BUFFALO SOCIETY OF NATURAL SCIENCES 57 



sented only by their impressions in half of a small concretion. Two 

 or three less complete ventral armors of Dinichthys are also known; 

 first, that described by von Koenen^^ in 1895, in which the left half of 

 the plastron is preserved; second, a much more complete armor con- 

 sisting of an antero-ventrolateral and both postero-ventrolaterals, 

 associated with other remains of a single specimen of Dinichthys 

 curius, described by Hussakof in 1905.^* 



The present specimen was discussed at some length in 1897 by 

 Eastman/^ who tried to determine the species from a consideration 

 of the known relations in size between the mandible and the ventral 

 plates in some dinichthyids. He conlcuded that the specimen might 

 provisionally be referred to D. newberryi. But it seems to us, after a 

 careful study of both the specimen and this discussion, that the matter 

 is altogether too hypothetical for a definite conclusion, and that there 

 is no use in this fine-spun reasoning since it cannot lead to anything 

 positive. For our part, we are content to describe the specimen 

 as a plastron of Dinichthys, without referring it to any one species. 

 If D. newberryi were the only species of Dinichthys in the Portage 

 shales, we would be justified in thinking that the present plastron 

 belonged to it; but since there are one or two other species, as shown 

 by fragmentary remains, there seems little justification for associating 

 the plastron with one of these rather than with another. The dis- 

 covery in the Portage of ventral plates in association with other 

 remains may, at any time, upset any mere theoretical conclusions 

 on this point. 



Figure 16, represents the plastron as it appears in the slab, and 

 figure 17, is a restoration of it. On comparing the two figures it will 

 be seen that there is very little that is conjectural in our restoration. 

 It is, in fact, hardly more than an outline drawing of the plates shown 

 in the specimen, with here and there a missing portion of the outline 

 added. 



On comparing our restoration with that given by Eastman^^ of the 

 same specimen it will be seen that there is considerable difference 

 between the two. In our restoration the postero-ventrolaterals are 

 broader and shorter, and united in the median line by small flanges 



22 vonKoenen, A.: Ueber einigeFischreste desnorddeutschen und bohmischen Devons . . . 

 Abhandl. k. Gesell. Wissen. Gottingen, xl, 1-37, pis. i-v. 



2* Hussakof , L.: On the structure of two imperfectly known Dinichthyids. Bull. Amer. Mus. 

 Nat. Hist., xxi, 400-414, pis. xv-xvii and 2 figs. 



2* Eastman, C. R.: On the relations of certain plates in the Dinichthyids, with descriptions of 

 new species. Bull. Mus. Comp. Zool., xxxi, 19-44, pis. i-v. 



2*1x10. cit., pi. i, fig. 2. 



