156 ' "TERRA NOVA" EXPEDITION. 



differs from these in certain details of scul^^turing on the carapace, but the evidence 

 afforded by a solitary immature specimen is insufficient to justify either the establish- 

 ment of a new species or an extension of the known range of C. undata from Norway 

 to New Zealand. 



PHYLLOCARIDA. 



23. Nehalia longicornis, G. M. Thomson. 



Nehalia longicornis, G. M. Thomson, 1879, p. 418, pi. xix, figs. 7-9 ; N. I. with var. magellanica, 

 etc., Thiele, 1904, p. 9, figs, on pi. iv ; N. I. magellanica, Thiele, 1905, p. 66, pi. ii, figs. 

 14-17; Thiele, 1907, p. 1, text-figs. 



Occurrence. — Station 130. Off Three Kings Islands, New Zealand. Plankton. 

 Square 18-mesh net at surface. Aug. 26-27, 1911, 8 p.m. to 6.30 a.m. One specimen. 



Station 135. Spirits Bay, New Zealand. Plankton. Square 18-mesh net at 

 3 metres depth. Aug. 31 to Sept. 1, 1911, 9 p.m. to 6.30 am. One specimen. 



Station 331. Off Cape Bird Peninsula, entrance to McMurdo Sound. 250 fathoms, 

 dredge. Jan. 14, 1912. One specimen. 



Remarks. — In the proportions of the rostral plate (2'1 : 1), in the form of the 

 ocular peduncle with its " sensory " tubercle, and in the armature of the fourth segment 

 of the antennule (1 spine, 7 or 8 setae), the specimen from McMurdo Sound agrees 

 almost exactly with Thiele's account of the "Discovery" specimens, and gives evidence, 

 as far as a solitary specimen may, for constancy in the characters of the local race 

 which Thiele refers to his subspecies viagellanica. 



The two specimens from the north of New Zealand are noteworthy, in the first 

 place, for the fact that they were taken with the surface-net. We have no record of 

 the depth of water over which they were swimming, but it is not likely to have been 

 great, and indeed many of the plankton-gatherings from this region contain animals 

 that are, at most, temporary migrants from the bottom-fauna. 



Both the New Zealand specimens appear to be immature, and one of them retains 

 the mucronate termination of the rostral plate regarded by Thiele as a juvenile 

 character. Both specimens have on the anterior margin of the fourth antennular 

 segment one strong spine followed by three or four setae, and so far agree with Thiele's 

 definition of N. longicornis as against the northern iV. bipes. They diverge remarkably 

 from this definition, however, in the narrow form of the rostral plate. In the specimen 

 which is presumably the more mature of the two, the proportion of length to breadth 

 is 2 '76 : 1, that is to say, the plate is considerably narrower than that which Thiele 

 fio-ures (1904, pi. iv, fig. 7) as typical for N. bijyes, the proportion measured from his 

 figure being about 2'3 : 1. In both specimens the eyestalk is short, the corneal area 

 occupies about half of its length, and the " sensory " tubercle is insignificant. 



If we attach primary importance (as Thiele seems to do) to the form of the rostral 

 plate as distinctive between AL bipes and N. longicornis, then these New Zealand 



