38 Bulletin American Museum of Natural History. [Vol. XIX, 



(PI. II, mx.), and these smooth surfaces must have been 

 applied to the premaxillas, unless the structure of this region 

 was entirely different from that of A. favirostris. The pre- 

 maxillae have quite certainly been similar to those of Loomis's 

 Thryptodus zitteli, and when present have nearly concealed 

 the ethmoid. On the other hand, Loomis is probably mistaken 

 in thinking that the premaxillae of his species have coalesced 

 with the ethmoid. His specimen and the type of A. aratus 

 explain each other, and the explanation is confirmed by the 

 types of A. favirostris. Whether or not the premaxillae of 

 A. aratus were coossified, we cannot be sure, No evidences 

 of a median suture appear. Nor is it absolutely certain that 

 the ascending processes of those bones were not united in A . 

 favirostris. 



The nasal cavities are situated above the level of the eyes 

 and a little in front of them. 



The bones of the upper surface of the head (Fig. 23) are in 

 general like those figured by Loomis in his Thryptodus zitteli. 

 The supraoccipital {s. oc.) does not separate the small and 

 square parietals {pa.). Laterad of the last named bones are 

 the large squamosals {sq.). The eye has been surrounded 

 by a complete ring of bones. Over the eye are two sculptured 

 supraorbitals (Fig. 23, 5. or.). Behind the eye is the smooth 

 postorbital {pt. or.); while in front is the prefrontal (prf.). 

 This bone is sculptured above, but it sends downward in front 

 of the orbit a long smooth process. Another elongated bone, 

 apparently the preorbital, occupies the area between the pre- 

 frontal and the maxilla. The number of the bones covering 

 the cheeks cannot be determined. They extend to the pre- 

 opercular. The preoperculum (p. op.) and operculum (op.) 

 are as represented in the plate. The other opercular bones 

 were doubtless present, but are not preserved. The post- 

 temporal (pt.) and the supracleithrum (su. cl.) are present. 

 Of the cleithrum (cl.) only fragments remain on the block. 

 An imprint of the coracoid is seen on the matrix below the 

 throat. The articulation of the pectoral fin is high, being 

 just below the vertebral column. If the precoracoid was 

 present, it must have been short. 



