REPORT ON THE DEEP-SEA FISHES. '27 i> 



closely like that of the pearl spot on the tail of Scopelus humboldtii. These phos- 

 phorescent organs or " Leuchtorgane " differ from the glass-bead organs in having no 

 gelatinous refractive media. They have the pigment capsule, the blood-vessels, and 

 the layer of polygonal elements, with metallic glance, like the glass-bead organs, and 

 inside this layer a layer of substance which is to be regarded (Ley dig, loc. cit., 

 p. 73) as a specialisation and expansion of the spindle-shaped plate of the glass- 

 bead organs. 



This inner substance ought to be homologous with the rod layer in Ipnops, but 

 unfortunately its structure as described is very indefinite, probably from the imperfec- 

 tion of the material used. In the phosphorescent organs of Scopelus rajinesquii it is 

 stated to be a grey layer of a peculiar transparent homogeneous fine granular substance. 

 In the pearl spot organ on the back of the tail of Sco2')ehxs huinboldtii, it is said not to 

 be composed of cell elements, but made of very finely granular material traversed 

 by channels through which the blood-vessels pass. Nevertheless, on side view, an 

 appearance is seen as if it were composed of pear-shaped elements (Leydig, loc. cit., 

 pp. 53, 54). From the figure given of the microscopical structure of this layer (Leydig, 

 loc, cit., Taf. X. fig. 61), and the description of its iieculiarities, it seems possible that it 

 may eventually prove to be made ujj of fine rods something like those composing the 

 rod layer in Ipnops. The small blood-vessels traverse the layer just as in the case of 

 the rod layer of Ipnops. Leydig describes the cell elements lying next to the tapctum 

 in the eye-like organs of some fishes as approaching in form and power of refraction 

 the crystalline rods of Arthropoda ; these must surely be identical with the rod bodies of 

 Ipnops (Leydig, loc. cit., p. 80). 



On the whole, it seems not unlikely that the remarkaljle head organs of Ipnops are 

 to be regarded as highly specialised and enormously enlarged representatives of the 

 phosphorescent organs on the heads of such allied Scopelidse as Scopelus rajinesquii 

 and Scopelus metopoclampiis (Leydig, loc. cit., Taf. x. figs. 55, 56). It may be 

 conceived that in Ipnops the supra-nasal and subocular phosphorescent organ of these 

 species on either side have united and become one, with the result of the total obliteration 

 of the eye. In Scopelus metopoclanipus there is a notch which nearly separates the 

 subocular organ into two parts. In the organs of Ipnops there is a similar notch due 

 to an incursion inwards of their pigmented borders on either side (see PI. LXVIL), but 

 it lies on the outer, not on the inner margin of each organ. 



The phosphorescent organs in Ipnops lie entirely outside the ca-sdty of the cranium, 

 although they sink over part of their anterior region into two cavities on either side of 

 the median septum of the skull. Posteriorly they lie quite close to the skin surface on 

 the top of the cranial wall. The margins of the rod layer and pigment layers are every- 

 where superficial. The exact extent to which the transparent roof of the organ is 

 ossified was not made out, as no specimen was available for maceration. The organs are 



