48 C. R. Eastman — Prelimmary Note on the Relations 



relief. As seen from the ventral aspect, the form is suggestive 

 of the nasal plate in Dinichthys^ owing to the semicircular 

 compression of the border around the head of the T. This 

 plate was overlapped for three-fourths of its length by the 

 anterior ventral s, as is also the case in Coccosteiis. 



Posterior dor so-lateral plate. — This element of the dermal 

 skeleton is represented in the collection by three perfect speci- 

 mens pertaining to the left side, and a fragmentary one from 

 the right side. As already remarked, portions of this plate 

 have long been known, J^ewberry having mentioned it as 

 early as 1875."^ The greater portion of two such plates is 

 figured by the same author in his Monographf under the desig- 

 nation of '' post-clavicular (?) plates." In the body of the 

 work (p. 142) this element is described as a " triangular bone 

 with which I have long been familiar, but I am not yet able 

 to '^x with certainty its position on the body." 



[N^ewberry suggested two places where this bone might have 

 fitted on : one behind and overlapped by the suborbital plate, 

 the other somewhere between the median dorsal and the 

 antero-dorso-lateral (" suprascapula "). This latter position he 

 regarded as the more probable, although hesitating to express 

 himself decidedly on this point. If the true relations of the 

 plate had been clear in J^ewberry's mind, he would have des- 

 ignated it by its right name in the letterpress, and would not 

 have styled it the " post-clavicular " in the legends for 

 Plate VI. 



A study of the material in the Museum of Comparative 

 Zoology has convinced the writer of several things, as follows : 

 The specimens figured by Newberry are incomplete, over 

 20^°" being wanting from either end ; the " free margin " is not 

 straight, but gently arcuate from end to end, and is indented 

 for a certain portion of its length by a contiguous plate (pos- 

 tero-lateral) ; the surface is not plane but warped, and on the 

 under side it is thrown into a few parallel undulations, the 

 axes of which are transverse with respect to the main axis of 

 the body when the plate is oriented in its natural position. 



The plate in question is the postero-dorso-lateral. New- 

 berry's figures 2 and 2a of Plate YI. are rights and lefts 

 respectively;:]: to bring them into their natural relationships it 

 is necessary to rotate them about their centers until their 

 upper (antwior) extremities are inclined toward one another 

 at an angle of about sixty degrees, and then separate them at a 

 distance equal to the width of the median dorsal shield. In 

 this position it will be noticed that the regular curve of the 



* Palseontologj of Ohio, vol. ii, p. 32. 



f op. cit, Plate YI., Figures 2, 2a. 



X These relations are reversed by Newberry ; hence the figures must be inter* 

 changed in position after they have been rotated so as to include the angle above 

 specified between their free margins, which are here represented as parallel. 



