EUPHAUSIA CRYSTALLOROPHIAS 221 



There is no dorsal lobe at the distal end of the first segment of the antennular 

 peduncle as in the other species described; there is a short sharp spine, difficult to 

 distinguish among the setae, on the outer distal corner (Fig. 17 a, b). The dorsal keel of 

 the third segment is well developed and of the same shape as that in E. frigida but per- 

 haps a little lower. 



The abdominal segments have no dorsal spines. 



The male copulatory organ shows many diff'erences from those of the four species 

 already described, but it is on the same pattern (Fig. 30 e). In E. liicens, E. vallentmi and 

 E. frigida the terminal process has a wide and somewhat foot-like base, the toe forming 



Fig. 17. E. crystallorophias. 



a, front part of carapace and first segments of antennules from above, x 12. 



b, left antennular peduncle from the side, x 14. 



the point of attachment, the heel being free on the outer side ; in E. siiperba the basal part 

 of the process is bent nearly at a right angle to the distal part. The terminal process of 

 E. crystallorophias is straighter and its base but little wider than the rest of it; but there 

 is on the outer side, near the base and pointing back towards it, a blunt protuberance. 

 The end of the process is divided into two very unequal parts. That which is the hinder- 

 and outermost is by far the larger ; it is long and tongue-like and curled forwards, and 

 its inner edge is serrate. The other part is finger-like, curved at the end, and lies within 

 the larger as though protected in its curl. 



The proximal process has a secondary process as do those of the four other species 

 (though it is reduced, and may be absent, in E. siiperba). The inner membranous part 

 of it is expanded so that its edge is convex ; it is also serrate. As in the other four species 

 there are at the end of the proximal process in E. crystallorophias two membranous 

 expansions, one behind and one before, the latter striated the former not, but their 

 proportions and shapes are very different. The foremost expansion is very small and 

 set comparatively far back on the process, the middle of its base opposite the secondary 



