Studies of the distribution and feeding ecology of small mammals in Ngong Hills
Nzui, Issa Aggunday Ramadhan
MetadataAfficher la notice complète
The study was carried out in Ngong Hills about four kilometres from Kiserian shopping centre off Magadi road in the University of Nairobi forest station 1°_ 26'S. 36° 38'E. at an altitude of 2250m. According to Pratt and Gwynne  the site was classified ~s a dry sub-humid grassland zone with average rainfall of 800mm per annum and average temperature of the study area was divided into 130 contiguous quadrats each measured 10xlOm. Their comers were demarcated with flagged sticks for easy recognition of boundaries and each Ofthem was given a reference number. Vegetation identification and distribution was carried out and a sketch of the four different plant communities was prepared. Only 126 quadrats were sampled due to a shortfall in the number of traps. A mixture of small mammal traps were used namely 30 large Shennan, 72 medium Shennan and 24 Longworth, A mark-recapture method was used whereby one digit per limb was amputated as per the system which was developed by French  and illustrated by DeBIase and Martin in their Manual of Mammalogv . The field studv was conducted between 21st October to 20th December 1993. During that period a total of 49 days were used for trapping. The trapping sessions were divided into 7 day-units of continuous trapping followed by a pause of one or more days. In all there were a total of 6174 trap-nights during which 213 animals were marked and subsequently recaptured severally. Overall recaptures were 2852 which was nearly 46%m success. A total of 8 species were involved and the number of animals marked per species were 137 Rhabdomys pumilio, 29 Otomys tropiealis, 13 Mus minutoides, 11 Lophuromys flavopuntatus, 9 Dendromus mesomelas, 5 Lemniscomys striatus, 3 Grammomys dolichurus and 6 Crocidura bicolor. R. pumilio was a cosmopolitan species: it occurred in 75 out of 126 quadrats besides being the most abundant species in the site. There was 110 difference in the species trap-responses between large Shennan and Longworth traps [t=12 l.13, P 0.05]. However the species was more trappable by medium Shennan traps than by either Longworth (t=12 8.5 p>-0.05) or large Shennan traps (t=12 7.8: p<O.O»). There was no sexual difference in trappability. Both 0. tropicalis and G. dolichurus were not caught in Longv..orth traps. Probably their entrances were too small to accommodate the sizes of both species. With regard to time span between captures in day as fo11ows:-R. pumilio 3.36, O. tropicalis 4, M. minutoides 3.35, L. flavopuntatus 3.72 and D. mesomelas was 3.76. Their biomass and densities were as follows:- SPECIES DENSITY ind/h BIOMASS/kg/h R. pumilio 108 2.7 O.tropicalis 23 2.76 M. minutoides 11 0.143 The distributions of the various species were confined mostly to areas where their preferred micro habitats occurred. R. puinilio was found in grass and light bush plant communities while 0. tropicalis was found in the ecotone zones between grassland and bush areas. Both G. dolichurus and D. mesomelas occurred in bush and gallery forest areas with G. doli churus showing preference to gallery forest while D. mesomelas mostly occurred in bushy areas. A1. minutoides inhabited grassland and light bush areas while L. flavopunctatus was mostly found in the thick bushes and forest. L. striatus were trapped in light bush areas and C. bicolor occurred in most of the areas except deep inside gallery forest. BothR. pumilio and O. tropicalis made distinct runways which were used as both home ranges and territories. L. flavopunctatus also made runways but in some parts they were not clearly defined. A1. minutoides and C. bicolor did not have discernible runways. L. striatus were few and were not caught in one place therefore attempts to find their runways were unsuccessful. All the above species were terrestrial and they lived in burrows, holes or cracks in the ground. Both 0. tropicalis and R. pumulio also made grass nests on the ground. On the other hand both D. mesomelas and D. dolichurus were arboreal and nocturnal and made nests on vegetation. They only descended to the ground occasionally. The feeding habits of the both R. pumilio and 0. tropicalis were established by:- 1. direct observations through binoculars 2. feeding experiments 3. faecal and stomach content examination 0. tropiealis was found to be a grazer feeding mostly on grasses such as Digiraria macroblephara, Sporobolus macranihelus, Cynodon aethiopicus and Panicum sp. They mostly fed on leaves and stems and they had caeca which helped them in microbial digestion of cellulose. R. pumilio was found to be omnivorous reeding on grasses, dicots and insects. As far as the other species were concerned, L. flavopunctatus fed on grasses and insects, M. minutoides fed on seeds and insects. The same case also applied 10 both G. dolichurus and D. mesomelas. But C. bicolor was insectivorous. During most of the early parts of the study period neither pregnant nor lactating females were caught. But one notable change was noticed ten days after the rains. Many females of R. pumilio and O. tropicalis were observed haying relaxed vaginal openings a phenomenon which was not apparent hitherto. The coincidence of the relaxed vaginal openings after the onset of rains probably indicated the species concerned were oestrous. Delany (1964), Okia (1973) and Southern and Hook (1963) in their studies of rodents recorded tendencies of various rodent species to have breeding peaks after the onset of rains. The observation at Ngong Hills could probably be an indication that the species were starting to breed. Examination of carcasses of females in December showed early pregnancies.
- MST-Zoological Sciences