studies on mni'ii\c Ostrai.'ods " 99 



resemblance to this appendage in the C y p r i d s than to the corresponding organ in the 

 other Halocyprids. 



Second antenna: — It is true that this antenna issues at the same phice ,,zu beiden 

 Seiten der Oberlippe", but in other respects it shows rather far-reaching differences in Myodo- 

 copa and Podocupa. It is this organ, from the structure of which G. 0. Sars has 

 given the names to the two groups, that really seems to me to be the best 

 support for G. W. MULLEr's classification. There are, however, a 

 number of facts that seem to deserve closer observation. According 

 to G. W. MtJLLER, an important difference between Myodocopa and Podocopa is that in the 

 former group a knee has been formed between the basale of the protopodite on the one hand and 

 the exopodite and endopodite on the other, while in the latter group a knee has been formed 

 between the coxale and the basale. In the families among Podocopa in which a knee is now 

 formed between the protopodite and the endopodite, i. e. in C y p r i d s, D a r w i n u 1 i d s, 

 N e s i d e i d s and Cytherids, in other words in all the families of this group except 

 Cytherellidae, the distal protopodite joint would be absent and so the knee would not be formed 

 between the basale and the endopodite, but between the coxale and the endopodite. The joint- 

 like process from which the second antenna in Podocopa issues would not originally have belonged 

 to this limb. It seems, however, as is shown on p. 79 above, far from impossible that G. W. 

 MUllbr is quite mistaken in this matter. As a matter of fact it seems not at all improbable 

 that the large distal protopodite joint in C y p r i d s, D a r w i n u 1 i d s, N e s i d e i d s and 

 Cytherids corresponds to the basale, and that a knee has thus been developed in these 

 forms at the same place as in the families belonging to Myodocopa; in this way the formation 

 of a knee between the coxale and the basale would only have arisen in the Cytherellids. 

 The part of the second antenna that seems specially to support the affinity of the Cypridinids, 

 Halocyprids and P o 1 y c o p i d s is the exopodite. The agreement in structure is, as 

 has been shown, too great to justify the assumption of convergence. It is, however, to be 

 noted that G. W. MUller himself assumed that this antenna had in the Protostracods 

 an exopodite of the same type as in the P o 1 y c o p i d s. Although this assumption is by no 

 means proved yet, it is nevertheless, as has been pointed out above, not impossible that it is 

 correct; on the contrary there seems to be a certain amount of probability that it is so, cf. p. 82. 

 If this is correct it is obvious that this character will lose a great deal of its classificatory 

 value. The endopodite of this family cannot be used as a basis for a classification of the sort 

 assumed by G. W. MOLler. The fact that this branch is developed as a clasping organ in 

 Cypridinids and Halocyprids seems to be of little importance. This is probably, 

 as is shown on p. 82 above, a convergence phenomenon; at any rate it is not impossible that 

 this is the case. ' . • 



Mandible: — According to G. W. MtJLLER's statement, 1894, Myodocopa is dist- 

 inguished from Podocopa especially by its extensive and often very movable mandibular palp. 

 This character is certainly of very slight classificatory value. As a matter of fact this limb 

 is subjected to not inconsiderable variations both in Myodocopa and Podocopa. The man- 

 dible of the Halocyprids reaUy approaches more closely to the t}i)o that is characteristic 



Second niilrniifj. 



Mandible 



