84 Bulletm American Museum of Natural History. [Vol. XIX, 



What is the value of these species? K. contracta is said to 

 be characterized by the flatness of the "maxillary bone," by 

 which is meant the palatine; but in the types the palatine 

 of E. merrillii is still flatter, especially at the anterior end. 

 To what extent the flatness of this bone in both species is 

 due to its original form and to what extent to distortion dur- 

 ing fossilization is hard to determine. There was probably 

 considerable individual variation among these fishes. As re- 

 gards the sizes of the teeth in different parts of the palatine, 

 the writer has found, in looking through the collection of 

 this Museum, so much variation that he regards the dis- 

 tinctions specified by Cope as of little or no value. Consider- 

 able differences are often to be observed on the opposite sides 

 of the head. Cope also found characters in the "tongue- 

 shaped pharyngeal bone," which is really the vomer; but 

 after diligent comparison on the part of the present writer 

 the attempt to find specific characters in this part was 

 abandoned. 



Prof. Cope has given extended and accurate descriptions 

 of E. nepaholica and E. semianceps ; but when we seek for 

 the characters by which they may be distinguished from 

 each other, they are found to be rare and elusive. The 

 character which is most definitely stated is found in the 

 structure of the mandibular teeth. In E. nepaholica these 

 are said to possess no cutting edge on the posterior face. 

 Unfortunately the crowns of the teeth are nearly always 

 broken off, so that it is impossible to apply this test. The 

 types themselves cannot, at least now, be distinguished 

 in this respect. In another specimen I find a mandibular 

 tooth with a posterior edge, but there is no other indication that 

 it is not E. nepaholica. Both species possess such two-edged 

 teeth on the palatines. On the lower J9,w of a very large 

 specimen in the Museum I find that the teeth are two-edged ; 

 and observation makes it certain that if such teeth belong 

 only to E. semianceps this was not a smaller species than 

 E. nepaholica. Stewart says that the mandibular teeth of 

 specimens which he identified as E. nepaholica seemed to 

 have both anterior and posterior cutting edges. 



