1886.] PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 509 



Habitat.— 'East coast of United States. Cape Cod to Florida. 



This little fish is generally common on our Atlantic coast, especially 

 southward in shallow grassy bays. The name bosci, first given it by 

 Lac^p^de, has priority over the commonly used alepidotum of Bloch & 

 Schneider. 



49. Gobiosoma multifasciatum. 



Gobiosoma multifasciatum Steindachner, Ichth. Beitrage, v. 183, 1870 (Lesser 



Antilles). 



Habitat. — West Indian fauna ; Lesser Antilles. 



We know this species only from the description of Dr. Steindachner, 

 who found it not uncommon about the Lesser Antilles. 



50. Gobiosoma zosterurum. 



Gobiosoma zosterurttm Jordan & Gilbert, Proc. U. S. Nat. Mus., 1881, 361 (Ma- 

 zatlan) ; Jordan & Gilbert, Bull. U. S. Fish Com., 1882, 108 (Mazatlan); 

 Jordan, Catalogue Fish. North America, 106, 1885 (name only). 



Habitat. — Gulf of California ; Mazatlan. 



This species is known only from the type taken by Professor Gilbert 

 at Mazatlan. The name zosterurum (belted tail) is intended to refer to 

 the dark stripe on the caudal fin. 



51. Gobiosoma longipinne. 



Gobiosoma longipinne Steindachner, Ichth. Beitrage, viii, 27, 1879 (Las Ani- 

 mas, Gulf of California). 



Habitat. — Gulf of California. 



This species is known to us only from Dr. Steindachner's description. 

 It differs widely from the foregoing species of the genus, and is prob- 

 ably an ally of the aberrant G. ios. 



52. Gobiosoma ios. 



Gobiosoma ios Jordan & Gilbert, Proc. U. S. Nat. Mus., 1882, 437 (Vancouver's 

 Island); Jordan & Gilbert, Syn. Fish. North America, 948, 1883 (Puget 

 Sound); Jordan, Cat. Fish. North America, 106, 1885 (name only). 



Habitat. — Waters about Puget Sound. 



The types of this species, two in number, were taken by Professors 

 Jordan and Gilbert in Saanich Arm, Vancouver's Island. They were 

 found in the stomach of a specimen of Hexagrammus asjaer, taken in 

 water of some depth. 



The species has evidently very little affinity with the type of Gobio- 

 soma, and when its skull is examined it may prove to be the tyx)6 of a 

 distinct genus, perhaps allied to Gillichthys. 



17. GILLICHTHYS. 



GiLLiCTHYS Cooper, Proc. Cal. Acad. Nat. Sci., 1863, 109 (miraiilis). 

 GiLLiA Giinther, Zoological Record, 1865 (name preoccupied), (mirahilia). 

 Saccostoma (Guichenot MSS.), Sauvage, Boll. Sci. Phiiom. Paris, 1882, 171 (name 

 preoccupied), {gulosum). 



Type Gillichthys mirdbilis Cooper. 



