No. 375.] AGASSIZ’S WORK ON FOSSIL FISHES. 183 
ishment that Cuvier felt in examining the Plesiosaurus I myself experienced 
when Mr. H. Miller, the first discoverer of these fossils, showed me the 
specimens which he had collected in the Old Red Sandstone of Cromarty. 
Any one who has attempted for himself to decipher the dis- 
torted and for the most part obscure remains from the Scottish 
Old Red can imagine the difficulties which the first students of 
such extraordinary forms labored under. He will understand 
that above all scrupulous refinement of observation is necessary ; 
that innumerable comparisons and attentive reéxaminations of 
even the most tattered fragments must be made in order to test 
his hypothesis of the association of parts. Considering the 
means at Agassiz’s disposal, his work must be pronounced 
nothing short of brilliant; it was remarkable alike for the origi- 
nality and insight displayed, and for the general correctness of 
his conclusions. That some of his generalizations should have 
been premature was an inevitable consequence of pioneer work. 
And if, after more than fifty years, certain of his views are 
found to require modification, or to be no longer tenable, what 
more was to have been expected ? 
To cite one or two instances by way of illustration, let us 
suppose we grant with Cope that the Ostracodermi are not 
fishes, properly speaking, but belong to a group at the base of 
the craniate Vertebrata, characterized by the lack of a lower 
jaw and of paired limbs; how does that detract any from the 
unerring judgment of Agassiz, who pronounced them first of all 
to be chordates, and assigned them a place among the most 
primitive of ganoids? If we criticise his restorations of 
Pterichthys and Coccosteus as.being crude and fanciful, we 
cannot accuse him, at all events, of misrepresentation. Just as 
it required the genius of a Traquair after many years of patient 
study to prove that the Platysomidz are in no sense whatsoever 
related to the pycnodonts,! so, too, it required the combined 
efforts of the best Russian, ‘German, and British talent to 
unravel the complicated structure of the coccosteids and ostra- 
coderms. Tremendous advances have since been made, almost 
as a matter of course, but it was Agassiz who first clearly 
1 R. H. Traquair, On the Structure and Affinities of the Platysomide. Zrans. 
Roy. Soc. Edinburgh, vol. xxix, 1879, pp. 343-391: 
