354 DISCOVERY REPORTS 



distinguishable from them. On the other hand, they are quite different from both the 

 nuclei of the connective tissue sheath of the photophore nerve (Fig. 24, ?i.coim.t.) and 

 its neuroglia nuclei (Fig. 24, ruigl.), both of which are highly distinctive. The connective 

 tissue nuclei have a much more diffuse chromatin arrangement, and their coagulated 

 nuclear fluid stains lightly with aniline blue, whereas the neuroglia nuclei are small, 

 with very densely packed chromatin. I believe that the cone of nuclei represents a 

 mass of chromatophores still further serving as a screen preventing the inward passage 

 of light. 



The massive nerve branch to the photophore (Fig. 24, phot.nv.) shows in the neigh- 

 bourhood of the ventral nerve cord some of its constituent elements — nerve fibres, 

 connective tissue sheath and its nuclei, and neuroglia nuclei. The neuroglia fibres are 

 not discernible with the fixative and stain employed. In the region of the photophore, 

 however, these elements are not so clearly distinguishable, and the nerve shows a 

 greater development of its connective tissue sheath. At the apex of the cone of chroma- 

 tophore nuclei, on the inner side of the photophore, the connective tissue sheath of the 

 nerve expands to form a cup-like investment of the organ (Fig. 24, r.). Since this 

 structure occurs in the position occupied by the fibrous reflector of the pleopod photo- 

 phores I have retained for it the term "reflector", although as far as can be seen its 

 function is largely usurped by the massive granular reflector. Just as in the pleopod 

 photophores, elongate vacuolated nuclei are found in the reflector at the level of the 

 photogenic cells (Fig. 24, n.r.c). 



As far as I can ascertain, the nerve fibres do not expand with the fibrous sheath and 

 become involved in the reflector, but pass through the cone of chromatophore nuclei 

 and impinge on the proximal granular masses of the photogenic units. The precise 

 mode of entry of the nerve fibrils to the photogenic units has not been discovered. 



To recapitulate briefly, the distinctive features of the uropod photophores are their 

 possession of intracellular lenses and a reflector of an entirely novel type. 



(c) The transverse streak behind the coxa of the fifth thoracic limb 



As has been previously stated, the transverse luminous streak behind the base of the 

 fifth thoracic limb of Systellaspis affinis occupies the same position as in S. debilis 

 and Hoplophoriis. In my specimen it consisted of a somewhat raised white opaque 

 band, extending laterally and dorsally and slightly anteriorly. It is indicated in the 

 photograph in Plate XXVI, fig. 2, in which it did not show very clearly, as a white 

 streak (phot.t.s.). On cutting out and sectioning the strip of tissue it proved, contrary to 

 expectations, to have an entirely different structure from the similar organs of Hoplo- 

 phoriis, and also from the photophores on the fifth thoracic limb itself. As will be seen 

 from Fig. 25 it has the same basic structure as the pleopod, and particularly the uropod, 

 photophores. It may be said at once that the proximal granular mass of white opaque 

 substance mentioned in the description of the uropod organs is present here, and 

 accounts for the external appearance of the organ. 



The elevation of the luminous streak above the adjacent integument is accounted for 



