COPEPODA 



15 



are found. The arrangement of hairs on the labial lobes and the somites behind seems, as far as made 

 out, to be like that of Cal. hyperboreus. In several specimens it was seen that the transverse line, which 

 limits the maxillar somite posteriorly, is continued just in front of the articular cavity of the maxilli- 

 pes and laterally into the membrane, dorsally and posteriorly limiting the head. Accordingly the 

 first, thoracic somite bear two pair of limbs viz. maxillipeds and pes I, and thus corresponds to two 

 somites (cf. Hansen 1893). 



By examining females from Denmark I did not find any difference from those described except 

 in the arrangement of the hairs of the ventral surface of the maxillar and maxillular somite, which 

 is less complex than those figured. 



The only difference between males and females seems to be the more prominent median 

 process of the labrum in the former sex. 



Textfigure i. 



Catanns finmarchiciis G. Q. Labial lobe seen from 



in front witli lamina labialis and serrula 6den- 



tata etc. X c. 170. 



Textfigure 2. 



C. fmmarchicus G. 9- (Stage IV— V). 



a. Pes V sin. (stage V) in anterior view X 85. 



b. Pes IV sin. (stage IV) in anterior view X 85. 



c. Pes V sin. (stage IV) in anterior view X 85. 



Y. (St. V). Size: Varying from 2-2 (Egedesminde, Davis Strait), 2-5 (73' L. N. 8' L. E.) — 4-8 

 mm. (Davis Strait). Gran's specimens (-=- urosome) varied from 2'i — 2'6 and the specimens from due 

 d'Orleans 3 — 4-5 mm. 



This stage, in which I have not been able to distinguish males and females, is easily disting- 

 uished by 4 abdominal somites of which the second is the longest, 27 as long as the first and 1-4 

 as long as the anal somite. The mouth-appendages are scarcely different except the maxillulae, 

 which have only 10 instead of 11 setae in the exopodite. The four pair of natatory legs are scarcely 

 different from those of the mature female, but pes V differs by fused Ri II — III without indication of 

 Se II but with the number of Si 7 as usual, and by the fused Re II — III with 5 Si and well 

 developed Se Re II (textfig. 2 a). Number of secretory pores seem to be alike f^ in pes II — III, but 



