MY SOMALI BOOK 223 



explained on the former supposition. While the white 

 patch on the rump of the aoul is probably an instance 

 of the latter. The absurd idea that such a patch 

 could be obliterative and have a concealing effect has 

 been convincingly disposed of by Colonel Roosevelt. 

 The oryx face markings may possibly come in here as 

 an aid to the young. 



But admitting the existence of recognition marks 

 in certain cases, there is no doubt that the theory 

 has been carried to extremes, and that it cannot apply 

 to many colour markings for which it has been held 

 accountable. Moreover, a great number of animals 

 do not possess these so-called recognition marks ; 

 while their existence in some cases seems to be in 

 direct conflict with the idea of concealing colouration 

 as applied to the identical animal. 



It seems to me, then, clear that neither sexual 

 selection nor the need for mutual recognition can be 

 of sufficiently general application to take the place of 

 the necessity for protection as the main factor in fixing 

 animal colouration. So far, that is, as our knowledge 

 extends, and admitting that every detail of its colouring 

 may not always be of direct utility to an animal. 



Are we then to revise our ideas of the severity of 

 the struggle for existence and reject natural selection 

 as the main factor in evolution, at any rate as regards 

 colour ? Did Darwin build no better than that ? 



I propose nothing so drastic. What we have to do 

 is to go back and see, firstly, whether Mr. Selous' 

 observations really go as far as appears at first sight 

 towards proving the non-existence of protective 

 colouration in many animals ; and secondly, in so far 



