264 EXPEDITION TO JAPAN. 



or purple, over nead and back, extending low down on the sides, and as far as end of soft 

 dorsal. Snout, breast, abdomen, and base of tail, a bright orange red, rather paler on the fins. 

 The soft dorsal and caudal shading off into yellow on the margin. A distinct bright ultrama- 

 rine blue stripe extends from the base of the first dorsal spine obliquely forwards under the eye, 

 and over the preopercle, narrowing as it descends, and terminating near the throat. 



22. GLYPHISODON VIOLACEUS, N. S. 



Notes. — From Lew Chew. (4 inches.) 



This seems to be a Glyphisodon, but the drawing is not distinct enough to determine its genus. 

 Its height is 2\ times in the length, regularly arched above and below, with subequal dorsal 

 and rather deeply forked caudal. Fins and scales rather large. Rays cannot be counted. It 

 is of a uniform dark dusky violet over the whole of the body and fins, darkest on the back ; 

 but without any bar or spot besides. Irides silvery. 



The G. melas of Kuhl and Van Hasselt, (Cuv. and Val. V, 472,) comes near it as far as can be 

 determined by the description. Their Pomacentrus littoralis seems also to agree with it in color, &c. 



Bleecker has described many new species of Pomacentri, and other allied genera, from Su- 

 matra, &c, but none of these appear to be identical with this species. 



Diengkitsch calls it Ya-ha-gi. 



23. GLYPfflSODON SMARAGDDJUS, N. S. 



PLATE VI, fig. 3. Life size. 



Notes. — From Lew Chew, (3J inches.) 



This species resembles the G. coelestinus of Solander, (Cuv. and Val. V, 464, pi. 135,) in 

 shape, and were it not for its decided green hue I should not hesitate to class it with that 

 species, which appears to be common in those seas. 



Its height is only twice in its length ; the dtirsal spines appear to be twelve in number, and 

 longer than in G. coelestinus, which has thirteen. The other fins as in that species. 



Color of a general verdigris green, paler beneath, and very dark on back and spinous dorsal. 

 Three or four bars of dark green across the sides, of which three are under the dorsal, and one 

 on the tail. Irides green also. 



24. ETROPLUS FUMOSTJS, N. S. 



PLATE VI, fig. 1. Life size. 



Notes. — From Lew Chew. (If inches.) 



This seems to be an Etroplus from its general form and great number of anal spines, which 

 were not counted. Its height is twice in its length ; outlines regularly arched. Dorsal and 



