VARIATIONS IN THE NUMBERS OF ANIMALS 145 



land was much greater than on ploughed land. If rooks had 

 been much rarer birds we might have found that they were 

 exclusively attached to ploughed fields at that time of the year. 

 23. Dugmore's description-*^^ of the African buffalo (Bos 

 coffer) in the Sudan shows in a very interesting way how the 

 habits of an animal may depend upon its numbers. The 

 wild buffalo which inhabits a large part of Africa used to be 

 one of the most abundant large animals in that country, but in 

 1890 a frightful epidemic of rinderpest swept the continent, 

 killing off, amongst other species, enormous numbers of 

 buffaloes, and almost exterminating them in many parts of 

 the country-. Before the great epidemic the buffalo used to 

 live in herds out in the open grassland and feed by day, but 

 " for many years after [1890] the few remaining animals fed at 

 night and retired to forests and dense swamps during the 

 day." This was still the condition of affairs in 1910, but 

 within recent years the buffalo have begun to increase again 

 greatly and appear to have gone back to their former habits. 

 This fact is confirmed by the obsers'ations of Percival ^-^^ and 

 Chapman *^^" in Africa. It will be noticed that the African 

 buffalo took about thirty-five years to recover its numbers. 

 It is interesting, therefore, to find that the greater kudu 

 {Strepsiceros bed) recovered in a much shorter time (ten to 

 twenty years) from the rinderpest epidemic, although at the 

 time it was almost ^^^ped out. This quicker recovery is 

 associated ^^ith the fact that it is a much rarer animal than 

 the buffalo, and so requires less time to regain its maximum 

 density of numbers. ^-^^ 



