/ha). The means for the ears weight ranged between 1.36t/ha to 0.34t/ha with tied ridges with 5t/ha FYM and monocropping recording significantly higher yields (1.36t/ha) followed by conventional tillage combined with 5t/ha FYM and monocropping while the lowest mean was registered by conventional tillage with 0t/ha FYM and intercropping. For the stover yield, the means ranged between 0.93t/ha to 0.44t/ha with tied ridges combined with 5t/ha FYM and monocropping recording significantly higher yields (0.93t/ha) and the second highest from conventional with 5t/ha FYM and monocropping. The lowest mean was gotten from tied ridges with 20kg N/ ha (0.44t/ha). For the grain yield, the means ranged between 1.11t/ha to 0.26t/ha with tied ridges with 5t/ha FYM recording the highest mean (1.11t/ha). Conventional tillage combined with 5t/ha FYM and monocropping produced the second highest yield (0.89t/ha) and the lowest grain yield was given by conventional with 0t/ha FYM and monocropping (0.26t/ha). 412 Table 1: Means of Effects of Tied Ridges, Fertilizers and Cropping Systems on the Grain and Yield Components of Maize Treatment Biomass Fresh Weight (t/ha) Total No. of Ears/ ha Ears Weight(t/ha) Stover Yield (t/ha) Grain Yield (t/ha) Conventional + 0t/ ha FYM + Monocropping 2.142abcdef 16542a 0.9630abc 0.6073ab 0.6426ab Conventional + 0t/ha FYM + Intercropping 1.160a 13326a 0.3395a 0.4669a 0.2592a Conventional +5t/ha FYM + Monocropping 2.585abcdef 17002a 1.1033abc 0.8211ab 0.8897ab Conventional + 5t/ha FYM + Intercropping 1.404ab 11028a 0.4090ab 0.4365a 0.3264a Conventional + 20kg N/ha + Monocropping 2.369abcdef 15623a 1.0316abc 0.6760ab 0.8067ab Conventional + 20 kg N/ha + Intercropping 1.342ab 12406a 0.3965ab 0.4714a 0.3079a Tied ridges + 0t/ha FYM + Monocropping 1.889abcde 14244a 0.6609abc 0.5140ab 0.5240ab Tied ridges + 0t/ha FYM + Intercropping 1.438ab 11947a 0.4246ab 0.4658a 0.3358a Tied ridges + 5t/ha FYM + Monocropping 3.869f 16542a 1.3565c 0.9271ab 1.1106b Tied ridges + 5t/ha FYM + Intercropping 2.647abcdef 14244a 1.00001abc 0.5318ab 0.8070ab 413 Tied ridges + 20kg N/ha + Monocropping 1.955abcdef 11947a 0.7747abc 0.5207ab 0.6244ab Tied ridges + 20kg N/ha + Intercropping 1.579abcd 142449a 0.6038abc 0.4367a 0.4774ab P value < 0.001 0.184 < 0.001 < 0.001 < 0.001 e.s.e 0.6816 0.1193 0.121 0.0802 0.1012 *Means with the same letter in each column are not significantly different 414 DISCUSSION Tied ridges combined with 5t/ha FYM gave significantly higher yields across all the treatments except for the total number of ears /ha. The increased yields with tied ridges may be attributed to reduced run-off since the rain water was trapped in the ridges and given time to slowly infiltrate in to the soil. As a result, nutrient loss was also reduced since run- off was minimized. The increased water infiltration increased moisture at the root zone; improving nutrient uptake hence higher yields. This is as opposed to conventional tillage where run-off took place carrying away soil nutrients reducing the amount of nutrients in the soil. The findings agree with those of Prinz, (2001) who also reported increased yields with use of tied ridges in maize production. The plots with FYM registered higher yields as a result of increased source of organic matter in the soil which also improved the soil nutrients. The organic matter bound the soil particles together improving the water holding capacity of the soil and also the structure. This resulted in increased soil moisture for crop use hence higher yields. In plots with no FYM, the organic matter content remained low resulting to less soil nutrient hence reduced yields. The observation agrees with the findings of other researchers (Swift, 1997; Mutegi, 2012; Das et al., 1992; Tamayo et al., 1997 & Jate, 2012) who also observed increased yields as a result of FYM application. Mochoge et al., 1997 also reported increased maize yields of 700 and 1720kg/ha from plots treated with FYM. Monocropping recorded higher yields since there was reduced competition for growth factors and therefore, the soil nutrients were adequate for the crop to perform better as opposed to where there was intercropping which led to completion of growth factors by the crops reducing the yields. The findings are in line with those of Hadter et al., (1991) who reported that, sole cropping produced higher yields than mixed cropping. Sawagogo kabore’ et al., (2008) also observed that, monocropping had a significant yield increase. Other related results were reported by Twomlow et al., (2010) who reported an increased yield grains in maize monocropping. Conclusion and Recommendations Tied ridges combined with 5t/ha FYM had a highly significant effect (0.001) on the grain yield and its components apart from the total number of ears/ha. The combination also gave the highest stover and grain yield as compared to the other treatment. In order to exploit the potential of tied ridges as a water conservation practice, it ought to be combined with other soil and water management practices like use of FYM and monocropping system so as to maximize the yields. Farmers may adopt use of tied ridges combined with 5t/ha FYM and monocropping so as to maximize maize yields in consideration to other soil and water management practices. Further research can be done to determine whether maize intercropped with a different legume apart from cow pea which was used in this study would give different results and also to explore other soil and water management practices suitable in this experimental site to exploit the effectiveness of tied ridges in conserving soil and water. 415 ACKNOWLEDGEMENT We acknowledge KARLO Katumani for facilitating the research work and also Kenyatta University for granting the opportunity for study to the first author. REFERENCES Baron and Okwatch,(2005). Run off harvesting for dry spell mitigation in maize results on-farm research in semi- arid Kenya, Agricultural management, 74: 1-21 Bossio, D., Geheb, K., and Critchley, W.(2010). Managing water by managing land: Addressing land degradation to improve water productivity and rural livelihoods. Agricultural Water Management, 97, 536-542. Jate, M. (2012).Impact of mineral fertilizer integration with farm yard manure on crop yield, nutrient use efficiency and soil fertility in a long- term trial Crop production Technologies. Dr. Peeyush Sharma (Ed). ISBN: 978-953-307-787-1 Mochoge, B.O, Kanyanjua and S. Obanyi (1997).Response of maize, maize-cowpea intercrop and pigeon pea to N and P fertilizer and farm yard manure. African Crop Science Conference Proceedings vol.3 pp 439-448. Uganda 108: 19-24 (Field Crops Abst., 51: 3970; 1997) NEMA, (2013) Machakos District Environmental Action Plan 2013, Government of Kenya. Ngigi, S., Rockström, J., and Savenije, H.H.G. (2006).Assessment of rainwater retention in agricultural land and crop yield increase due to conservation tillage in Ewaso Ng‟iro river basin, Kenya. Physics and Chemistry of the Earth, 31, 910-918. Powlson, D. S., Gregory, P. J., Whalley, W. R., Quinton, J. N., Hopkins, D. W.,Whitmore, A. P., and Hirsch, P. R. (2011).Soil management in relation to sustainable agriculture and ecosystem services q. Food Policy, 36, 72-87. Shisanga, C., Mucheru, M., Mugendi, D., and Kung‟u, J. (2009).Effect of organic and inorganic nutrient sources on soil mineral nitrogen and maize yields in central highlands of Kenya. Soil and Tillage Research, 103, 239-246 Stolte, J., Wang. E, van Oosterom EJ, Meinke H, Asseng S, Robertson MJ, Huth N, Shi, X., and Ritsema, C. J. (2009). Introduction: Soil erosion and nutrient losses in the Hilly Purple Soil area in China. Soil and Tillage Research, 105(2), 283-284. Swift, M.Biological management of soil fertility: an integrated approach to soil nutrient repleshment. Proceedings of International Seminar on Approaches to Replenishing Soil Fertility in Africa-NGO Perspectives Tamayo, V.A.AR Munnoz and A.C Diaz, (1997). Organic fertilizers to maize; on alluvial soils in moderate climate. Act alidades Corpoica . 416 CONFERENCE ORGANIZING COMMITTEE OVERALL COORDINATOR 1. Prof Waceke Wanjohi - Dean, School of Agriculture and Enterprised Development MAIN COMMITTEE MEMBERS 1. Dr. Daniel Kyalo Willy - Chairman 2. Dr. Purity Nguhiu - Vice Chairman 3. Ms. Everline Samita - Secretary 4. Mr. Lawrence Kiguro - Member 5. Ms. Joan Sang - Member 6. Dr. Nicholas Korir - Member 7. Dr. Catherine Muui - Member 8. Dr. Eric Bett - Member 9. Dr. Ibrahim Macharia - Member 10. Dr. Ruth Gathu - Member 11. Dr. Gabriel Mwenjeri - Member 12. Dr. Wilson Thagana - Member 13. Dr. Maina Mwangi - Member SUB-COMMITTEE MEMBERS Liaison, Resource Mobilization and Logistics Budget and Finances Dr. Daniel Kyalo Willy - Chairperson Dr. Maina Mwangi - Chairperson Prof. Benson Mochoge Dr. Nigat Bekele Dr. Eric Bett Dr. Keneddy Mwetu Dr. Maina Mwangi Dr. Isaac Osuga Dr. Joseph Gweyi Ms. Everlyne Samita Dr. Benjamin Ndanga Dr. Purity Nguhiu Dr. Mukiri Githendu Venue/Halls Branding Dr. Richard Osiyo Dr. Ruth Gathu - Chairperson Prof. Fuchaka Waswa Dr. Nicholas Korir Dr. Duncun Kirubi Dr. Nigat Bekele Registration and Protocol Publicity and Website Development Dr. Catherine Muui - Chairperson Dr. Eric Bett - Chairperson Ms. Everlyne Samita Dr. Kenneddy Mwetu Ms. Mwende Ngie Dr. Nicholas Korir Ms. Sylvia Henga Dr. Ibrahim Macharia Ms Wanjiku Mbugua Ms. Rachael Ng’ang’a Rachael Wanjiru 417 Registration and Protocol Overall Conference Secretariat Dr. Catherine Muui - Chairperson Dr. Jane Mwangi - Chairperson Ms. Everlyne Samita Mr. Cyrus Gichaga Ms. Mwende Ngie Ms. Wanjiku Mbugua Ms. Sylvia Henga Ms. Karen Gacheri Ms Wanjiku Mbugua Ms. Lucy Wangu Ms. Rachael Ng’ang’a Mr. Kennedy Kiprotich Rachael Wanjiru Ms. Nancy Ndichu Abstracts Review, Book of Abstracts Development, Full Papers Review and Publications Security and Health Care/First Aid Dr. Purity Nguhiu - Chairperson Dr. Gabriel Mwenjeri - Chairperson Prof. Fuchaka Waswa Mr. Leornard Mburu Dr, Jayne Mugwe Ms. Sylvia Henga Prof. Bernard Njehia Dr. Keneddy Mwetu Exhibitions Dr. Wilson Thagana Dr. Wilsom Thagana - Chairperson Dr. Daniel Kyalo Dr. Nicholas Korir Dr. Ruth Gathu Dr. Maina Mwangi Transport, Accommodation and Catering Sub-Committee Dr. Duncan Kirubi Dr. Ibrahim Macharia - Chairperson Dr. Nicholas Korir Dr. Lucy Ngare Dr. Catherine Muui Dr. Leornard Mburu Ms. Rachel Kimani Audio Visual Equipment, Power and IT support Mr. Ezekiel Chepkwony Prof Fuchaka Waswa - Chairperson Dr. Duncan Kirubi Mr. Ezekiel Chepkwony 418 LIST OF PARTICIPANTS S/No. Name Email contact Institutional Affiliation 1 Acheing, Carolyne cachienga@gmail.com Kenyatta University 2 Achienga, Mildred Ada millieada@gmail.com Kenyatta University 3 Ajaku, Daniel ICRISAT-Nairobi@cgiar.org ICRISAT 4 Bekele, Nigat bekele.nigat@ku.ac.ke Kenyatta University 5 Bett, Erick bett.erick@ku.ac.ke Kenyatta University 6 Chepkwony, Ezekiel ezekielchepkwony66@gmail.com Kenyatta University 7 Chimalizeni, Michael m.chimalizeni@cabi.org CABI 8 Danga, Benjamin benjamindanga@yahoo.com Kenyatta University 9 Gakunyi, Lydia Mugao mugaolydia@gmail.com Kenyatta University 10 Gatambia, Eliud eliudgatambia@yahoo.com KALRO-THIKA 11 Gathu, Ruth gathu.ruth@ku.ac.ke Kenyatta University 12 Gathua, Fidelis Wahira fidelisgithua@gmail.com Kenyatta University 13 Gichaga, Cyrus gichaga.cyrus@gmail.com Kenyatta University 14 Githendu, Mukiri mukirigithendu@yahoo.com Kenyatta University 15 Githinji, Joseph .N. Joseph_ndegwa@wvi.org World Vision Kenya 16 Gitonga, Hannah W. gitongaw@yahoo.com KALRO-THIKA 17 Henga, Sylvia ssilviahenga@gmail.com Kenyatta University 18 Juma, Peter Shango pshango@kephis.org Kenyatta University 19 Kabuage, Lucy lwkabuage@gmail.comKabuage Kenyatta University 20 Kagunyu, Anastasia Anastasia.Kagunyu@kalro.org KALRO-THIKA 21 Kamotho, Grace gracekamotho@yahoo.com Karatina University 22 Karanja, Ann Njeri lanonjeri@gmail.com Kenyatta University 23 Kariuki, George gmmkariuki@yahoo.com Kenyatta University ( 24 Kariuki, Duncan Ducco.kariuki@gmail.com Kenyatta University 25 Karuga, Charlse info@equitybank.co.ke/alice.mburu@equitybank.co.ke Equity Bank -KU 26 Kasina, Muo Muo.Kasina@kalro.org KALRO-THIKA 27 Kiguro, Lawremce lawrence_Kiguro@wvi.org World Vision Kenya 28 Kimani, Racheal Kimani.rachael@ku.ac.ke Kenyatta University 29 Kimonyi, Patrick marketing@oshochem.org Osho chemicals 30 Kimotho, Veronica hshiraku@biovisionafrica.org BioVision 31 Kiprotich, Kennedy leilelken@gmail.com Kenyatta University 32 Kirubi, Duncan kirubi.duncan@ku.ac.ke Kenyatta University 33 Kitonga, Lydia lydiakitonga@gmail.com Kisii/ Egerton University 419 34 Kiviva, Gabriel Kenyatta University 35 Kombo, Ceasar cmkombo@gmail.com KALRO-THIKA 36 Korir, Nicholas Korir.nicholas@ku.ac.ke Kenyatta University 37 Lagat,Chemutai R. roselagat76@gmail.com Kenyatta University 38 Leruk, Fatuma KLMC 39 Macharia, Ibrahim mach4ibra@yahoo.com Kenyatta University 40 Magero, Tabitha tmagero04@gmail.com Kenyatta University 41 Maina, Joyce Kenyatta University 42 Maindi, Catherine N. catherinemaindi@yahoo.com Kenyatta University 43 Maranga, M. Lewish Kenyatta University 44 Matere, Nancy J jilandere2011@gmail.com Kenyatta University 45 Mawanga, William wv_kenya@wvi.org World Vision Kenya 46 Mbae, Jane j.jane10@student.ku.ac.ke Kenyatta University 47 Mbalanya, Mavuze M. maxwellmavuze@gmail.com Kenyatta University 48 Mbatha, Peter mbathapeterm@yahoo.com Directorate of Veterinary Services(MoALF) 49 Mbaya, Miriam mirium.mbaya@ku.ac.ke Kenyatta University 50 Mbugua, Wanjiku Wanjiku.mbugua@ku.ac.ke Kenyatta University 51 Mburu, Leonard leonardmburu@yahoo.com Kenyatta University 52 Mburu, Njung’e mburu@convifirm.com Kenyatta University 53 Mochoge, Benson Bmochoge1974@hotmail.com Kenyatta University 54 Morgan, Lydia wv_kenya@wvi.org World Vision -K 55 Mugwe, Jayne jaynemugwe@yahoo.com Kenyatta University 56 Mulwa, Joseph josemulwa0009@gmail.com KALRO-THIKA 57 Mumo, Mathew mattmumo@yahoo.com Kenyatta University 58 Munga, Leornard mungalk@yahoo.com Kenyatta University 59 Murage, Francis M mucirafrancis@yahoo.com Embu University 60 Muriithi, Alfred murealphy@yahoo.com Kenyatta University 61 Muriithi, Lucy wv_kenya@wvi.org World Animal Protection 62 Muriu, Felister W. fmuriu@seku.ac.ke South Eastern Kenya University 63 Mutua, Roger ruiru@kcb.co.ke KCB Bank- Ruiru 64 Muui, Catherine Muui.catherine@ku.ac.ke Kenyatta University 65 Mwangi, Elizabeth elizamwangi@gmail.com Kenyatta University 66 Mwangi, James nguhiuja@yahoo.com University of Nairobi 67 Mwangi, Jane drmwangi12.jm@gmail.com Kenyatta University 68 Mwangi, Magdalene M mmutumi@gmail.com University of Nairobi 420 69 Mwari, Mary Mary.Mwari@kalro.org KALRO-THIKA 70 Mwaura, George G. gg.mwaura1@gmail.com Kenyatta University 71 Mwenesi, Karika Sheila wv_kenya@wvi.org World Vision Kenya 72 Mwenjeri, Gabriel gmwenjeri@gmail.com Kenyatta University 73 Marystella wekhanya nangoniswek@gmail.com Kenyatta University 74 Mwungu, Chris Miyinzi miyinzic@gmail.com ICIPE 75 Namibiru-Mwaura, Evelyne e.nmwaura@aasciences.ac.ke Africa Academy of Sciences 76 Nang'oni, Marystella W nangoniswek@gmail.com Kenyatta University 77 Ndenga, Charlse charlesndenga@yahoo.com Kenyatta University 78 Ndichu, Nancy shikundiwa@gmail.com Kenyatta University 79 Ngare, Lucy Ngare.lucy@ku.ac.ke Kenyatta University 80 Ngie, Mwende mwendengie@gmail.com Kenyatta University 81 Ngigi, Benson info@agrochem.co.ke Agrochemicals Association of Kenya 82 Nguhiu, Purity N. puritynguhiu@yahoo.com Kenyatta University 83 Njau, Njeri njaunjeri@gmail.com EDULINK 84 Njehia, Bernard Njehia.bernard@ku.ac.ke Kenyatta University 85 Njeru, Charity N. charity.njeru@yahoo.com Kenyatta University 86 Njoki, Felister felister.njoki@ku.ac.ke Kenyatta University 87 Nyaga, Jeremiah wv_kenya@wvi.org World Vision Kenya 88 Nyairo, Newton Nyairo.newton@ku.ac.ke Kenyatta University 89 Nyakundi, Moffat O mophatnyakundi@gmail.com Kenyatta University 90 Nyambura, Grace M rycmwangi@gmail.com ICIPE/Kenyatta University 91 Nzamu, Janet janetnzamu123@gmail.com Kenyatta University 92 Odindo, Samuel Jerry kabokasam@gmail.com MEFOSEP-KISUMU 93 Oduor, Richard oduor.richard@ku.ac.ke Kenyatta University 94 Ogega, Stephen stepheneogega@students.ku.ac.ke Kenyatta University 95 Ogolla, Andrew Egesa ayechoogola@gmail.com Kenyatta University 96 Olajide, Rasak r.olajide@gmail.com University of Ibadan, Nigeria 97 Ombuna, Christine christineombuna@gmail.com KIRDI 98 Omondi, Bethwel kabokasam@gmail.com MEFOSEPT-Kisumu 99 Omondi, Raphael raphaelomondi@worldanimalprotection.org World Animal Protection 100 Ong’ute, Hillary hillarywills00@gmail.com Kenyatta University 101 Onkendi, Edward eddyoara1959@gmail.com University of Pretoria 421 102 Osiemo, Jamleck osiemojamleck@gmail.com University of Nairobi 103 Osiyo, Richard rocoruko@ymail.com Kenyatta University 104 Osuga, Isaac Isaac.osuga@yahoo.com Kenyatta University 105 Rabach, Vincent Otieno rabachv@gmail.com Kenyatta University 106 Rose, Caroline World Animal Protection 107 Rutto, David Kiprono ruttodkl@gmail.com University of Eldoret 108 Rwito, Doris Karamana.doris@ku.ac.ke Kenyatta University 109 Samita, Evelyne everlynekakai@yahoo.com Kenyatta University 110 Sang, Joan joan_sang@wvi.org World Vision Kenya 111 Thagana, Wilson thagana.wilson@ku.ac.ke Kenyatta University 112 Theuri, Mathew Theuri.mathew@ku.ac.ke Kenyatta University 113 Wachira, Benson wachira.benson@students.ku.ac.ke Kenyatta University 114 Wainaina, Gituro wgituro@uonbi.ac.ke University of Nairobi 115 Wainaina, Joyce Director.hri@kalro.org KARLO-THIKA 116 Kathleen M katynach2@gmail.com Kenyatta University 117 Wamalu, Anglina wv_kenya@wvi.org World Vision -K 118 Wamalu, Angline wv_kenya@wvi.org Biovision Trust 119 Wambui, Mary Marywambui.gitau90@gmail.com EDULINK 120 Wangu, Lucy Wangui.lucy@ku.ac.ke Kenyatta University 121 Wanjiru, Carolyne G carolinegithiomi@gmail.com ICIPE 122 Wanjiuku, Salome wv_kenya@wvi.org World Vision Kenya 123 Wanjohi, Waceke wanjohi.waceke@ku.ac.ke Kenyatta University 124 Wanzetse, Evans Wanzetse.evans@ku.ac.ke Kenyatta University 125 Waswa, Fuchaka wfuchaka@yahoo.com Kenyatta University 126 Williams, Tennyson TennysonWilliams@worldanimalprotection.org World Animal Protection 127 Willy, Daniel Kyalo willy.kyalo@ku.ac.ke Kenyatta University 128 Yamo, Victor O. victoryamo@worldanimalprotection.org World Animal Protection Kenyatta University School of Agriculture and Enterprise Development Vision To be a dynamic and inclusive center of excellence in teaching, learning, research and providing agricultural services to humanity. Mission To provide a quality agricultural education and training; promote scholarship, service, innovation and creativity in sustainable agriculture. The Dean, School of Agriculture and Enterprise Development Kenyatta University P.O Box 43844-00100, Nairobi, Kenya Tel: +25420-8703634 Mobile: +254772 -865845 Email: dean-agriculture@ku.ac.ke or dean-saed@ku.ac.ke Website: www.ku.ac.ke/schools/agriculture Kenyatta University School of Agriculture and Enterprise Development Vision To be a dynamic and inclusive center of excellence in teaching, learning, research and providing agricultural services to humanity. Mission To provide a quality agricultural education and training; promote scholarship, service, innovation and creativity in sustainable agriculture. The Dean, School of Agriculture and Enterprise Development Kenyatta University P.O Box 43844-00100, Nairobi, Kenya Tel: +254 20-8703634 Mobile: +254772 -865845 Email: dean-agriculture@ku.ac.ke or deansaed@gmail.com Website: www.ku.ac.ke/schools/agriculture