THE DEYONIAN EOCKS OF CANADA. 83 



obscure impressed Hue or narrow groove extends along the median line to about the 

 midlength, where it bifurcates, after which each branch curves outward and forward 

 for a short distance, then slightly inward, until both become too indistinct to be traced 

 any farther. 



The parietal plate is doubly flexuous, or doubly and faintly convex with a shallow 

 sinus between each convexity, on each of its lateral margins, and widens rapidly back- 

 ward into a broadly truncated but slightly siniious base, whose breadth is much greater 

 than the length of the plate as measured along the median line. 



Immediately behind the cranial buckler there are three small plates, which evidently 

 correspond to the supratemporals in Dr. Traquair's diagrammatic representation of Tristi- 

 chopterus and to the occipitals of Osleolepis as restored by Pander. Unfortunately, these 

 plates are preserved only in a single specimen of Etisthenopteron (No. 4), and in it they 

 are displaced in such a way as to partially overlap one another and so much broken at 

 the edges that their exact shape cannot be ascertained. 



The orbit is placed very far forward, and in one specimen (No. 4), on and partly 

 around the upper margin of the left eye, there are indications of what seem to have been 

 five rectangular, or somewhat wedge-shaped and closely contiguous, thin, radiating 

 plates, which are higher than broad. These jDlates are probably homologous with the 

 circumorbitals of Traquair's restoration of Dapedius, as well as suggestive, in a general 

 way, of the still more highly specialized sclerotic plates in the eye of Ichthyosaurus and 

 Megalosaurus. In this connection it may be well to mention that in a head of Pliane- 

 ropleuron curtum, collected by Mr. Foord in 1881, the space originally occupied by the 

 pupil of the left eye is completely surrounded by a circle of similar impressions, a cir- 

 cumstance which seems to shew that in this species there was a continuous series or ring 

 of circvimorbitals. This specimen, which was overlooked when the description of 

 P. curium in the first part of the present paper was written, will be found described more 

 in detail on pages 91, 92, and figured on Plate X. 



In Eusthenopteron, as in Osf.eolepis and Tristichopterus, there appear to have been three 

 suborbital plates around each eye, but only two of these are distinctly shewn in the 

 specimens of E. Foordi in the Survey collection. 



The exact shape of the inferior suborbital is clearly defined in specimen No. 3, and 

 in another Fragment not specially enumerated, while its relative position is well exhibited 

 in specimens Nos. 1 and 4. In the former it appears as a narrowly elongated plate, 

 whose maximum length is more than three times its greatest height or breadth. It is 

 clearly the largest of the three suborbitals and extends as far forwards beyond the orbit 

 as to the anterior termination of the maxilla, and beyond the orbit behind, to about two- 

 thirds of the entire length of the maxilla. Its upper margin is concavely excavated a 

 little in advance of the midlength, apparently to form part of the orbit ; its lower margin 

 is long, straight and sutnrally connected with the upper margin of the maxilla ; while 

 its narrow anterior and somewhat broader posterior extremities are both obliquely 

 subtruncated and widen outward. 



The posterior suborbital, whose shape and position are best seen in specimens Nos. 

 1, 3 and 4, is also longer than broad, and its front margin would seem to have been more 

 or less truncated or excavated to form part of the orbit, though this part of the plate is 

 unfortunately either imperfect or crushed out of shape in the very few specimens in 



