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



It appears highly probable that the whole structure is very transparent in the fresh 

 state, and that the delicate cells, which appear granular in spirit specimens, are then 

 difficult to distinguish. Perhaps Ussow has failed to see their contours and nuclei. The 

 radial fibres described by him entirely correspond to the fibres observed by me. 



Leydig represents this structure in his figures as consisting of radial tubes, which he 

 considers as glandular. 



Between the squirrel-tail-like columns of cells with the central fibre, very slender 

 membranes, likewise containing nerves as well as capillaries, are found. These membranes 

 coalesce with the watch-glass-shaped external portion of the membrane and are thickened 

 near their base. They divide the fibres with their cell columns completely, as may be 

 observed in longitudinal, but much better in transverse sections which are properly 

 stained. 



These polygonal tubes are identical with the tubes observed by Leydig and are not so 

 distinct as the gland-tubes of the spherical part, but similar to them. 



Ussow has not observed them, whilst Leydig and myself have found them without 

 exception. According to Leydig, however, they are filled with spherical granular cells, 

 similar to those in the proximal spherical portion of the phosphorescent organ. Leydig's 

 material was not very good and this statement must therefore be accepted with caution. 



c. Innervation. 



Leydig^ has discovered the mode of innervation of these organs. According to him, 

 a thick nerve enters it by penetrating the pigment coat and intima, near the central 

 constriction, which divides the spherical from the cup-shaped portion. It then spreads 

 out and joins the granular mass in the centre. 



To these observations, which I can corroborate, I have to add that the nerve is a 

 branch of a spinal nerve, and where there are two composite organs on each side of each 

 segment, the nerves supplying them join a little way above the upper organ of the two. 

 They are the thickest branches of the spinal nerves, and all other branches originate 

 from them in such a way as clearly to indicate that they form the stem of the distal 

 two-thirds of the spinal nerve. The two composite phosphorescent organs appear to be 

 situated at the terminations of the two main branches into which the spinal nerve 

 divides. 



I have mentioned above that Leydig's " central granular mass " is identical with the 

 granular secretion in the centre of the spherical glandular portion, together with the disc 

 of large granular cells situated in the constriction. The nerves of course terminate in, 

 the disc, and not in the secretion. Here they become non-medullated. Other, but very 

 much smaller and quite insignificant nerve-fibres are also found attached to the sphericid 



' F. Leydig, Die augenalmlichen Organe der Fische. 



