RECAPITULATION. 497 



by generalised forms corresponding to the diagrams of 

 primitive ancestors invented by the ingenious morphologist. 

 As we are forced to modify our earlier views by additional 

 evidence we become more and more convinced that it is 

 not length of time which is so important in organic evolu- 

 tion, /.e'., in the modification of structure, but change of the 

 conditions of life. There can be no doubt that the hetero- 

 cercal condition of the tail in fishes has been derived from 

 the diphycercal, but the latter is scarcely less common at the 

 present day than it was among the fishes of the Palaeozoic 

 age. In both cases it appears to exist in connection 

 with certain habits, certain modes of swimming, which are 

 not confined to one kind of habitat. Thus we find it in more 

 or less perfect development in ClUainydoselacIie, which is a 

 deep-water shark, in Protopterus which is a Dipnoan living 

 in rivers, and the Ganoid Polypterus, likewise a fiuviatile 

 fish. In all these cases the movement of the fish seems to 

 be rather slow, and horizontal in direction. 



To return to the heterocercal tail. It exists both at the 

 present day and in former periods, back to the Palaeozoic, 

 among the fishes now united under the name Teleostomes, 

 formerly separated into the tw^o divisions Ganoids and 

 Teleosteans. Instead of these two divisions we have now 

 the two sub-divisions Crossopterygians and Actinopte- 

 rygians, distinguished by the structure of the paired fins, 

 the former having fringed fins, the latter fan-shaped fins. 

 The heterocercal tail is quite as exclusively characteristic of 

 the Actinopterygians as the structure of their paired fins. 

 Both types have persisted from the Palaeozoic period to 

 the present day, but the Teleosteans with homocercal tails 

 are undoubtedly connected by a series of intermediate 

 forms with the heterocercal Actinopterygians. The question 

 then which we have to consider is whether there is evidence 

 that the earlier heterocercal Actinopterygians sought their 

 food upon the bottom. 



In answer to this question we can say that fossil forms 

 similar to the sturgeon are very ancient. In Palceoniscus 

 the snout is slightly produced beyond the jaws but not so 

 much as in the sturo'eon. It is found in the Permian. But 



