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



The really phosphorescent apparatus is to be sought in the cells which occupy the 

 ■cavity of the cup, and which are evidently connected with the ganglion-cells by means 

 of the radial fibres issuing from the disc. 



The structure is intermediate between the simple form in the simple ocellar 

 phosphorescent organs and the more complicated structure of the composite organs with 

 special reflector, to be described below. 



The fish can at its option incite the organ, which under ordinary circumstances 

 is non-luminous, to phosphorescence. The voluntary impulse is transmitted by the 

 thick nerve of Leydig to the disc of ganglion cells, which excite the phosphorescent cells 

 in the cup to action by means of the radial nerve-fibres. 



The phosphorescence of the latter requires, however, the secretion of the gland in the 

 sphere as fuel, in a manner similar to that which has been described in the simple organs. 



The secretion passes through the disc and the radial fibres to its destination, where 

 it is consumed to pi'oduce light. 



e. Development. 



Although no transitional forms have been observed by me, I do not doubt that these 

 composite phosphorescent organs have been developed from the simple ones by continued 

 differentiation of the parts. During this progress of development the shape, which is 

 very varialile in the simple organs, has been determined. Their number has been 

 reduced and likewise determined, together with their position. From a great number of 

 undiff"erentiated organs scattered irregularly over the surface of the body, a small 

 number of more highly difierentiated organs, with a definite position, has been 

 evolved. 



4. Composite ocellar organs, with special reflector. 



a. Distribution. 



These most highly differentiated phosphorescent organs have been found by me ia 

 Argyroixlecns Jiemigymnus, Sternoptyx diaphana, and Scojpelus henoiti, whilst Ussow ' 

 lias seen them in Argyropelecus hemigymmis and designates them as " dritsenahnliche 

 Organe." Leydig,^ who designates them as " glasperlenartige Organe," has found them 

 in Gonostoma demidatum, Argyropelecxis hemigymnus, Scop)elus rissoi, Scopeliis huin- 

 holdtii, Scopehis henoiti, Scop>eli(s J>onap>artii, Scopelus rafinesquii, and *Scopc/».s- 

 meto2)oclam2ms. 



The number of these organs and their distribution on the body are very regular, and 



' M. U.ssow, loc. cit., p. 108. - M. Ussow, loc. rit. 



