lo University of Michigan 



with aberrant dentition; we have examined numerous typical 

 specimens from rather deep w-ater in the North Atlantic. 



Regan's record^^ of this species from Muscatine, Iowa, is 

 probably erroneous, although based on specimens of P. mari- 

 nns so labelled by Meek (some of which we have re-exam- 

 ined). It seems more likely that labels were transposed than 

 that marinus should have been taken in the Mississippi basin. 



4. Entosphenus spadicens Bean 



This species is sharply differentiated by its dentition, which 

 seems to be more generalized than that of other members of 

 the Entospheniis-lMiiipetra series. We make it the type of a 

 new" subgenus, Tctrapleiirodon (defined on p. 6). In addi- 

 tion to the types, we have examined specimens taken by 

 Duges at Tanganzicuaro, Mexico, and by Nelson in Lago de 

 Chapala. 



5. Entosphenus tridentatus tridentatus Richardson 



We find it necessary to divide this West American species 

 into a northern and a southern subspecies, as in the specimens 

 examined from the Columbia River system and northward 

 the myotomes between gill slits and anus vary from 68 to 74, 

 whereas in those from the Klamath basin and southward the 

 myotomes number 57 to 67. 



Petromyzon lividus Girard"^ and P. astori Girard-^ fall into 

 the synonymy of the typical subspecies of Entosphenus triden- 

 tatus as thus restricted. 



21 Ann. Mag. Nat. Hist, Ser. 8, Vol. 7, 191 1, p. ic 



22 Pac. R. R. Exp., Fish., 1858, p. 379. 



23 Pac. R. R. Exp., Fish., 1858, p. 380. 



