203 DISCOVERY REPORTS 



It has been assumed from the fact that Nebaliopsis has only been obtained from deep 

 hauls that it lives at great depths. However, specimen C from St. loi is the first 

 example which has been collected at a known depth. All other specimens were collected 

 by nets which were open on their way up to the surface. Specimen C was collected at 

 a depth of approximately i J miles. 



With regard to localities, the Valdivia specimen is the only representative from 

 tropical waters, the remainder having been collected from sub-tropical to antarctic 

 localities. Of the Challenger specimens, one was collected in the south-west Indian 

 Ocean and the other in the South Pacific. The remainder are all Atlantic specimens. 



There is a peculiar mistake in the map published by Thiele (1905, p. 67) in which he 

 records Nebaliopsis as having been collected from a point south of Australia. This is 

 presumably the second Challenger specimen from St. 289 in which the longitude has 

 been plotted E instead of W. 



From the measurements quoted in Table I, it will be seen that there is considerable 

 variation in the length of the carapace. This is partly only apparent and results from the 

 difficulty of accurate measurement owing to the distortion of most of the specimens, but 

 that there is a real variation can be seen by comparing the ratios taken from Ohlin's 

 figure and that from specimen F^, both of which were undamaged. 



There is similarly a marked variation in the structure, but not the size, of the anterior 

 limbs. Thiele gave a very complete description of the setation and detailed anatomy of 

 the limbs of his specimen. I have compared the antennules, antennae and mandibles of 

 five of the Discovery specimens with Thiele's description and, while I found a general 

 agreement, there was a very marked variation. It would serve no purpose to record the 

 setation of each limb of the Discovery specimens, but there are several general con- 

 clusions which can be stated from the comparison. 



The larger specimens are markedly more setose than the smaller. This is seen clearly 

 in the mandibular palp. In specimen A (total length 19-5 mm.) the middle joint carries 

 two setae on its median face and the distal joint a row of fifteen, while specimen C (total 

 length 30 mm.) bears an irregular row of ten setae on the middle joint and a row of 

 twenty on the distal. A more marked diff"erence is seen by comparing specimen C with 

 D, but from other characters it is clear that specimen D is immature. 



The fourth joint of the antennule is produced on its median face into a long pointed 

 process (Thiele, 1904, PI. 3, fig. 41), which carries a row of marked teeth on its anterior 

 edge. These teeth become obliterated in the larger specimens. 



The scale of the antennule, which, according to Thiele, carries a curved row of about 

 fifteen setae, becomes more setose in the larger specimens and loses its characteristic 

 shape (Thiele, 1904, PI. 3, fig. 40). Thus, in specimen D, the scale carries a curved row 

 of seven setae and terminates in a triangular point. In specimen B it carries fifteen setae 

 and is an ovoid plate. 



Specimens D, E and F^ are undoubtedly larval specimens. The flagellum of the 

 antennule is in each case six- instead of twelve-jointed. The most marked difference is in 

 the pleopods. In each case the fourth pleopods consist of a pair of minute knobs, while 



