CLASSIFICATION OF DEVONIAN FISHES 447 
for analogies too far to discover in the stiff-walled lungs of Lepidosiren, 
a structure more nearly representing the ossified air bladder of the 
Ccelacanths than any with which we are at present acquainted, among 
recent or fossil fishes. Furthermore, Lepédostren is the only fish 
whose teeth are comparable in form and arrangement to those of 
Dipterus Though Lepidostren may not be included among the Cros- 
sopterygide, nor even in the order of the Ganoidei, the relations just 
pointed out are not the less distinct ; and, perhaps, they gain in interest 
when we reflect, that while Polypferus, the modern representative of 
the rhombiferous Crossopterygide, is that fish which has the most 
completely lung-like of all air bladders, Lepdosiren, which has been 
just shown to be, if not the modern representative of the cycliferous 
Crossopterygide, yet their “next of kin” is the only fish which is 
provided with true lungs. These are unquestionable facts. I leave 
their bearing upon the great problems of zoological theory to be 
developed by every one for himself. 
The preceding discussion of the affinities of the Devonian genera, 
Osteolepis, Diplopterus, Glyptolemus, Glyptopomus, Gyroptychius, Holo- 
ptychius, Glyptolepts, Dendrodus, Phaneropleuron, Dipterus, was an 
indispensable preliminary to the consideration of the main question 
with which I proposed to deal in the present essay, viz., What, and 
how many, groups of fishes are represented in the Fauna of the 
Devonian epoch? a Fauna which presents a surpassing interest, when 
we recollect that it comprises the oldest assemblage of vertebrate ani- 
mals, of which we possess a more than fragmentary knowledge ; that 
its constituents abound in certain localities ; and that, for many years 
past, they have been the subject of careful and repeated collection and 
investigation. An examination of the data collected up to the pre- 
sent time has led me to the following conclusions, some of which are 
already current, while others are new :— 
1. No vertebrate animal higher in the scale than fishes is as yet 
certainly known to have been found in any rock of Devonian age. 
In fact, until demonstrative stratigraphical evidence of the Devonian 
age of the well-known Elgin beds is obtained, the bearing of the pala- 
ontological evidence against that conclusion is too strong to allow of 
its being entertained. 
2. Of the six orders of the class Pisces, three, namely, the DIPNOT, 
MARSIPOBRANCHII, and PHARYNGOBRANCHII, are certainly not repre- 
sented by any known Devonian fish. In endeavouring to estimate 
the value of this negative fact, we must recollect that no fish belonging 
1 Prof. Pander has drawn attention to the resemblance of the teeth of his genus 
Holodus to those of Lepidostren, but it is not clear that he regards Holodus as a Ctenodo- 
dipterine. 
