The habitat and plant feeding of 64 well-habituated, individually identified adult male and female yellow baboons was studied for 5 years at Mikumi National Park, Tanzania. Variation across the years showed that a study of only one or two years would have been incomplete and misleading. The list of baboon food species obtained from Mikumi is considerably larger and more diverse than any previously reported. One to six plant parts were eaten from each of more than 180 species. The 25 most common tree genera all contained species used for food. Of the 50 most common grass, shrub and herb genera, 93% included plant foods. Using months in which a species was eaten during at least one year of the study, 21 staple species were eaten during a mean of 8.86 months and 7 were eaten in all 12 months. Although many foods were from commonly available plant species, 15 such species were only rarely eaten. The number of parts of a species eaten per month and an estimate of the amounts eaten per month both varied with temperature and rainfall, being lowest near the end of the cool, dry season. There were substantial differences from year to year in the timing and amount of food production of many species; nevertheless, the same broad feeding pattern was repeated in each of the 5 years of the study. Despite yearly variation in food availability, 14 or more staples and other common foods were eaten in any given month. If crops of many of these foods were to fail, a large number of less commonly eaten species could be substituted. Baboons are eclectic feeders that appear to be optimizing their diet by selective feeding from among a wide array of available foods in an ever-changing floristic environment.

Aim Detailed knowledge of species distributions, endemism patterns and threats is critical to site prioritization and conservation planning. However, data from biodiversity inventories are still limited, especially for tropical forests, and even well recognized hotspots remain understudied. We provide an example of how updated knowledge on species occurrence from strategically directed biodiversity surveys can change knowledge on perceived biodiversity importance, and facilitate understanding diversity patterns and the delivery of conservation recommendations.Location Eastern Arc Mountains (EAM), Kenya and Tanzania.Methods We surveyed amphibians, reptiles, birds and mammals during 2005-2009, targeting mountain blocks that had been poorly surveyed or unsurveyed by the early Noughties. We combined new and old data to produce a database of species presence by mountain block spanning four decades of research. Species richness is regressed against survey effort, funding, ecological and human disturbance factors to analyse the best predictors of vertebrate richness across mountain blocks. Similarity among species assemblages among blocks is analysed using dissimilarity analysis.Results New surveys raised the number of endemic and regional endemic vertebrates by 24% (from 170 to 211 species), including 27 new species of which 23 are amphibians and reptiles. Vertebrate richness is best explained by forest area, but rainfall is also important, especially for amphibians and reptiles. Forest elevational range is important for mammals and for block-endemic birds. Funding explains 19% of the variation in total species richness, while survey effort generally explains < 10% of variance. Cluster analysis shows that species assemblages are partitioned by geographical proximity among mountain blocks.Main conclusions The biological value of the EAM has been underestimated, and strategic surveys are important even in well-recognized hotspots. The exceptional and global importance of these mountains for endemic vertebrates is highlighted, supporting the development of a network of Nature Reserves and the proposed inclusion within UNESCO's natural World Heritage Sites.

Long‐term solutions to crop raiding by elephants (Loxodonta africana) should be based on an understanding of their behaviour and ecology. The real and perceived risks from humans have been shown to affect elephant behaviour. This is evidenced by elephants predominantly raiding crops at night, avoiding the height of human activity. If such human avoidance behaviours are apparent, it might also be expected that elephants avoid risks associated with higher visibility and increased human activity as may occur during the full moon. However, elephant nocturnal crop‐raiding behaviour in relation to lunar cycles has largely been a neglected factor in studies of human–elephant interactions. In this study around Mikumi National Park, Tanzania, we apply circular statistics in this context for the first time to show a significant decrease in crop raiding during the full moon and apply this method retrospectively to data from another site in West Africa with similar results. Additionally, a greater proportion of farms raided was guarded during the full moon than any other moon phase. Our results indicate that variations in crop raiding with lunar phase could be a general feature of elephant behaviour and thus could be used to design and time mitigation efforts.

Ageing and social status effects on reproductive condition are examined in 106 wild female baboons from 1974 to 1997. The mean duration of each reproductive state (follicular and luteal phase components, pregnancy and lactation) was examined per female, using a repeated measures analysis of variance across each of four age classes (3-6, 7-10, 11-14, and 15-20+ year olds) and dominance rank quartiles. The sex-skin swelling portion of the follicular phase increased whereas the fully swollen portion decreased in duration with advancing age and with low dominance rank. The detumescing portion of the luteal phase increased with advancing age, being most pronounced in lowest ranking females. The number of cycles to conception decreased with age, with no dominance rank effect. However, owing to variation in lactation duration, the interbirth interval was unaffected by age but was shorter in high ranking females. The miscarriage rate was unaffected by age but increased as female rank declined. Age and rank effects became significantly more pronounced in response to a crash in the study population between 1987 and 1993. In conclusion, age- and rank-related reproductive changes in baboons, like those in humans, appear to represent attempts to compensate endocrinologically for accelerated ovarian atresia and increased environmental hardship.

No abstract