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



rows, on the head, &c., are slightly larger and more conspicuous than the others. The 

 average measurements are : — 



They arc accordingly larger than the simple organs, yet much smaller than the much 

 more highly differentiated composite phosphorescent organs, with reflectors, found in 

 Argyropelecus, Scopelus and Sternoptyx. They are intermediate between these two 

 kinds in size and difi"erentiation of structure. The whole organ is covered, with the 

 exception of the exposed outer surface, by a dense layer of pigment very similar to that 

 described above as forming the sacs which enclose the simple phosphorescent organs. 

 This layer is about 0'02 mm. thick, and it is most dense at the stricture ; round the 

 lower spherical part its thickness is greater than in the wall of the cup. Towards the 

 margin (PI. LXIX. fig. 3) it becomes very much thinner. This pigment is the same as 

 that in other parts of the skin. There is a layer of pigment just below the outer surface 

 and there are thin layers of it parallel to the otiter surface further down which join the 

 pigment coat of the phosphorescent organ. The nerves and bloodvessels penetrate the 

 pigment coat. A thin but very distinct membrane (PI. LXIX. fig. 3, m) is found within 

 it, which clothes the whole of the inner surface of this coat. It is very thin, being 

 only 0"0015 mm. in thickness, but it is conspicuous in sections in consequence of its high 

 refractive power ; it replaces functionally the complicated reflectors observed in the 

 more highly diff"erentiated phosphorescent organs. 



The outer surface is closed by a continuation of the ordinary cuticle, which is convex 

 (PI. LXIX. fig. 3, h) and appears as an immovable cornea. Below this, two other 

 thin membranes are found, the outer one of which is structureless like the cuticle, whilst 

 the inner one consists of cells (PI. LXIX. fig. 3). In the specimens at my disposal, these 

 membranes were generally folded and appear collapsed, l)ut it seemed to me that in 

 the living state the space between the innermost and intermediate membrane (PI. LXIX. 

 fig. 3, cl) is probably filled with a kind of corpus vitrcum, traces of which have been 

 observed Ijy me in some of the sections. In that case this structure would be a le7}s. 

 Its development, however, seems subject to specific variations. Ussow ' in his figure of 

 the organ in Stomias anguilliformis indicates a structure corresponding to that observed 

 by me particularly in Astronesthes niger. I am not prepared to say to what extent the 

 diS"erences between these structures in difierent species may be due to difierences in 

 the state of preservation of the various specimens. 



* M. Ussow, he. cil, pi. ii. fig. lOB. 



