NO. 1787. SCT^NOID FTSHES OF JAPAN— JORDAN AND THOMPSON. 245 



Body oblong, more or less elevated and compressed, the dorsal 

 and ventral outlines unlike; mouth low, horizontal or oblique, often 

 with conspicuous slits and pores; lower jaw usually included; teeth 

 in two or more series, the upper series above enlarged, and sometimes 

 the inner below; no true canines; preopercle entire, crenulate or with 

 a few small bony serree near the angle; no barbels; gill rakers more 

 or less shortened, relatively few, and often thick; lower pharyngeals 

 moderate, separate, with bluntish teeth. Dorsal fins well separated; 

 soft dorsal long; anal short; caudal fin usually rhomboidal; second 

 anal spine moderate or strong; pseudobranchise present, air bladder 

 large and complicated; scales moderate, normal. 



As here understood, a very large genus, mostly of the Old World, 

 comprising a great variety of forms, which differ widely among 

 themselves, but which form an almost continuous series from Pseu- 

 dotolithus (related to Corvula and Bairdiella) on the one extreme to 

 Cheilotrema at the other. 



In Pseudotolithus and Nibea the preopercle has bony serrations, as 

 in Micropogon, and the gill rakers are relatively long and numerous, 

 as in Corvula. In Pseudotolithus the lower jaw projects. In Argy- 

 rosomus (Pseudoscisena) the lower teeth are few rowed and unequal, 

 while the slits and pores about the mouth are little developed, and 

 the soft dorsal is scaleless. 



In the other subgenera the lower teeth are subequal, in bands, and 

 the mouth is smaller, its pores and slits more developed, and the gill 

 rakers are shorter. 



In Jolmius (including Bola) the body is elongate and the soft 

 dorsal scaly. In Scixna ( = Corvina) the body is deep and the soft 

 dorsal naked. In Cheilotrema (including Rhinoscion) the body is very 

 deep and the soft dorsal scaly. Callaus has the general traits of Argy- 

 rosomus, but with the slits and pores more developed, the gill rakers 

 slender and short, the caudal lunate. In most of the others the cau- 

 dal is rhomboidal or irregularly S-shaped. It is possible that Argy- 

 rosomus (including all with the lower teeth unequal) may be set off 

 from the other forms as a distinct genus as in Bleeker's system. In 

 any event, the many species form an almost continuous series from 

 one extreme to the other. The interrelations of these forms have 

 been fully discussed by Jordan and Eigenmann." In the same paper 

 the reasons are given why Scisena should take the place of Corvina, 

 the two having the same type, Scixna umhra Linnaeus, by the 

 restriction of Cuvier in 1817. Scixna umbra of Linnaeus is a com- 

 plex species, but the diagnosis copied from Artedi is based on 

 the "corb," a fact which justifies Cuvier's first restriction, for which 

 his later view could not properly be substituted. Still more impor- 

 tant is the fact that Scixna umbra of Hasselquist,'' from whose long 



oBull. U. S. Fish Comm. for 1886, 1889, p. 395. &Iter Palestinum, 1757, p. 352. 



