Odhiambo, Richard Oketch2012-02-282012-02-282012-02-28http://ir-library.ku.ac.ke/handle/123456789/2861The QL 737.R 666 O3Anthropogenic activities that enhance subdivision, conversion and fragmentation of tropical rain forests continue to pose a serious threat to biological diversity. The rain forests of Taita Hills, other Eastern Arc mountain forests and the coastal forests of Kenya and Tanzania are under severe threat due to demand for hardwood products and agricultural land. These forests are of biological importance due to their high generic and species endemicity and geographically as important catchment areas. In order to make informed conservation decision, an inventory of the species and their ecology is important. This study assessed the diversity of small rodents in seven forest patches of the Taita Hills. Aspects of population ecology and behaviour of an endemic forest dependent rodent, Praomys taitae, were studied in two forest fragments, Ngangao and Chawia. Diversity was assessed through removal study using baited traps. Praomys ecology was studied using a Capture-Mark-Recapture (CMR) technique. A total of 848 captures of eight species of rodents and shrews were made in 7220 trap nights during the removal study. Trap success varied between between fragments, being lowest in Ndiwenyi (6.8%) and highest in Chawia (19.5%). Chawia had the highest number of rodent species and Ngangao the least. Praomys taitae comprised the highest proportion of rodents caught in all the fragments accounting for 96% of rodent captures in Ngangao. Praomys and Graphiurus showed a wide distribution being captured in all fragments. Other species captured included, Mus spp. (six fragments), Grammomys and Dendromus (four fragments), Aethomys and Mastomys (three fragments), and Tatera (two fragments). Species diversity was highest at Macha (D = 3.2354 and H' = 1.3432) and least at Ngangao (D = 1.0931 and H' = 0.2089). The same pattern was recorded for evenness being 0.4575 and 0.0645 in Macha and Ngangao forests, respectively. A canonical variate analysis on the morphometric and craniodental variables of the two coat colour forms observed in Praomys in Taita Hills did not show any significant variation. Likewise, a cytogenetic study of individuals of the two morphotypes showed a similar diploid number 2n=48. The structure of the karyotypes were identical as far as both the autosomal and sex chromosomes are concerned. The chromosomes appeared as either telocentric (chromosome 1-17) or metacentric (chromosome 18-48). There were a total of 963 captures of 380 individual rodents and 58 individual shrews in 6000 trap nights during the CMR study. Rodents captured included six species belonging to two families. These included five murids (Praomys taitae, Grammomys sp., Mus spp., Rattus rattus and Mastomys sp.) and a myoxid (Graphiurus sp.). Praomys contributed over 80% of rodent captures in both forests. The highest species number (seven) was recorded at Chawia and least (three) at Ngangao. The overall mean trap success was 15.1% being in Chawia (24.5% than in Ngangao (7.5%). In Chawia fore I recorded higher values in terms of Population estimates (2=54.743902, df=1, P<0.001), individual captures (2=42.711111, df=1, P<0.001) and biomass (2=40.333333, df=1, P<0.001) as compared to Ngangao forest. Overall, the mean (±SEM) probability of capture was 0.588±1.278 and 0.557±3.73 for Ngangao and Chawia forests, respectively. Apart from April estimates in Chawia (2=11.84507, df=1, P<0.001), there were no significant differences between the monthly population estimates in the grids of the different forest fragments. With an exception of April and July (2=6.81818, df=1, P<0.01) and April and August (2=4.8285714, df=1, P<0.05), there was no significant variation in the number of Praomys between months in both fragments. Likewise, the number of individuals caught did not vary significantly from month to month in Chawia. In Ngangao, significant (P<0.05) variation did occur April and the other months. Both removal and CMR studies showed that Praomys taitae breeds throughout the year. The sex-ratio was not significantly different from the expected 1:1. In P. taitae, both males and females were trapped within small home ranges during each trapping occasion. Except for Ngangao females and Chawia males (t=-2.0943, df==37, p<0.05; t=-2.3872. df=37, p<0.05 for home range sizes and lengths, respectively), there was no significant difference between the home ranges of males and females. The present study showed that disturbed forests supported higher densities and biomass of rodents than intact forests. This may be explained by the invasion of disturbed forests by Savannah species such as Mastomys. This study also observed that the abundance of the forest dependent rodent, Praomys taitae is negatively affected by disturbance. Similarity in morphometric and cytogenetic data among Praomys from the different fragments would suggest one panmictic population. Praomys was, however, found to move within small home ranges and maintains a 1:1 sex ratio irrespective of habitat quality.enMuridae--Kenya--Taita Hill//Rodents--Kenya--Taita HillsThesis