86 TAGE SKOGSBERG 



see above, that the P r o t o s t r a c o d s had a fifth limb of about the same tvi^e as that 

 of the recent genus Macrocypris, but with a well-developed vibratory plate on the protopodite. 

 It seems to me very difficult to decide with certainty which of these views is correct. On 

 the one hand I consider it by no means impossible that the foliaceous type in the Cypridinids 

 may be original; this assumption agrees, of course, with the hypothesis that is almost univer- 

 sally adopted nowadays, namely that the foliaceous type is the original one for this limb in 

 the Crustacea and that the rod-shaped limb is a secondary type developed from the foliaceous 

 one — in most cases via the biramous stage. On the other hand I think it far from 

 impossible that G. W. MtlLLER is nearest to the truth and that the foliaceous type is of a 

 secondary nature in the Cypridinids, that in this group this limb was shortened in 

 connection with its development as the most important or at any rate one of the most important 

 masticatory organs. It is obvious that G. W. MULLER based his assumption on the agreement 

 found between the fifth limb in the Halocyprids on the one hand and this appendage in 

 Cyprids, Darwinulids, Nesideids and C y t h e r i d s on the other. This agreement is 

 certainly striking, but this is such a relatively simple organ that I can by no means consider it quite 

 impossible that the resemblance is due to convergence. See also below, the sixth and the 

 seventh limbs. 



It is possible, however, tliat the foliaceous type is the original one and that the 

 rod-shaped type was developed from it, without it being necessary to assmue that the 

 resemblance between the fifth limb in the Halocyprids and the Cyprids, etc. is 

 necessarily the result of convergence. This presujjposes, however, that the ancestors of the 

 Cyprids, Darwinulids, Nesideids and Cytherids branched off from the 

 ancestors of the Halocyprids after the latter had been differentiated from the ancestors 

 of the C y p r i d i n i d s. 



With regard to G. W. MCller's assumption that the original number of joints on 

 the exopodite* of this limb was fom- I only wish to point out that this is partly based on 

 presumably incorrect homologizations. This writer states that the Cypridinids have four 

 joints on this branch; in doing so he comited the basale of the protopodite as the first endo- 

 podite joint, the first and second joints of the exopodite as joint no. 2; on the other hand this 

 writer has not paid attention to the fact that there is sometimes an additional joint distally 

 of joint no. 4, sensu G. W. MtJLLERl. According to this author the Halocyprids also 

 have four joints on this branch; he arrived at this number by counting the endopodite as the 

 first exopodite joint; cf. G. W. MOller, 1894, p. 60. It is, however, to be noted that in one 

 genus of this group, which was not known to this writer when he put forward the assmnption 

 discussed here, namely the genus Thaumatocypris, the exopodite has four joints; cf. G. W. 

 MOller, 1906 a, pi. VI, fig. 3. This is noteworthy, as this genus is in many respects to be consid- 

 ered as the most primitive among the H a 1 o c y p v i d s. With regard to the uncertainty 

 of the homologization of this limb compare ]^. 54 above. 



G. W. MfjLLER's assumption that this limb was developed as a seizing organ in the 

 males of the ancestors of the forms which he groups together under the name of Podocopa 



* Exjjlaiui'd by G. VV. Mlllkh as an oiuluiioilile. 



