158 



bean found on the coast of North America, as far north as New 

 York. 



The great protractility of the mouth made the illustrious Cuvier 

 place it in a family he had called Menidce ; but in my paper on 

 the fishes of the Cape of G-ood Hope, written in 1861, I protested 

 against this character being considered as of first importance, as 

 it is to be found in sorts of almost all famihes. 



The species I describe here partakes of the characters of 

 Ditrema, by the large number of the anal rays, which in Gerres 

 are not generally more than eight or nine. 



GERRES MELBOTJRNEKSIS. 



The height of the body is contained two and a-half times in the 

 total length, without the caudal ; the eye is very large, and only 

 contained two and one- third times in the length of the head, when 

 the mouth is not extended ; when it is, the distance from the 

 extremity of the snout to the anterior margin of the eye is a little 

 less than the diameter of the orbit. The teeth are numerous and 

 villiform ; there are none on the vomer or palatine. The profile 

 is very convex over the eye and the snout. When the mouth is 

 not extended, it is rather bent downwards. The head is con- 

 tained, in this state, four times in the total length, including 

 the caudal ; the operculum is entire, as is also the pra3operculum, 

 and the praeorbital. The scales are large, being only thirty- 

 seven or thirty-eight on the lateral line ; this line is marked by a 

 succession of short, elevated, oblique lines, and is strongly curved 

 near the head. The spiny portion of the dorsal is much shorter 

 than the soft one ; it is formed of nine rather feeble spines ; the 

 first one is shorter than the following, which are about equal, the 

 posterior ones being something longer than the others. The rays 

 number sixteen ; the first are about of the same length as the 

 last spines, but they become rather longer as they go backwards. 

 The caudal is forked ; it is formed of eighteen long rays and of 

 several shorter ones on each side ; anal with three spines, rather 

 stronger than those of the dorsal ; the first is the shortest, and 

 the third the longest; the rays number seventeen, the first of which 

 is longer than the third spine, and the others decrease slightly 

 towards the tail. The ventrals are inserted a little behind tha 

 pectorals ; they are formed of a long slender spine and of five 



