456 CLASSIFICATION OF DEVONIAN FISHES 
ferior arches were ossified. Assuredly this is very unlike what we are 
accustomed to see among the Teleostei, but it must be recollected that 
it is at least equally unlike what we find in the Ganoids, if we except 
those of the same epoch ; and, on the other hand, there are some re- 
cent Teleostei, though there are no known Ganoidei, whose vertebral 
columns and skulls exhibit a correspondingly low stage of 
organization.! 
In the second place, arises the question whether, since we know 
that a true Ganoid, Ama, completely simulates the outward form of 
a Clupeoid Teleostean, while retaining all the essentials of its order,— 
may not Coccosteus be also a true Ganoid which simulates the out- 
ward aspect of a Siluroid? To this question it is, perhaps, impossible 
to give any answer, save by asking another, viz.:—Why should not a 
few Teleosteans have represented their order among the predominant ° 
Ganoids of the Devonian epoch, just as a few Ganoids remain among 
the predominant Teleosteans of the present day? When it is con- 
sidered that an ichthyologist might be acquainted with every fresh- 
water and marine fish of Europe, Asia, Southern Africa, Southern 
America, the Indian Archipelago, Polynesia, and Australia, and yet 
know of only one Ganoid, the Sturgeon, a fish so unlike the majority 
of its congeners, that a naturalist might be well acquainted with 
almost all the fossil Ganoids, and yet not recognize a sturgeon as a 
member of the group,—it will not seem difficult to admit the existence 
of a Teleostean among the Devonian Ganoids, even though that Tele- 
ostean should in some, even important, points differ from those with 
which we are familiar. 
At any rate, I think the prémd facie case in favour of the Tele- 
ostean nature of Coccosteus is so strong, that it can no longer be 
justifiable to rank it among the Ganoids, “sans phrase,’ but that even 
those who will not allow it to be Teleostean must attach to it the 
warning adjunct of zzcerte sedzs. 
No one doubts that wherever Coccosteus goes, Pterichthys must 
follow, and though the structure of the last-named fish is, in some re- 
spects, more difficult of interpretation than that of the former, in others 
it is strikingly Siluroid. For example, I know of no piscine structure 
that is even remotely comparable to the proximal joint of the pectoral 
limb of Pterichthys, except the corresponding articulation of the pec- 
toral spine and fin of the Siluroids. And again the example of Ostra- 
czon shows that the box-like cincture of the body of Prterzchthys is by 
no means foreign to the Teleostean group, though it cannot be par- 
alleled by fishes of any other order. Whether the other “ Placodermi” 
1 See on this point, however, the remarks at p. 457 under (3). 
