94 E. T. NEWTON ON FISHES' TAILS. 



find highly specialized forms, some being representatives of living 

 genera and having their tails formed on precisely the same plan as 

 those which we call homo-cereal among recent Teleostei. It is not 

 proposed to carry these comparisons any further at present, but it 

 is hoped that the accompanying table (p. 96), which shows some of 

 the facts above referred to, will help to put them in a clearer light. 

 This study of the fossil forms in the light of development leads 

 us to the following conclusions : — 



1. The Dipnoi having its earliest representative Dipterus, with 

 a heterocercal tail, and its recent examples with diphycercal tails, 

 shows a retrogression and not an advance. 



2. The Elasmobranchii have probably remained much the same 

 through all time since their first appearance in the Ludlow to the 

 present day. 



3. The Ganoidei, in their earliest known representative, Cepha- 

 laspis, were heterocercal, and therefore somewhat advanced, while 

 in the Devonian Rocks we have all the forms of tails which are now 

 living, with the exception possibly of the Amia, and might, there- 

 fore, be held to show a persistence of type through time, and not 

 an advance. 



On the other hand, although heterocercal tails are found in these 

 older rocks, yet the highest type is more embryonic than many that 

 are met with in the Secondary strata, where forms occur as high, or 

 even higher than, those of the present day. And consequently the 

 Ganoids of the Secondary strata are of a higher type than those of 

 the Primary rocks. But with regard to the Secondary Ganoids, it 

 must be remembered that we do not here, any more than in the 

 older formations, find a gradual advance from the forms of the 

 lowest Secondary Rocks to the highest ; for in the Lias we have 

 fishes with tails as highly organized as we find in the Purbecks, 

 and these agreeing as closely as possible with recent forms. So that 

 we cannot say the Ganoids have advanced in structure between the 

 Lias and the present day. 



4. Teleostei, when first we recognise them with certainty, in the 

 later Cretaceous rocks, they have an organization as high, so far 

 as we can tell, as any living representative of the group. 



It is a remarkable fact that the Teleostei are first known at the 

 time when the Ganoids are declining, and one is naturally led to ask 

 whether the one has descended from the other. At present this cannot 

 be answered ; but it seems quite possible, so far as the form of the 



