Eastman: Triassic Fishes. 145 



lower down than where we might suppose it to he ahmg what appears 

 to be the baek; that is, the median dorsal line of tlu' fish does not 

 coincide with the uppermost margin of the fossil, as is proved by the 

 row of dorsal ridge-scales which may be traced continuously in these 

 specimens, but might readily be obliterated in others. These examples 

 illustrate what may happen in Seminotus and other forms, where the 

 true dorsal and ventral contours are often difficult of recognition. 



Other characters which have been relied ui)on for distinguishing the 

 deep-bodied and slender-bodied species, or varieties as we prefer to 

 regard them, are the position of the median vertical fins and certain 

 details of the squamation. It has been claimed, for instance, that in 

 C. gracilis the "dorsal and anal fins are subeciual in size and almost 

 completely opposed"; whereas in C. rcdfichli the dorsal fin is said 

 to arise "opposite to the middle of the anal." This distinction, 

 however, will not hold. Intergradations occur, and as a matter of 

 fact, the alleged distinction applied in the reverse sense to one of the 

 original cotypes of C. gracilis, in which dorsal and anal are not "almost 

 completely opposed," liut the former arises o])positc the middle of the 

 latter. 



As for supposed differences in scale-characters, it has been asserted 

 in the definition of the type-species that "the scales are smooth, none 

 deeper than broad, tliose of the flank in the abdominal region very 

 finely serrated." In the variety which has been called C. redfieldi 

 the amended diagnosis reads: "Scales mostly smooth, but sometimes 

 in part longitudinally striated, the striae terminating in the coarse 

 serrations of the posterior border which characterize the principal 

 flank-scales; many of the flank-scales deeper than broad." 



E.xamination also shows that in respect to such characters no rigid 

 distinction can be made between the type-species, C. gracilis, and the 

 deep-bodied variety, which has been commonly recognized as belonging 

 to a separate species. The condition of smoothness in scale characters 

 is a variable one, and appears to be due largely to differences in age, 

 amount of wear, state of preservation, and in some cases to the opera- 

 tion of chemical agencies which have eroded the external surface or 

 covered it with a thin glaze. Differences in age and wear will also 

 account for degrees in coarseness or fineness of the striations along 

 the posterior margin of the flank-scales. To sum up, therefore, a valid 

 specific distinction between C. gracilis and C. redfieldi can scarcely 

 be maintained, but on grounds of convenience it may be well to retain 



