76 MEMOIRS OF THE CARNEGIE MUSEUM 



66. Supra-anal photophores angulated. 



d. Second ventral photophore not elevated. 



e. A single posterolateral. Second precaudal little above the first, far below 

 the lateral line. 

 /. All postero-anals behind the anal fin; 4 to 6 in number . M, evennanni. 

 //. First postero-anals above the anal fin; 8 to 11 in number. 



M. californiense. 

 ee. Two posterolaterals. Anterior postero-anals above the anal fin. Second 



precaudal widely separated from the first M. reinhardti. 



dd. Second ventral elevated, vertically above the first. 



g. No suborbital photophore. First two supra-anals in line with the supra- 

 ventral M. pterotum. 



gg. A large suborbital photophore. First two supra-anals in line with the 



second (elevated) ventral M. suborbitale. 



aa. Four photophores above the lateral line, near the dorsal profile M. valdiviae. 



5. Myctophum afiine (Liitken). 

 Scopelus affinis Liitken, Vidensk. Selsk. Skr., (6), 1892, p. 252, fig. 10; Atlantic 



and Indian Oceans. 



Seventeen immature specimens were secured at the surface from Ma- 

 tsushima Bay and to the southward. In none of these specimens are the supra- 

 anal photophores angulated and the precaudals obliquely placed and distant, 

 as figured for this species by Brauer (Die Tiefsee Fische, 1906, p. 191, fig. 106), 

 nor have these conditions been present in any of the very numerous examples 

 examined from various parts of the Pacific. The supra-anals are in a perfectly 

 straight line, or the middle one of the series (never the lower one) is rarely very 

 slightly advanced. 



The number of anal photophores is subject to wide variation in this form, 

 if the entire range of the species in all oceans is taken into account; but in any 

 one locality, or group of contiguous localities, the variation is much more re- 

 stricted, and is grouped about the mode in a symmetrical manner. It is im- 

 portant to determine the variation curve in as large a number of individuals as 

 possible, as a contribution to the regional variation of pelagic forms. The 

 lantern-fishes are especially well adapted to this study, as the photophores vary 

 regionally under conditions, which seem to leave the remainder of the organism 

 unmodified, and serve therefore as a more delicate gauge of slight divergence 

 than is usually available. 



The number of specimens in the present collection is too small to enable us 

 to determine finally the range of variation of the antero-anal and postero-anal 

 photophores in Japanese waters, but seems to indicate a close correspondence 



