Tsavo National Park lies in an area of very dry country in South Eastern Kenya. Elephants live there in large numbers and in the early 2 1970's, they had a mean population density of about 1 km^-2 . They have inflicted widespread destruction on the woodlands which originally covered the Park, leading to a complete change in the appearance of the habitat. This caused considerable anxiety about the future of the Park and all its animal populations so research was begun to try to elucidate the root causes of the "elephant problem" and to predict its consequences. The purpose of the present study was to find out what happens to dead wood in Tsavo and to gauge the importance of termites in the ecology of the area. Section 1 gives a brief introduction to the history of Tsavo and sets out the main questions posed when this study was begun. In Section 2 the environment of the study area is described in terms of its topography and geology, the soils and their origins, the floristic and structural composition of the vegetation during this study, the climate and its variations in time and space and the role of fire in shaping the habitat. The species of termites found in the study area are listed in Section 3, followed by brief descriptions of their natural history. Section 4 contains all the information on the turnover of dead wood. The standing crop and rates of wood fall and removal were measured directly at one site and extended by means of an index to measure consumption of dead wood by termites at other sites. Consumption is found to be related to rainfall. The results are discussed in relation to similar studies made elsewhere. A model is constructed relating the consumption of dead wood by termites to rainfall and comparing this with the consumption of grass and browse by large herbivorous mammals. A method is described for measuring the relative importance of different termite species as consumers of dead wood.

The land-bird community of Tsavo East National Park was studied for two years. There are two wet seasons each year: November-December and April- May, but the latter were very poor in both study years. Six habitat types were designated on the basis of canopy cover of woody vegetation. Riverine vegetation was also studied. Most ofthe open habitats extant are derived from woodland by destruction of trees, a process largely caused by elephants. Food supplies are seasonally variable. Insects and grass seeds are most abundant in December and January. In the open habitats the few fruits also peak then, but Commiphora spp., which is only in woodland and is the most abundant fruit in Tsavo, peaks July to September. The ecology of each species of land-bird is described and most are found to be ecologically separate from each other. The annual cycle of numbers of birds in each habitat is closely correlated with the food supplies. In the open habitats birds are almost all insectivorous, granivorous or both, and peak numbers (about five times the dry season density) occur in December and January. In woodland, numbers remain fairly constant through the year and frugivores are more prominent, especially around August. The wet season peaks in all habitats are caused by immigrations of different species, both from elsewhere in Africa most of which come in to breed, and species which breed in the Palaearctic. The August peak in Woodland consists mainly of species which have resident populations. The results are particularly related to the recent habitat changes in the Park. They are compared with other bird studies in African savannas and other components of the Tsavo ecosystem. Finally, the means by which birds are able to survive in the harsh and unpredictable conditions of Tsavo are discussed.

The illegal bushmeat harvest has been identified as a reason for declining wildlife populations throughout much of Africa. For many years the trade was thought to exist primarily in Western Africa. The illegal use of bushmeat in Eastern Africa, including Kenya, went undocumented and unstudied. In 2004, the Born Free Foundation published a study which claimed illegal sale of wild game was substantial in butcheries throughout Nairobi, Kenya. In an effort to determine other markets of the commercial trade, the goal of this study was to analyze local restaurants in Voi, Kenya for illegal bushmeat sale. The town of Voi was selected due to recent published reports in the popular press, its proximity to Tsavo National Park, and its access to a major highway. Samples were collected and analyzed using mDNA sequencing analysis of the cytochrome B gene. None of the collected samples were identified as illegal game meat. The restaurants in Voi, Kenya were not a commercial outlet for illegal bushmeat trading in the local economy during the period of this study. The results from this study provide valuable baseline data which can be used in future research to help determine possible vectors of the bushmeat trade.