EASTMAN : MYLOSTOMID DENTITION. 213 



M. variabile, and also founded a second species, M. terrelli, upon the evi- 

 dence of a solitary mandibular plate. He noted the general correspon- 

 dence between the Mylostomid and Dinichthyid type of mandible, and 

 observed that the former occurred together with two distinct varieties of 

 "flattened tabular dental plates . . . exhibiting the same microscopic 

 structure," which were properly referred to the upper dentition of the 

 same species. He made no effectual attempt, however, to work out the 

 arrangement of the pavement teeth, merely observing that their " sides 

 are straight or bevelled, apparently for co-adaptation, and by this char- 

 acter favor the conclusion that the dentition consisted of many pairs of 

 plates, constituting a tesselated pavement ; the crowns of the teeth 

 below being convex, those above concave." 1 The front margin of the 

 mandibular plates was also considered by this author to show evidence 

 of co-adaptation with other dental structures ; and in order to satisfy the 

 hypothetical requirements thus created, a pair of " premandibular " ele- 

 ments was not only postulated by him, but two specimens figured in his 

 monograph 2 were actually referred to this position, albeit with some 

 reservation. We will return later on to a discussion of these so-called 

 " premandibular " plates, under a separate heading. 



Newberry's investigations of Mylostoma served to acquaint us in all, 

 as he supposed, with four pairs of dental structure, one of which was 

 correctly identified as belonging to the lower, and two to the upper jaw, 

 while the position of the fourth was acknowledged to be uncertain. His 

 reasons for referring these various pairs to a single species are that they 

 were found to occur together, and to exhibit identical structure and 

 surface markings. The circumstances of their discovery are not related 

 in detail, but it is significant that all specimens of Mylostoma known to 

 Newberry, excepting the type of M. terrelli, were obtained by one 

 collector from a single horizon and locality, namely, the Cleveland shale 

 of Sheffield, Ohio. Newberry's suggestion that several pairs of plates 

 besides these four took part in the complete dentition, and that the 

 upper series formed a tesselated pavement, shows that he had only a 

 vague and illusory idea of their arrangement. He even confused the 

 right and left mandibular plates. His work was essentially that of a 

 pioneer, and as such is praiseworthy, although necessarily imperfect. 

 It was at Dr. Theodore Gill's suggestion that he undertook the first 

 comparisons between the dentition of Arthrodires and Dipnoans, the 



1 Newberry, J. S., The Palaeozoic Fishes of North America. Monogr. U. S. G. S., 

 16 (1889), p. 166. 



2 Ibid., p. 161, 165, Plate 16, Fig. 4. 



