356 DISCOVERY REPORTS 



indifferent to any of the stains used, and appears brilliant white by reflected light. 

 Resting within the distal concavity of each proximal granular mass is a basal cap, 

 corresponding precisely with those observed in the carapace organs of Hoploplwrus. 

 The basal caps are thin, with almost plane distal, and proximal convex, surfaces. As 

 has been pointed out earlier (p. 352) these basal caps correspond to the granular zone 

 of the pleopod photophores of Systellaspis and Hoplophorus. 



Lying beneath the photogenic cells is a tissue layer, corresponding with the fibrous 

 reflector of the pleopod photophores, and consisting of a fibrous sheet in which lie 

 large vacuolated nuclei elongated in a plane parallel to the surface of the photophore. 

 It will be recalled that the fibrous reflector of the pleopod photophores is apparently 

 derived from the connective tissue sheath of the photophore nerve, or at any rate is 

 continuous with it (p. 354), and that the nerve fibrils impinge directly on to the granular 

 mass (p. 354). With this in mind, careful search has been made in the neighbourhood of 

 the fibrous layer of the photophore under discussion for any indications of nerve supply. 

 Possibly owing to the fact that the fixation of this specimen is not as good as might be 

 desired, no definite indications of such supply were recognized. I believe, however, 

 that a study of specially preserved material will reveal that these photophores possess 

 their own special innervation in relation to the basal caps in much the same manner as 

 has been seen in the carapace organs of Hoplophorus. 



Underlying the fibrous reflector layer occurs a zone of scattered connective tissue in 

 which muscle strands (Fig. 25, m.str.), fine blood channels (Fig. 25, b.c.) and larger 

 haemocoelic spaces (Fig. 25, haem.) are recognizable. 



Lastly, the basal layer (Fig. 25, b.l.) of the excised strip of tissue is that lining the 

 main ventral haemocoelic space of the posterior part of the thorax. It shows ill-defined 

 cell boundaries and possesses nuclei which are quite different from any observed in the 

 photogenic streak. 



It will be appreciated that a photophore of this type differs fundamentally in structure 

 from that found in the corresponding position in the three species of Hoplophorus 

 examined. 



{d) The photophores on the fifth thoracic limb 



For the purpose of description of the photophores occurring on the thoracic limbs, 

 the proximal spot on the propodus of the last thoracic limb has been selected. All other 

 photophores on the thoracic limbs that have been examined are identical, and from the 

 point of view of their structure do not require further attention. 



These photophores present an extraordinary appearance which is only paralleled 

 by the organ behind the base of the last thoracic limb, and perhaps by that in the limb 

 itself, of Hoplophorus (p. 337). The main mass of the organ is, however, quite unlike 

 those of Hoplophorus just mentioned, being composed of the white opaque granular 

 substance mentioned in describing the uropod photophores (p. 351) and the streak at 

 the base of the fifth thoracic limbs (p. 355). It is present in such quantity that the 

 photophores are readily visible in the sections, even without the use of a microscope, 

 as white masses (see Plate XXV, figs. 4, 5). 



