240 ORDER ACANTHOPTERYGII. 



seen in the Moluccas, where they inhabit, creeping 

 and leaping over the mud, either to escape from their 

 enemies, or to seize upon the small shrimps, which 

 constitute their chief food. 



Some of them have the concave, disc-like ventrals 

 of the true gobies '. 



The ventrals of others are divided nearly to the 

 base 2 . 



I would also separate the 



1 Gobius Schlosseri, Pall. Spic. viii. pi. i. f. 1 4 ; to which must 

 be added the Gob. striatus, Schn. xvi., left among the gobies, though 

 it is hard to say why, since it is a true Periophthalmus. 



2 Gobius Kcelreuteri, Pall. Spic. viii. pi. xi. f. 13. Per. ruber, 

 Schn. Per. papilio, Schn. pi. xxv. 



N.B. Both the gobies and the periophthalmi, with divided ventrals, 

 according to the system of M. de Lacepede, would be Gobiomores ; 

 if, together with this division of the ventrals, they had but one dorsal, 

 they would be Gobiomoroides ; but the species arranged under these 

 two jjenera have not all their characters. The Gob. Gronovii, Gra., 

 Marcgr. 153, does not belong to this family, it is our genus Nomeus, 

 of the family of the Scomberoides. The Gobiomoroide pison, Gob. 

 pisonis, Gra., Amore pixuma, Marcgr. 166, Eleotris, 1. Gron. Mus. 

 16, has not the character of this genus, for it has two dorsals, both 

 in the figure of Marcgr., and in the description of Gronovius, and 

 by its ventrals is an eleotris. 



Bloch., Ed., Schn., p. 65, separates from the gobies, and forms 

 into the genus eleotris, different from that of Gronovius which bears 

 the same name, those species whose ventrals are merely united like a 

 fan, without being infundibuliform ; but in those which I have ex- 

 amined, the membrane which unites the external edges in front, is 

 merely somewhat shorter in proportion, which has prevented it from 

 being observed, and for this; reason I leave them among the gobies. 



