46 NEW YORK STATE MUSEUM 



in contact mesially, and are notched at the anteroexternal instead of at the 

 posteroexternal angles. The greater part of the exposed surface of each 

 plate is feebly rugose and marked by a sharply bent sensory canal, thus 

 recalling well known Arthrodire conditions, and confirming their interpreta- 

 tion as maxillae." In the majority of specimens, these plates, and also the 

 much more delicate pair of mandibular elements, are more or less displaced, 

 but this circumstance alone furnishes no reason for supposing that the two 

 pairs did not work directly against each other, or that the right and left 

 halves of either pair were capable of independent motion, as in Arthropods. 

 Neither are we compelled by the form of the mouth parts to make any 

 anomalous suppositions as to their movements or relations to one another, 

 although it is evident that they are merely dermal ossifications, and there- 

 fore of different nature from the jaws of higher vertebrates. 



The pattern of the cephalic sensory canal system is considerably dif- 

 ferent from that in Asterolepis and Pterichthys. No transverse commissure 

 unites the lateral canal of each side across the occipital plates; but in front, 

 just at its incurved flexure on the lateral plate, a conspicuous branch is 

 given off, which runs forwards and outwards to the margin of the shield, 

 being very likely continuous with that on the maxillary plates on the lower 

 side of the head shield. On the median occipital two fainter canals are 

 observable, forming an angle with each other but not always uniting 

 behind, and extending outwards and forwards over the lateral plates until 

 they become confluent, or almost confluent, with the lateral canal near its 

 central point of flexure. Near the same point, on the inner side of the 

 main groove, a small ear-shaped mark is often, but not always to be seen. 

 The main groove is continued nearly parallel with the periphery until it 

 reaches the center of the premedian plate, where it is sometimes interrupted 

 for a short distance. 



The articulations of the body plates are as in Pterichthys, the anterior 

 median dorsal [<?. in. d. \ overlapping the antero-dorsolateral \a. d. I. ] of 



'Professor Patten, however, insists that these pLates are mandibiUar, and that the 

 smaller, S-shaped pair are maxillary [Av. cit. p. ii8]. 



