196 THE VOYAGE OF H.M.S. CHALLENGER. 



surface net, but disappear during the daytime, when they evidently descend to a depth 

 to which only a moderate amount of light penetrates. A few, like Scojjehs macro- 

 lepidotus, and, perhaps, Scopclus glacialis, undoubtedly belong to the bathybial fauna, 

 but with recrard to the other species admitted in this Eeport, I consider it equally 

 probable that they accidentally entered the dredge during its ascent. Only a few 

 specimens were captured in this manner, much fewer than of Argyropelecus, a fact 

 which is no doubt due to their greater activity, by which they are enabled to make their 

 escape on perceiving the approach of the net. 



Scopclus macroleindotus (Johnson). 



Of this fine and large species, which hitherto was known from Madeira, six specimens, 

 from 6^ to 8 inches long, were obtained at the Kermadec Islands, Stations 170 and 

 170a, in 520 and 630 fathoms. 



Scopclus glacialis. 



Scopelm glacialis, Reinh., Nilss., Kroy. 



Scopelus miilleri ^ (GmeL) CoUett, ISTorges Fisk., p. 152; Norsk. Nordh. Exped. Fisk., p. 158. 

 „ „ Goode and Bean, Bull Mus. Comp. Zool., vol. x., 1883, p. 222. 



This species, which has been not rarely observed as a surface fish in the northern seas, 

 is stated by Collett to have been obtained by the North Atlantic Expedition in a 

 dredge, which was worked at a depth of 1110 fathoms. The U.S. steamers took it 

 frequently at depths of 300, 400 and 600 fathoms ofi" the southern shores of New England 

 and off the coast of South Carolina. All the specimens were in a more or less injured 

 condition, as if they had been dragged a long way through the water, and as Collett 

 found in their stomach an Ostracod {Conccecia horealis) wliich is not foimd at a less 

 depth than 300 fathoms, it is highly probable that this species is bathybial. 



Scopclus antarcticus (PI. LI. fig. D). 



Scopelm antarcticus, Giinth., Ann. and Mag. Nat. Hist., 1878, vol. ii. p. 184. 



D. 15. A. 20. P. 13. V. 9. L. lat. 38. 



The height of the body is two-ninths of the total length (without caudal) ; the length 

 of the head is contained thrice and two-thirds in it, the depth of the head equals its 

 length without snout. The diameter of the eye is somewhat more than one-third of the 

 length of the head ; distance between the posterior margin of the orbit and the prse- 

 opercular edge one-third of the diameter of the eye. Snout short, obtuse, with its upper 



1 There is no advantage to be gained in abandoning a well established specific name for an uncertain nomenclature. 

 Oar best thanks are due to Hr. Collett for having unravelled the confused liistory of the first knowledge of these 

 northern Scopeli, but, even if no confusion had taken place in Strom's account, I should hesitate to supersede a name 

 given by an original obsei"ver like Reinhardt, by another which is proposed by a mere compiler. 



