TEMPORAL VARIATION OF PREDATION ON RODENTS AND SHREWS BY SMALL AFRICAN FOREST CARNIVORES

Analysis of 904 small carnivore seats collected over a two-year period in a seasonal forest in the southwestern Central African Republic yie lded the remains of 732 shrews, rodents (<1 kg), and bats comprising a t least 34 species. Most seats belonged to the long-nosed mongoose, He rpestes naso, the numerically dominant carnivore in the study area. Sh rews were the principal small mammal prey with respect to number of in dividuals and frequency of occurrence. The mean monthly frequency of o ccurrence of shrews was 21.7% and of small rodents (<100 g) was 20.8%. Both shrews and small rodents occurred significantly more often in se ats collected during dry season months, and live-trapping of rodents o ver a 22-month period revealed similar peaks in availability. In contr ast to many mammalian carnivores, shrews did not appear to be used as an alternative prey source: not only did consumption levels of both pr ey types fluctuate in synchrony, but the availability of rodents appar ently had no effect upon the likelihood of including shrews in the die t. Preferential use of small mammals by long-nosed mongooses was teste d for the 10 rodent species for which abundance indices were calculate d. Mongooses preyed selectively on several rodent species (Praomys spp ., Deomys ferrugineus, Stochomys longicaudatus, and Grammomys rutilans ), while the most common rodent in the study area, Hylomyscus spp., wa s found in seats at lower frequencies than expected. In accordance wit h the habitat specificities of these 'preferred' rodent prey, it is su ggested that mongooses and perhaps other small carnivores were foragin g in micro-habitats characterized by higher-than-average vegetation un derstorey thickness where the simultaneous rate of encounter with arth ropods, their main prey, and shrews was apt to be high.