Permanence and Evohtiion. 165 



ancestors, or whether they suppose them to have 

 greatly changed. If they think the former, 

 they are involved in great difficulties ; if to 

 escape these they resort to the latter alternative, 

 they undermine the whole geographical argu- 

 ment for evolution. E.g., if South America and 

 South Africa were ab initio perfectly separate, 

 how do we account for the presence in both of 

 atruthious birds ? If they were recently con- 

 nected by land now submerged, then what 

 becomes of the argument from the generally 

 different facies of their faunas ? 



Again, how is it explained that the mamma- 

 lian fauna of Australia consists almost entirely 

 of marsupials ? To give this any meaning on 

 the hypothesis of evolution, it must be supposed 

 that Australia was separated from Asia ever 

 since the beginning of placental mammalia, i.e., 

 at the very beginning of the tertiary period at 

 least. For marsupials were once dominant over 

 the whole world, therefore no local product of 



