DR. A. GUNTHER’S DESCRIPTION OF CERATODUS. 
553 
ends the extremities of the interspinous bones as in the Dipnoi. But these bones are 
very much reduced in number, and enlarged, the vertical fins themselves being entirely 
separate from one another. The peculiar termination of the vertebral column, with an 
unequal development of the caudal lobes, represents a most curious intermediate con- 
dition between the diphycercal tail of the Sirenidce and the heterocercal of Dipterus. 
To this is added “ the complete ossification and segmentation of the vertebral column, 
in which respect this genus stands alone among the contemporaneous fishes.” Unfor- 
tunately the head and base of the paired fins are destroyed in the only two specimens 
known ; and it is chiefly the last-named character which prevents me from associating 
this genus with the Dipnoi. 
However uncertain the affinities of the last-named genera must appear from a zoolo- 
gical point of view, I believe that I have shown that Ceratodus clearly proves the cor- 
rectness of Professor Huxley’s view regarding the similarity of the Lepidosiren- limb 
with the fringed fin of certain Crossopterygians, and that Ceratodus , Lepidosiren, and 
Dipterus are most closely allied forms, and must remain together in the same suborder. 
Consequently if we retained the suborder Crossopterygians with the limits assigned to it 
by its author, it would comprise four recent Ganoid genera, viz. Polypterus, Ceratodus , 
Protopterus , and Lepidosiren ; or, in other words, these four genera would require to be 
regarded as more nearly allied to each other than to the other recent Ganoids. I am 
not prepared to adopt this view. Muller was certainly right in regarding the condition 
of the skeleton as a character of primary importance for the systematic division of Ganoids, 
and in supporting his opinion by the analogous case of the Chondropterygians, of which 
the Plagiostomes have the skeleton completely divided into segments, whilst the “Holo- 
cephala with their notochordal skeleton form a division distinguished in a marked manner 
also in other respects” (Ganoid, p. 150). Thus also I have had repeated occasion to 
draw attention to identity of structure in Ceratodus and Acipenser ; and when we consider 
that the notochordal or segmented condition of the skeleton is systematically of as great 
importance at least as the state of development of an air-bladder into a lung, we must 
admit that the points of affinity of Ceratodus to Acipenser counterbalance those existing 
between Ceratodus and Polypterus , and consequently that the fishes named are repre- 
sentatives of cosubordinate divisions of Ganoidei. 
Thus, from an examination of the living forms, I am inclined to withdraw from the 
Crossopterygians some of the component parts of this suborder ; and I am encouraged 
in this by the following additional consideration. With the knowledge obtained by 
means of Ceratodus , we are now able to define more distinctly two types of “ fringed 
fin,” already indicated by Professor Huxley, who terms them “ acutely lobate ” and 
“ obtusely lobate.” Fringed fins of the former type are long, pointed, like those of 
Ceratodus , or even narrower ; they are covered with small scales along the middle, and 
surrounded by a cutaneous fringe containing innumerable fine fin-rays. There can 
scarcely be any doubt that such fins in fossil fishes were supported, as in Dipnoi, by an 
axial cartilaginous skeleton, extending from the base to the extremity. I have never 
