NO. 2149. JAPANESE MACROUROID FISHES— GILBERT AND HUBBS. 137 



more slender and the anterior edge rounded in those genera charac- 

 terized by the total absence of serrations. Two serrations were found 

 on the dorsal spine in a paratype of C oelorhynchus productus. 



The classifications of genera adopted by Jordan and Evermann/ 

 Goode and Bean,^ Dollo,'' and by other authors are based chiefly upon 

 that of Giinther, proposed in the report on the deep-sea fishes of the 

 Challenger expedition (p. 124). 



The characters used by us to separate the genera related to 

 Coryphaenoides (here referred to the subfamily Coryphaenoidinae) 

 are small, but their value lies in their constancy, which has been 

 tested in a large majority of the species. 



Many nomenclatural changes follow as a result of this rearrange- 

 ment of the genera. Dolloa {=3Ioseleya), Albatrossia, and Bogos- 

 lovius are regarded as synonyms of Nematonurus. Challnura can 

 not be separated from C ory phaenoides^ and the most of the numer- 

 ous species heretofore referred to Macrourus are now regarded as 

 species of Coryphaenoides on the one hand and of Llonurus on the 

 other. These changes are further indicated in the list of species 

 and in the generic descriptions. 



The number of branchiostegal rays is a character of great value 

 in this group, and is used in the present classification in the primary 

 subdivision of the subfamily Coryphaenoidinae. One group of 

 genera, undoubtedly related to C ory phaenoldes, has six branchios- 

 tegal rays. The species of these genera have the anus immediately 

 anterior to the anal fin, with the exception of Coryphaenoides hyo- 

 stomus, Abyss icola 'inacrochir, and several aberrant species of C oelor- 

 hynchus. Another large group, similarly interrelated, has seven 

 branchiostegal rays, and the anus remote from the anal fin. Alcock* 

 is the only author who has used this interesting correlation in the 

 classification of the group, but he used it merely in his key to the 

 species which he referred to 3Iacncrus, and based no genera upon 

 this correlation. The genus Hymenocephalus alone has seven 

 branchiostegals and the anus immediately before the anal. This 

 correlation confirms the isolated position of Hymenocephalus among 

 the Coryphaenoidinae, as further indicated by the presence of ventral 

 striae, the large lateral and subterminal mouth, the long forward ex- 

 tension of the gill openings, the smooth dorsal spine, the thin and 

 weakly armed scales, the presence of two ventral lens-shaped bodies, 



iThe Fishes of North and Middle America, Bull. U. S. Nat. Mus., No. 47, pi. 3, 1898, 

 p. 2561. 



~ Oceanic Ichthyology, Smiths. Inst., Sp. Bull., 1895, p. 389. 



3 Expedition Antarctique Beige, R&. du Voy, du S. Y. Belgica. Zoologie : Poissons, 

 1904, p. 32. 



* Alcock, Journal of the Asiatic Society of Bengal, vol. 63, pt. 2, 1894, pp. 126 and 127 ; 

 Goode and Bean, Oceanic Ichthyology, 1S95, p. 532 (reprinted from Alcock) ; and Alcock, 

 A Descriptive Catalogue of the DeepSea Fishes in the Indian Museum, 1899, pp. 108 

 and 109. 



