328 THE AMERICAN NATURALIST. — [Vor. XXXV. 
to our knowledge of Paleozoic fishes than those of any other 
single investigator. And the propriety of electing a paleontologist 
to preside over its sessions is abundantly confirmed by the brilliant 
address of sixteen closely printed pages with which its sittings were 
opened. 
In this essay Dr. Traquair has done for the class of fishes what 
Marsh attempted for American fossil vertebrates in general, and 
what Osborn accomplished for the Mammalia, in their addresses 
before the sister association! in this country some years ago ; and 
these three summaries, taken together, constitute a very important 
chapter in the literature of vertebrate paleontology. Although not 
retrospective in an historical sense, Dr. Traquair's paper is in effect 
a clarified review, expressed in terms of evolution, of the leading 
philosophic deductions which the science of ichthyic paleontology 
has afforded up to the present time. Its significance consists in a 
clearer recognition of the relationships between different groups of 
fishes, together with a more precise indication of their lines of 
descent; and it contains also critical observations on various dis- 
puted points, such as the origin of paired fins, development of dental 
plates and dermal armor from shagreen-like scales, atrophy of the 
lower jaw and shoulder-girdle, modification of the caudal fin, and 
similar issues. And in conclusion it is stated that *the paleon- 
tology of fishes is not less emphatic in the support of the doctrine 
of descent than that of any other division of the animal kingdom”; 
also that “we do not and cannot know the oldest fishes, as they 
would not have had hard parts for preservation, but we may hope to 
come to know many more old ones, and older ones still, than we do 
at present." 
The author devotes special consideration to those ancient forms 
whose structure he has so ably elucidated in earlier memoirs. AS 
for the ostracoderms, a most enigmatical group, which he at one 
— supposed were derived from the primitive elasmobranch stem 
(owing to the resemblance of Thelodus scales to shagreen), an 
independent origin is now admitted to be possible; but the idea is 
discredited that they had any share in the evolution of more recent 
types of fishes. The Ceelolepide, to which **Cephalaspis as well 
as Pteraspis and its allies are traceable," were certainly shark-like | 
1) i . 
P Marsh, O. C. Introduction and Succession of Vertebrate Life in Am 
roc. Amer. Assoc. Adv. Sci., Nashville Meeting, 1873. Madison 
sborn, H. F. The Rise of the Mammalia in North America, iid. 
Meeting, 1893. 
