42 TA(1K SKOGSBERG 



fully developed in this primitive form than in the sub-family Cyfridininae and, as in the latter, 

 it was armed with numerous bristles. 



\^Tiich of the explanations described above. C. Claus's, G. W . MOllrr's of 1890 and 

 that of the same investigator of 1894 is to be considered correct — judging from the view of 

 the method of homologization applied by me above to the families Polycopidae, Saraiellidae 

 and Cypridinidael Or is it possible — judging from this ]ioint of view — that any of them 

 on the whole is to be considered quite correct? 



No very definite answer to these questions can be given — at least at present. The 

 strongly modified tvpe of this limb and the complete absence of any known intermediate forms 

 seem — as has been shown above and as G. W. MULLER has also previously pointed out — to 

 make it almost impossible to carry out a certain homologization of this organ at the present time. 



It seems to me very improbable that, as G. W. Mt'l.LER assumed in 1894, the comb 

 has developed only from the part denoted above as the second joint of the exopodite. on account 

 of the fact that this organ issues near the base of the limb. The same reason militates to an 

 even greater extent against G. W. MOller's assumj^tion of 1890. It seems to me most probable 

 that the proximal part of the comb has been formed by what I termed above the protopo- 

 dite, its distal part by what is called above the first and second joints of the exopodite. The 

 long bristle (or the two long bristles) with the short bristles situated near it (them) on the lateral 

 side of the comb presumably belong, according to my idea, to what I have called the third and 

 fourth exopodite joints; on the other hand it seems quite impossible to decide whether they 

 belong to only one of or to both these joints. It is uncertain, however, whether this homo- 

 logization is more correct than that worked out by G. W. MOller. It is at present basefl 

 only on such weak arguments as the relative positions of the different parts. 



The vibratory plate on this limb is of about the same type as in the families Polycopidae, 

 Sarsiellidae and Cypridinidae and must certainly behomologized with this organ in these families. 

 According to the explanation given above it is consequently to be considered as an epipodial 

 appendage. Of. the adjoining figure IV: 3, 4. 

 iininnjitruhir. Family H(docypridae: At the first glance there seems to be a great gap between the type 



of the fifth limb in the families Sarsiellidae and Cypridinidae and that found in the same organ 

 in the family Halocypridae. While in the two first-mentioned families this appendage is deve- 

 loped as a more or less typical foliaeeous liml). in Halocypridae it is a typical rod-shaped limb. 

 Still it is possible to show, although with a certain amount of doubt, which parts of this limb 

 are homologous in these three families. 



If the homologization employed above is applied to tiie fifth limb of the H a 1 o c y p r i d s 

 we shall find the following results: The vibratory plate is to be considered as homologous to th(» 

 same organ in the preceding families and is consequently to be denoted as an epipodial 

 appendage. The three distal joints are j)robably to be homologized with the process that 

 is denoted by me above as the exopodite. The protopodite, which is proportionately 

 almost as large as in, for instance, the Cypridinidae, is sometimes divided into two joints, 

 which are often only weakly marked off from each other, a proximal one, on whicli the 

 vibratory plate issues which is to he consi(l(>red as a procoxale and a coxalc, and a distal 



