92 CATALOG OF FOSSIL FISHES IN THE MUSEUM 



tubercles are all low, smooth and rounded, and may be described 

 almost in the words appHed by Whiteaves to those of the Manitoba 

 specimens: 



The surface markings . . . consist of numerous small, smooth and 

 rounded tubercles, which are imequal in size and irregular in their distribution, 

 though the largest average 2 mm. in diameter at the base, and from 2 to 5 mm. 

 in their distance apart at the summits. The greater part of the beveled or 

 outer margin is smooth to the naked eye, but around its outer limits there are 

 indications of short and irregular radiating grooves and ridges.*" 



It appears therefore, even on the evidence of the ornamentation 

 alone, that the Conodont specimens do not belong to Trachosteus. 

 There is, also, the further evidence from the shape of the plates. In 

 Trachosteus the antero-ventrolateral is short and broad, and very 

 different in form from the specimens in hand. 



We have also compared the specimens with the genus Glyptaspis, 

 using for comparison the type specimens which are in the Newberry 

 collection in the American Museum. In Glyptaspis the antero- 

 ventrolateral is fortunately known, so that we are able to compare 

 plates of the same kind. But the denticles of the two forms are 

 very different. In Glyptaspis tbey have a tendency to run into lines, 

 and this is particularly well shown on the antero-ventrolateral. In the 

 Conodont specimen of this plate, on the other hand, the denticles 

 never run into lines, but always remain discrete. Furthermore, in 

 Glyptaspis the antero-ventrolateral has a broad, smooth border on its 

 outer or ornamented face, whereas in the Conodont antero-ventro- 

 lateral there is no such border, the tubercles running clear to the mar- 

 gin on one side, and on the other terminating in an irregularly delimited 

 smooth area, not sharply defined from the ornamented portion. It 

 seems, therefore, that the Conodont specimens cannot be placed in 

 the genus Glyptaspis. 



There then remains only the genus Aspidichthys of forms with 

 tubercles comparable with those of the present specimens. In Aspi- 

 dichthys clavatus Newberry, the tubercles are much larger than in either 

 the present specimens or the Manitoban plates. However, their 

 height, roundness, distribution, and the occurrence of smaller denticles 

 amongst the larger ones, is in Aspidichthys very much as in the spec- 

 imens in hand. The mere difference in size of the denticles would 

 not exclude the present specimens from that genus, since it is known 



*°Loc. cit., p. 3?5. 



