KEPOKT ON l(;ilTriY()LO(}Y. 401 



appears to bo questionable. Dr. Girard, in liis nionograpli of tlie genus, puljlished 

 by the Smitlisoniau Institution, has assci-ted that when young some species of Coftiis 

 "exhibit tecth-hke asjierities on the pahitiues. This occurs chiefly amongst tliose having 

 four jointed rays to the ventrals: in C. Wilsonii, C. Bairilii, and (J. Meridlonalis. C. 

 (jracUis is the only one of the division with tliree jointed rays Avhere similar as])erities 

 have been noticed." I'liis assertion has not, however, been confirmed by my iiixesti- 

 gations. An examination of the types of the Cottoids described by Dr. Girard, in his 

 "Monograph", has demonsti'atcd that the presence or absence of teeth in the palatine 

 bones is constant in tlie various species. In the Coffus Richardsonii, C. Wilsonii, and 

 C 7»mr?io?«a/i6-, teeth are always found on the palatines, in the oldest as Avell as the 

 youngest individuals. The Cottits Bairdii cannot be at present tound; l)nt the same is 

 doubtless the ease with that species. Many other specimens preserved in the Smith- 

 sonian Museum exhibit the same constancy in their dentition. 



As to the Cottus gracilis, it is said by Dr. Ayres, in his Memoir on the Identity of 

 the Noi'th American Cotti with the Cottus ff oh i/) of Europe, that of the very numerous 

 specimens of the Connecticut Cottus (C. gracilis Heckel), which lie had examined, lie 

 had seen a single one in which there Avere a few scattered teeth on the palatines, like 

 those of the vomer ; in others, those bones were edentulous. It is probable that that 

 instance is alluded to by Dr. Qirard in liis mention of palatine teeth having been dis- 

 covered in the Cottus gracilis. An isolated fact like that recorded by an observer \\'lio 

 has failed to appreciate the distinctive characters of species of this group cannot, 

 however, be urged as a valid objection to the importance of such characters. Nor 

 could the circumstance that some Cotti have teeth when young, which they lose w^ith 

 affe, militate ag-ainst assiyniniy a certain value to a plan of dentition whi(;h is constant 

 through life, as well in the young and old. The difference of development alone 

 would be a character of importance. But there does not appear to be even such 

 dilference between the dentition of the young and old. In those specimens which 

 Dr. Girard described, the dentition is constant. Palatine teeth are even found in indi- 

 viduals which are much larger than any without. Such is the case with the species 

 now under consideration ; such is the case with other species equally large from the 

 Western States. 



If the above \aews ai-e correct, it would then appear to be advisable to separate 

 the Cotti with palatine teeth, and place them in another genus, or, at least, a subgenus, 

 to which the name of Potamocottus may be given. This grouj) will emljrace the Cottus 

 imnctulatus as its type, and, in addition, Cottus meridiomdis Girard, C. Bairdii Girard, 

 C. WilsQnii Girard, and C. Richardsonii of xVgassiz, as well as Cottopsis gulosus of 

 Girard. The genus Potamocottus would bear the same relation to Uranidea that Bnjttus 

 does to Pomotis, or Scorpccna to Scarpfcnopsis of Hoeckel. 



The genera Uranidea, Potamocottus, and Cottopsis agree very closely together, 

 l)oth in supei-ficial and anatomical characters, ami diflfer in the most decided manner 

 from Cottus and the related genera ; to express this divergence, the genera in (pies- 

 tiou should be segregated in a grouj) which may be named Uranidem. • 

 51 B u 



