REPORT ON THE DEEP-SEA FISHES. 299 



part. There are generally two or three such nerves, which apparently have nothing to 

 do with the large nerve supplying the central disc. 



Ussow's representation of the innervation of the organs in Stomias anguilUfoi'mis, 

 where a single nerve is said to enter at the back of the spherical part, does not coincide 

 with Leydig's observations or my own. 



The innervation of these organs corresponds to that of the simple phosphorescent 

 organs above described, and in every case the nerves supplying them are branches of 

 spinal nerves. 



d. Function. 



It is not very difficult to perceive the functions of the different parts of these 

 organs. The pigment coat prevents the light from issuing in any other direction than 

 that indicated by the axis of the organ. 



The membrane which reflects the light produced within the organ is, in con- 

 sequence of its shape, very well adapted for the purpose of concentrating the light in 

 one cone. 



The optical action of this reflecting membrane is illustrated in PI. LXIX. fig. 4. 

 The spherical part reflects all the rays of light which are thrown in a centripetal 

 direction, until they pass from it into the cup-shaped portion. As the cup has the 

 shape of a rotation-paraboloid, all the light which issues from the focus of the parabola 

 will be reflected parallel to the axis of the optical system, whilst light coming from other 

 points in the neighbourhood of the focus will be reflected in a more or less simdar 

 direction, so that all the light produced within the organ issues from it in the shape 

 of a cone, the axis of which is the continuation of the axis of the optical system. The 

 light passing through the concavo-convex lens is further concentrated and may be emitted 

 from the fish as an intense flash in the direction of the axis. 



There are muscles surrounding these phosphorescent organs, but it is doubtful 

 whether the organs are movable. If they were, the, fish could of course throw flashes of 

 light from them in any desired direction, and by setting tlie axes of aU the organs 

 of one side in the same direction, might produce a. " broadside " of light-flashes 

 sufficient to illuminate objects at some distance. 



There can hardly he any doubt that the spherical part of the orgtm is filled \\ith a 

 glandular structure. The cells of the layer found just wdthin the membrana propria are 

 probably ganglion cells in connection with the nerves which supply this part of tlie 

 organ. The gland-tubes are morphologically very similar to tliose in the simple organs, 

 and produce a granular secretion by the same process of direct conversion of cell- 

 substance into secretion. The secretion accumulates in the space below the disc, and the 

 disc itself is evidently composed of ganglion cells, which are in connection with the large 

 nerve, to be designated after its discoverer, Leydig's nerve. 



