104 TAGR RKOaSBERa 



the second and third endopodite joint of this limb. I do not understand at all what G. W. 

 MfTLLER means about the other character. In the first place the number five is not at all uni- 

 versal for these bristles in the Cyprids; cf., for instance, figs. 1, 2 and 13 of pi. 13 in G. 

 W. Mt'LLER's work of 1894. The first two figures show the second antennae of Paracypris rara 

 (G. W. MUller). ^ and ?; the former is characterized by seven, the latter by six bristles at this 

 place. Fig. 13 of Macrocypris succinea G. W. Mt'LLER, shows four bristles at the corresponding 

 place. If we examine more closely G. W. MOller's figure of the second antenna of Cytherella 

 sordida G. W. MOller, 1894, pi. 32, fig. 3, we shall find that the first endopodite joint has a 

 dense row of powerful bristles along the whole of the inside of the distal boundary; I think as 

 manv as eleven can be coimted. The same was true of a species of this genus that I had an 

 opportunity of investigating; there I counted twelve bristles. A row of similar bristles is also 

 found, in addition, distally on the inside of the first exopodite joint of this antenna in the 

 (j' V t h e r e 1 1 i d s. From a structural point of view too the distal bristles on the first endopodite 

 joint in Cytherella show no resemblance to those similarly situated in the family Cypridae. 



Fifth limb: — With regard to the strength of the evidence afforded by the last- 

 mentioned character of G. W. Muller's it will perhaps be sufficient to point out, first, the great 

 dift'erence that exists between the male fifth limb in Cytherellids and Cyprids, 

 secondlv. that in the families most closely related to the C y p r i d s. namely the N e s i- 

 d e i d s and tlie C y t h e r i d s, this appendage is not modified in the males as a clasping 

 organ, but is cU'veloped as a typical crawling leg and, thirdly, that in the males of Cytherella 

 the sixth limb too is developed as a clasping organ, perhaps resembling in its t^^e the fifth limb 

 of the Cyprids even more than the fifth limb does. 



This seens to show that we have every reason to consider that the classificatory 

 evidence afforded by the characters brought forward by G. W. Mt'LLER is rather uncertain. 



On the same page of the work quoted above G. W. Mt'l-LER states that there is possibly 

 a close relationship between Cytherellids and I) a r w i n u I i d s ..in der Vermehrung 

 der Borsten des 1. Tastergliedes der Mandibel und ihrer Anordnung zu einem Kamm". A 

 similar comb of bristles also occurs on the maxilla of Cytherella. but not, on the other hand, 

 on this limb in D a r w i n u 1 i d s. Might not this be explained as a sign of relationship with 

 — AsteropeV. I think that we shall not arrive very far with such uncertain assumptions, 

 it is obvious that G. W . Ml'LLER himself does not attach much value to his surmise. He writes 

 (p. 190) as follows: ,,Legt man Werth auf diese Beziehung, so wird man zu der Annahmegedrangt, 

 (hiB die D a r w i n u 1 i d e n die Vorlaufer der C y t h e r e 1 1 i d e n waren, ihnen nnhe stehen ; 

 dies halte ich aber wegen der iibrigen starken Abweichung fiir unwahrscheinlich." 



I do not mean by this to say that Cytherelliformes are not more closely related to Cypri- 

 jormes than they are to any other group. As a matter of fact I consider that this is by no means 

 impossible. What I wish to say is that the position of this group, which is aberrant in almost 

 all respects, is very uncertain and that so far no evidence has come to light that allows 

 ns to assign to it with any degree of certainty a place in the natural system of the Ostrocods. 

 Cypnformcs. With regard to the classificatory position of the four families belonging to Cypriformes 



G. W. MCLler assumes that the Cyprids are nearest to the original type; the N e s i d e i d s 



