572 PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 



tator by Block & Schneider =:P/il7^/29>^^^s• donnitator (Lac.) C. & Y.) aud G. 

 Imireuteri {=P€rioplitlialmus Imlreuteri (Gmoliti) Blocli & Schneider). 



Of these species, the first, (jronovii has no rehition to Gohius, and 

 does not correspond to the definition of Gobiomorus, as the gill mem- 

 branes are free from the istlimns. Its association with the Gobies is an 

 error which originated with (Imelin. It may therefore be omitted from 

 consideration. The remaining aiodern genera included in GoMomorus, 

 viz, Eleotris Bioch & Schneider, 1801 (Subgenus Valenciennea Bleeker, 

 ISoG), FMhjpnus Guv. & Yal., 1837, and Periophihalntus Bloch & Schnei- 

 der, 1801, are all subsequent to Gobiomorus, and in place of one of them 

 the latter name must be retained. It has not as yet been restricted by 

 any author, so far as we know. It seems to us best to consider as the 

 type of Gobiomorus^ G. dormitafor LaCepede, and therefore to use the 

 name Gobiomorus instead of Philt/pnus. A serions practical objection 

 to the consideration of taiboa {sfriyatus) as the type of Gobiomorus lies in 

 the uncertainty whether this species is really congeneric with Eleotris 

 gijrinns, (which species must, we think, as ^'■Eleotris ^nsonis," be consid- 

 ered the type of Eleotris). In Bleeker's system, strigatus is made the 

 type of a distinct genus ( Valenciennea Bleeker) and j)laced at a distance 

 from Eleotris, but no diagnostic features of importance have been made 

 known by which it may be distinguished. 



3. GOBIOMOROIDES. 



The genus Gobiomoroides was proposed by La Cepede (Hist. Xat. 

 Poiss., ii., 592, 1800)' with a definition identical \\\th. that of Gobiomorus 

 except that "uue seule nageoire dorsale" is substituted for "deux 

 nageoires dorsales." Its type is Gobiomoroides piso La Cepede, a 

 species which is considered by La Cqpede identical with Gobius pisonis 

 Gmelin, the '■'■Eleotris''^ of Gronow. 



Gobius pisonis Gmelin is identified by Cuvier & Valenciennes with 

 Eleotris (/ijrinus, with considerable doubt, however, as the descriptions 

 and figures of the former species are both incomplete aud erroneous. 

 The identity is probably too doubtful to w^arrant the use of the specific 

 nnme pisonis for (jyrinus. La Cepede's description of G. piso, is, however, 

 nottakenfromGmelin,butfromadriedfish " given by Bollaud to France.'' 

 This.j>pecimen has 45 rays in the dorsal which is continuous, 23 in the 

 anal, and the lower jaw has a series of canines besides the cardiform 

 baud. Whatever this fish may be, it is not an Eleotris, and the name 

 Gobiomoroides cannot be used for Eleotris gyrinus, even if it be shown 

 that this species is identical with Gobius pisonis Gmelin. 



4. KYPHOSUS. 



The genera Kypliosus (La Cepede, Hist. Nat. Poiss., iii, 114, 1802), 

 Pimelepterus (1. c. iv., 429, 1803) : Dorsuarius (1. c. v., 482, 1803), and 

 Xyster (1. c. v. 484, 1803), are identical, as has been shown by Cuvier 



