496 The American Naturalist. [June, 
is patent from the specimen that they are only the posterior 
cranial angles, produced in the manner already described. 
One of them is preserved in its natural position, and the other, 
that belonging to the right side, has become dismembered and 
lost. The supposed “articular glenoid cavity, possibly for the 
condyle of a mandible,” which is stated to be “one-half in the 
cranium and one-half in the nuchal element,” pertains entirely 
to the head shield, and abuts directly against the cranial roof. 
Turning now to the osteology of the back of the head, very 
peculiar conditions are encountered. The boundaries of the 
Fig. 5. Macropetalichthys sullivanti Newb. X 4. 
large central plate are readily determinable, and are about the 
same as Cope has shown for M. rapheidolabis.® Epiotic and 
marginal presumably constitute the postero-lateral angles. 
But superimposed upon the dorsal plate and hinder part of 
the cranium is a system of tuberculated derm plates which are 
arranged independently of the bones beneath. There is first 
of all a median superficial element which covers the same 
space as the central, but in addition to this ‘it preserves its 
® Loc. cit., pl. xxix, fig. 4. 
