THE DISTRIBUTION AND CHARACTERS OF REPTILES. 



-M7 



(by the amount of intergrading within the group as to make this 

 statement : 



I am convinced that if all the tortoises belonging to the gcuiiictnca 

 group at present living in South Africa could be gathered together . . . 

 there wor.hl be an almost ini]ierceptible passage from one so-called species 

 to another : there would be numerous specimens of which it would be 

 impossible to say to which of two species they should belong. 



However this statement can only be considered as a possi- 

 bihty, for as yet nothing truly intermediate between ocuHfera and 

 verreauxii has been brought to light, though Miss Wilman, of 

 the Kimberley Museum, has secured numerous tortoises from 

 a district where both these forms occur. In my opinion there is 

 real discontinuity in this case at the present day. 



No doubt, those of us who deny the occurrence of mutations 

 in specific characters and yet admit that species are real entities, 



ooZ, GEOMETRICA 



Distribution of Tcstudo gcontctrica and allies. 



will find some difficulty in explaining the structural discontinuity 

 between species. Probal)ly the simplest explanation is to suppose 

 that a complete range of intermediates has existed, but such inter- 

 mediates have either been directly cut out or have merged into 

 one of the divergent stocks. This is indeed the \ iew actually held 

 by De Vries. That an actual c-utting out has occurred in some 

 cases is obvious enough from the distribution data : for instance, 

 the genus Tctradactylus in Western Cape Colony extends from 

 Capetown to Knysna. and is quite tmknown in Eastern Cape 

 Colony, though it is not uncommon in parts of Natal, Zululand, 

 and the Transvaal. The snake Amphorhiniis multimacidatus is 

 common on the high veld of the Transvaal and in the neighbour- 

 hood of the Cape Division, but is quite unknown in intervening 

 districts. But such cases are rather rare, and generally speaking 

 the areas of nearly related species either closely adjoin or actually 

 overlap. It may be presuined that the extent of isolation in habit 

 that obtains between two divergent stocks will largely determine 

 the chances of survival of intennediates : if strong ditiferences of 



