NO. 1581. FISHES OF JAPAN— JORDAN AND RJCJIARJJSON. 657 



rather wider than in specimens of C. Jcuwu from AustraUa; two short 

 but strong preocular spines, a low postocular spine, with a very shal- 

 low cross-furrow behind it; nuchal spine not reaching to front of 

 dorsal; opercular spine barelj^^ crossing gill opening; humeral spine 

 about to vertical from fourth dorsal spine; second and third dorsal 

 spines subequal; posterior margin of spinous dorsal nearly straight; 

 pectoral to tenth ray of soft dorsal; upper pectoral appendage missing 

 tip of ventral by a distance equal to two-thirds diameter of eye; dor- 

 sal scutes moderate, their points not long and sharp. 



Color in life olivaceous, the deep lustrous blue of the inner side of 

 the pectoral fins very conspicuous ; preserved specimens have the back 

 (m adults) more or less coarsely mottled or blotched ; a plain line sepa- 

 rates the upper color from the paler whitish (silvery) of the belly; 

 young specimens have the back crossed with three or four broad, 

 obliquely disposed, and rather obscure cross-bands; spinous dorsal in 

 adults with some dusky on outer margin; soft dorsal with its outer 

 third dusky; caudal with a diffused dark blotch posteriorly; pectoral 

 inky blue inside, with a narrow pale outer margin, and with its inner 

 lower third, or entire lower half, darker in color than the rest of the fin 

 and marked with from 6 to 20 light-colored spots. 



(kumu, the native Maori name of this fish in New Zealand.) 



We are unable to separate this common Japanese fish from the 

 species Kumu, which is one of the best known food fishes of New 

 Zealand and Australia. 



Specimens of this species from Port Jackson (Sytlney) dift'er from 

 the Japanese form in having the snout less emarginate and its angles 

 less produced, scarcely extending over the retracted maxillaries; in 

 Port Jackson specimens the interorbital space is slightly narrower (7 

 to 7.10 in head) than in Japanese individuals; a single specimen 

 from Sydney (6} inches long) has the interorbital space as broad as in 

 some Japanese specimens (6.50 in head). The coloration of the pec- 

 toral fins varies in ' these specimens, the one just noted and the 

 small one from Port Jackson having only 8 to 12 spots, gathered in a 

 dark blotch on the lower inner third of the fin, as in the figure of 

 Lesson; while the larger Port Jackson specimen has the spots 18 or 20 

 in number and scattered over the low^er two-thirds of the fin. This 

 amount of variation in the markings of Australian specimens would 

 seem to break down the distinction between humu and sjnnosus as 

 made out by Gtinther (Cat., II, p. 204), following McClelland. If a 

 separate species or subspecies (spinosus) is to be made of the Japanese- 

 Chinese form it is evidently impossible to establish it in color mark- 

 ings. Nevertheless, we shall not be surprised if slight but perma- 

 nent differences in form should be made out, in winch case the 

 Japanese species will stand as Ohelidonichthys spinosus. 



