684 A NEW ENDOPARASrriC COPEPOD, 



are, with the exception of the labrum, the only structures which 

 break its even surface. 



The antennules are uniramous, blunt-ended, two-jointed limbs. 

 Tlie blunt distal end of the second joint bears two long, stiff 

 bristles. 



The antennae are biramous, composed of a single-jointed proto- 

 podite, and of two rami, each two-jointed. Both rami are blunt- 

 ended, and bear each two bristles very similar to those of the 

 antennules. 



The mandibles differ from the antennae only in being smaller, 

 and in that the inner side of the last joint of the anterior ramus 

 bears two or three of the long bristles, as well as the usual two 

 on the blunt end. 



This completes the account of the development, as far as I am 

 able to take it; the stages between the nauplius and the adult, are 

 probably passed through as a free-living period of the life-history. 

 It, therefore, only remains to review this development, and to 

 classify Ubius. Before passing to review the development, I would 

 like once more to draw attention to the absence of the digestive 

 gland, or rather to its occupying a very primitive embryonic situa- 

 tion, and to suggest that this may be found to be correlated with a 

 very abbreviated ontogeny. 



Review of the Development. — The cleavage is of Korschelt and 

 Heider's first type (3, p. 108), characterised by them as being very 

 rare among the Crustacea. This type of cleavage, however, occurs 

 in two otlier copepods of Avhich the development has been studied. 

 I refer to Cyclops{2), and ChondracantJins(Q) . In both of these, 

 not only is the cleavage total, but it follows very much the same 

 lines as in Ubius. There may be a difference in the arrangement 

 of the cells in the eight-cell stage in Chondr acanthus. 



In all these three copepods, invagination takes an almost identi- 

 cally similar form, which form reappears in Lucifer (1), though 

 here toned down, as it were, and connecting the extreme modifica- 

 tion of the above copepod type with typical embolic gastrulation, 

 I am unable to find tissue-differentiation appearing at so early a 

 stage in Ubius as Hacker was able to demonstrate in Cyclops. The 



