522 CENTRISCID.E. 



h-n. Half-grown. China Seas. Presented by Vice -Admiral Sir E. 



Belcher, C.B. 

 0. Adult. 

 p, q. Adult : stuffed. 



Description. — This species does not differ in any essential point 

 from C. scolopax, so that we can dispense with a lengthened descrip- 

 tion ; it will be readily distinguished by its body, which is more 

 elongate than in the other species, by its much shorter spine, and, 

 finally, by conspicuously smaller scales, which advance nearly to the 

 middle of the length of the rostral tube. 



The length of the head is somewhat less than its distance from the 

 caudal fin ; the skin which covers the iris is scaly nearly in its entire 

 circuit ; the margins of the orbit are smooth. The upper and lf)wer 

 profiles of the body are very little convex. The spinous dorsal fin 

 commences much nearer to the occiput than to the base of the caudal 

 fin ; the length of its second dorsal spine varies a little in different 

 individuals : — 



In a specimen 00 lines long (Japan) it is 6i lines long. 



„ „ 57 ,, (?Mediterranean) „ 7 



52 „ (Japan) „ 51 



„ ,, 41 „ (Sierra Leone) . . ,, 5^ 

 „ ,, 39 ,, (Sierra Leone) . . ,, 5^ 

 „ „ 41 „ (Madeira) „ ^ 



>» >> "" >> >j 4^ 



„ „ 23 „ (Mediterranean) „ 3^^ 



„ ,, 18 „ (Mediterranean) „ 2i 



„ ,> 17 „ (China) „ 2| 



The distance of the soft dorsal fin from the caudal is more than 

 the length of its base. The anal commences immediately behind 

 the vent, or behind the vertical from the fourth dorsal spine. Caudal 

 emarginate. Back reddish or greenish, sides and belly silvery. 



These characters will sufficiently prove the distinctness of this 

 species from C. scolopax. Its occun'ence in the Atlantic and in 

 the Japanese seas is a new proof of the similarity of a part of the 

 fauna of the latter to that of Southern Europe. The specimens from 

 both seas are so similar, that we consider them as one and the 

 same species ; as, however, there exists a trifling difference in the 

 length of the dorsal spine, it being a little shorter in the Japanese 

 specimens, this character will be taken advantage of to separate 

 them specifically. We propose, to those who are inclined to do so, 

 the name of Centriscus japonicus. 



3. Centriscus humerosus. 



Richards. Voy. Ereb. 8f- Terr. Fishes, p. 50. pi. 34. figs. 5, 6. 



D. 7 I 15. A. 17. C. 4-1-4-1-5 + 5. 



The height of the body is somewhat less than the distance of the 

 operculum from the base of the caudal. The second dorsal spine is 



