III. COTTOIDS OF NOKTH AMERICA. 47 



are broad and proportionally much developed, for their tip reaches backwards, the 

 fourth ray of the second dorsal, and the first ray of the anal fin. The base of 

 insertion of these fins is crescent-shaped and close to the gill opening. In this 

 respect C. Alvordii differs considerably from C. meridionalis. The rays are fifteen 

 in number, all undivided, the eight lowermost alone overrunning the membrane of 

 the fin. 



Br. 16. D VII. 1C. A 13. C 3. I. 4. 4. I. 2. V I. 4. P 15. 



The lateral line, nearly straight, vanishes already under the seventh ray of the 

 second dorsal, whence the series of minute pores cannot be followed to the tail 

 without an eye-glass, when it is seen along the back as far as the posterior ray of 

 the second dorsal, and there disappears entirely. 



The anal opening is a little nearer the extremity of the jaws than the base of 

 the caudal fin. It is provided posteriorly with a tongue-shaped membrane, broad 

 at its origin, and very acute on its extremity, so that when seen in a profile view, 

 it has the appearance of an ordinary ray. Our figure, however, gives a clear idea 

 of it, as a solution of continuity exists between it and the first ray of the anal. A 

 close examination will likewise show a structural difference. 



The head and back are yellowish-brown, clouded on the sides. The belly and 

 fins are yellowish ; the first dorsal alone having two black spots on its posterior 

 half. 



This species inhabits the streams emptying into Lake Huron, in the vicinity of 

 Fort Gratiot, where it was collected in 1850, by Major B. Alvord, and presented 

 by this officer to the Smithsonian Institution, where the original specimen is pre- 

 served. 



VI. COTTCS MERIDIONALIS, GIRARD. -f Uv^U 



PLATE I. Figs. 9 and 10. Vfl-v. 



Sjn. Cottus meridionalis, GIRARD, Proc. Amer. Assoc. Adv. Sc. II., 1850, p. 410; and, Proc. Bost. Soc. 



*Nat. Hist. III., 1850, p. 189. 



Nothing is more dissimilar than the general outline of this species and that of 

 C. Bairdii, and yet the absolute length of both species is exactly the same on the 

 specimens figured. Indeed, the species which G. meridionalis resembles most in 

 shape, is C. Alvordii, and still there are differences which could not for a moment 

 be overlooked ; besides the fact of their remote habitat, which would likewise cause 

 doubts as to their identical relationship. It is not without a certain resemblance 

 with C. viscosu-s, and did it not belong to a different division of the genus by a 

 hidden character, its stout and contracted body would scarcely suffice to distin- 

 guish it from the latter. 



The greatest depth is contained but five times in the total length ; four times 

 only when we exclude the caudal fin, the length of which is equal to the depth. 

 The body tapers very rapidly away; its least depth, on the peduncle of the tail, 

 enters fifteen times in the total length. The greatest thickness is considerably less 



