Cretaceous Fishes from Mount Lebanon. 407 



number, with moderately robust arches. Behind the chavi- 

 cular arch there are traces of smootli, expanded, postchivicuhir 

 ph^tes, while the pectoral fin, which exhibits about 18 rajs, is 

 truncated distallj, and when adpressed would probably reach 

 the insertion of the pelvic pair. The pelvic fin preserved is 

 about two thirds as large as the pectoral, and shows 8 stout 

 rays, all divided distally. The dorsal fin arises considerably 

 further from the caudal fin than from the occiput, opposite the 

 origin of the pelvic pair. Its two foremost rays are slender 

 spines, the second longer than the first ; the third ray is still 

 longer and articulated, though not divided, distally ; the 

 following rays, which are at least ten in number, but too 

 crowded for precise counting, are all both divided and articu- 

 lated distally, and gradually decrease in length backwards. 

 The anal fin is shown to be two thirds as elevated as the 

 dorsal, and about 16 supports can be counted at its base. 

 The caudal fin is very deeply forked. The squamation 

 seems to have been uniform, all the scales cycloidal and 

 deeply overlapping. 



An example of the same species in the British Museum 

 (no. 48155 />), which is considerably elongated by distortion, 

 is important as having the mouth widely open, and thus 

 displaying the jaws. Nearly the wdiole of the upper border 

 of the gape is shown to be formed by the stout arched 

 maxilla, which bears a single close series of minute conical 

 teeth, and above this bone there are two large supramaxillaries, 

 as in the herring. Impressions of teeth like those of the 

 maxilla are also seen on the border of the dentary bone. 



A third specimen, apparently of the same fish, in the 

 Edinburgh Museum, is described and figured (lac. cit. p. 567, 

 pi. xxxii. fig. 5) as the type of Sardinius crassapinna, Davis. 

 A direct comparison of this fish with the original of fig. 4 

 seems to leave no doubt that their differences are due solely 

 to accidents in preservation and the mode of crushing. The 

 so-called S. crassapiuna is evidently much shortened, while 

 the specimen now under consideration is shown to be some- 

 what elongated by distortion. They may thus be placed in 

 one and the same species. 



Assuming that this determination of the specific identity of 

 the three specimens just meritioned is correct, it becomes 

 clear that although the jaws of Sardmius crassapinna re- 

 semble those of the typical Osmeroides^ the fish is distinct 

 from the latter genus at least in the slenderness of its 

 branchiostegal rays, the comparatively small number of its 

 abdominal vertebrae, and the relatively large size of its paired 

 fins. It is indeed in all these respects generically identical 



28* 



