COPEPODA 59 



also been examined at se\eral other stations (f. iust. Gnnolfsvig Iceland), where I often found that the 

 small specimens had the nrosonie comparatively longer (f. inst. ant. div. 1-9 x uros.), but with several 

 exceptions. According to Sars the anterior division of Ps. gracilis is comparativel}- more slender; I 

 found, certainly, that a specimen of t\-pus ?iiajor had the anterior division 2-3 as long as wide, one of 

 t\-pus gracilis 27 as long as wide, and one of typus cloiigahis 2-4 as long as wide, but as several 

 exceptions were found, I was not able to use this character. I first got the impression that the 

 characters found in the comparatively long antennulae and natatory legs in several specimens of 

 the typus gracilis were useful, but a stud)' of greater material made it evident to me that they 

 were too variable to be of any value. The feature which marks out some specimens as gracilis 

 i.s the curiously prominent forehead, which is in any case very seldom found in the biggest as 

 well as in the smaller specimens. But as all transitions were found between the different shape of 

 head, even in specimens which in other respects were most like gracilis, it was impossible for me to 

 accept more than a single species. 



Description, f?. I was not able with certainty to find any secretory pores in the natatory legs. 

 The labruiii proper, which is the most posterior of a row of more or less prominent elevations 

 found behind the insertion of the rostral filaments (text-fig. 13 a), has along its hinder margin, on each 

 side, a group of fairly long hairs, and in the middle, somewhat in front, a hairy protuberance. The oral 

 surface of the labrum possesses four groups of short hairs, as seen in fig. 8 (PI. I). The lamina la- 

 bialis has along its posterior margin three indistinct rounded eminences ; orally the two usual longitudinal 

 series of fairly short hairs are observed, of which the more lateral is longer and with longer hairs. 

 Behind the lamina a transverse row of short hairs or teeth is observed and a similar but longitudinal 

 one is found inside the serrula 6-dentata. Between the labial lobes, which are covered with hairs, the 

 arrangement of which were not made out, about four groups of hairs are found, of which the hind- 

 most, which has longer hairs, stands more laterally. Between this and the lateral margin behind the 

 labial lobes 2 oblique groups of hairs are present. 



Behind the area labialis proper is found on each side a large group of fairly long hairs in addition 

 to a few hairs more laterally, and in the middle a triangular group, placed more posteriorly and with 

 its hindmost, wider part fused with a rather irregular group of setae placed between the insertion of 

 the maxillae. 



fc?. As far as the full-grown males, of which comparativelj- few were examined, were concerned, 

 I could not find other differences between the different specimens than that of size, varying from i-i 

 to 1-4 mm. Canu and Sars have given descriptions and figures of the rudimentary mouth limbs. In most 

 specimens an indication between the head and first thoracic somites was seen as figured by Canu as 

 well for the male as for the immature specimens. 



Y. (Stage IV). Size: r2— rg mm. The last thoracic somite is generally more produced than 

 that of the mature females, and dorsally a more or less marked limitation between the two last 

 somites is found (text-figs. 12 b— c). The urosome is comparatively short and sometimes only one third 

 of the anterior division; according to Kraefft (figs. 15 — 18) the two first somites are better separated 

 in the males than in the females; generally no such differences were observed, but in some young 

 females, especially of the smaller "variety", the two first somites were somewhat produced below, 



