82 TAOK SKOOSmORO 



parative omV^rvoloirv. It sopius, however, to be by no means impossible, perhaps even rather 

 probable, that (i. W. Mt'M.Ki; was pretty near the truth when he assumed that the Prot- 

 o s t r a c o li s had a nine-jointed* exopodite, perhaps of about the same type as in tlic 

 recent P o 1 y c o p i d s. The reasons that seem to me to support this are, first, that the 

 forms which are now characterized l)y an antenna of this sort, C y p r i d i n i d s, 11 a 1 o- 

 c y p r i d s and P o 1 a- c o p i d s, are in several respects certainly to be considered as the 

 most primitive s t r a c o d s, and, secondly, that the agreement in structure in these groups 

 is altogether too great to justify us in assuming a convergence, in the case of an organ of so 

 complicated a structure as this branch. At any rate the ancestral forms of these three groups 

 probably had a similar exopodite on this limb. 



The o n 1 \' statement about t li e second antenna of the P r o t- 

 o s t r a c o d s that we can make at p r (> s e n t with almost c; o m p 1 e t e 

 certainty, is, according to my opinion, that it possessed a well- 

 developed protopodite and a comparatively strong exopodite 

 and endopodite. 



We ought perhaps to note the great resemblances there is between this limb in the 

 Cytherellids and the type that is presumably to be considered as the original one for 

 the two following limbs, the mandible and the maxilla. Perhaps this indicates that the 

 second antenna of this group represents a comparatively primitive type. 



Was the endopodite in the males of the ancestral forms of C y p r i d i n i d s, H a 1 o- 

 cyprids and Polycopids developed as a seizing organ, as G. W. Mt'LLER has assumed? 

 In the recent Cypridinids the whole of the end joint of the endopodite is pressed 

 against the preceding joint in seizing. In the H a 1 o c y p r i d s, on the other hand, the end 

 joint has a process in the shape of a seizing arm, which issues proximally on the joint. The 

 males of the Polycopids do not have the endopodite developed as a seizing organ; 

 whether the wart-like process near the point of the endopodite which is found in this group 

 (cf. G. W. MtLLER, 1894, pi. 7, fig. 10) is homologous with the seizing arm in the H a 1 o- 

 c y p r i d s, as G. W. MCller has assumed, I must leave undecided; there is no certain proof 

 of this homologization, G. W. Mt'LLER does not give any and I have not found any myself. 

 In the face of these facts — that the Polycopids, in several respects presumably 

 the most primitive of these three groups, do not have this branch developed as a seizing 

 organ and that the seizing function has been carried into effect in different ways in the 

 Cypridinids and the Halocyprids — one must say that this assumption of 

 G. W. MUller's is to be considered as rather uncertain. 

 Mandible. Mandible: — G. W. Mt'LLER's assumption that this limb in the Protostracods 



was composed of two protopodite joints and three endopodite joints seems fairly probably; 



* G. W. MCller points out, 1894, p. 36 that ..manche Arten weisen darauf liiii, daB die Zahl ursprunglich 

 groBer war ( Philomedes)" . A careful study of some forms of this genus has convinced ine that the chitinous structures 

 on the end-joint of this branch (of the males), to which this writer refers, cannot probably be explained as traces of 

 a tenth joint. In addition it is to be noted that, even it the genus Philomedes originally had a ten-jointed exopodite 

 on this limb, it is obvious that we have no right, because of this, to conclude that the Protostracods 

 origin.-illy had ten joints. 



