THE RADIATING ORGANS OF THE DEEP SEA FISHES. 199 



radially in a para tangential plane. In the central part of the lower portion 

 of the disc numerous large cavities, which appear empty in sections (Fig. 13 

 cv), are observed. At the base of the disc a layer of scattered pigment cells 

 (Fig. 13 p) occurs. No particular structures could be made out in the 

 tissue forming the disc itself In transverse sections (Fig. 13) through it 

 one sees paratangentially orientated plate-like or spindle-shaped elements, 

 probably connective-tissue cells, sections of nerves and blood vessels and 

 here and there granular elements which look like nerve cells. 



As in the case of Bassozetus (p. 198), also here in Leucicorus hisciosus the 

 possibility of the discs examined being incomplete must be taken into con- 

 sideration. If the discs as observed were covered by a layer of cylindrical 

 cells these organs of Leucicorus would become very similar to those of Ha- 

 losaurus described below. Such a cylinder cell layer may have beer 

 present on these organs during life ; I failed however to find any trace of 

 it in the sections. 



Halosaurus radiatus Garman. 



Plate 11, Figs. 53-57. 



This species has been described by Garman ('99, pp. 298, 364, Plate 60, 

 Figs. 2, 2 a, Plate 84, Figs. 3-6). The radiating discs are situated on the 

 head and, according to Garman ('99, p. 298, Plate 60, Fig. 2, Plate 34, Figs. 

 3-6), also in a longitudinal row on each side of the body. They lie under 

 the skin and are spindle-shaped, the long axis being disposed paratangen- 

 tially. The largest are 3 mm. long. On each side of the head there are 7 

 maxillary, 7 orbital, 8 ethmoidal, 4 occipitoparietal, 4 mandibular, and 9 

 opercular radiating discs. Those on the body could not be made out in the 

 somewhat injured specimen at my disposal. 



In Halosaurus lYidiatus the radiating discs are obviously connected with 

 the slime-canal system. They are in fact local differentiations of the walls 

 of certain slime canals. The cephalic slime canals in which the organs of 

 the head lie, extend as usual paratangentially below the outer skin. These 

 slime canals are enclosed in a stout, apparently structureless sheath (Plate 11, 

 Fig, 56 s). Within this a layer of connective tissue (Figs. 56, 57 c) is met 

 with, in which slender, probably spindle-shaped cells arranged paratangen- 

 tially can clearly be made out. This connective-tissue layer is stouter on the 

 lower side, that is, in the floor, than on the upper side, that is, in the roof of 



