166 



lAC.F- sKOGsriicm; 



Sub-taiiiil 



1 I. l'hilii)iu'ili)i(if 



III. SarsirUindr 

 1 \ . Asti'iufjiiuic 



Of tlu'se genoni <-'. W . .M('LLEJ{ writen: 



Fsi'itilopJidvuicdr-y 



thrt'c L;t'iH'i:i 



(inr ;j,i'!ius 

 twi) genera 



as a svtiiiiisin nl 



I'luloinrdcs 



I'si'iahiplii/onit'dcs 



Hutidvnmi 



Sarsidla 



Aster ope 



Ci/dastcrofX'. 



TdriujiiiKidiin 

 I'aramekudoii 

 S/ reptoleberis 

 Sarsielln ,, ,, ,, ,, Eurypylus 



V Nemalohaninia 

 umber uf species. Ill liis sviioptic Work ill ,,D a s T i c r r e i r h", 1912, this uutlior records 155 recent 



species of this sub-order, 105 of which would be ,, certain", 50 ,, uncertain". Tlie proportion 

 between ,, certain" and ,, uncertain" is, however, much more in favour of the hitter category, 

 a fact that I have unfortunately been only too often reminded of during my study of this group. 

 After this work of G. W. MtJLLER some additional species, thougii only a few, have been incoi- 

 porated in tlic literature of this group. 



.\aturat system. Remarks: — Tlio difference between the two above-mentioned divisions of this sub-order 



worked out by G. 8. Bhadv — A. M. NoiiMAX and G. W. MOller is, as is seen at the first glance, 

 not profound. Two divergencies are to be noted. First G. W. Mt'LLER has removed the genera 

 PkUomedes and PseudoplMomedes from G. S. BltAU'i's and A. M. NORMAX's family Cyprklinidae 

 and of these has formed a new systematic unit, tlie sub-family Philomedinae, ranged with the 

 sub-family (Uipridininae, which includes all the remaining genera of the above-named family, 

 and with the sub-families tiarsiellinae antl Astewpinae. Secondly the same author has adopted 

 the genus Rutiderma in the new sub-family Philumsdvtae, which genus had formerly been 

 distinguished by G. S. Bhaio and A. M. NOR.MAN as a representative of a special family Ruti- 

 dennatidae, ranged with the ( ' y j) r i d i n i d s, S a r s i e 1 1 i d s and A s t e r o p i d s. 



Which of these divisions is preferable? Is any of them quite natural or is none at 

 all suitable to be accepted without alteration? 



A thorough study of the forms belonging here has led me to the following conclusions: 



The separation attempted by G. W. Ml'ller of the genera PhUomedes and Pseudophilo- 

 medes from the genera Cypridina, Pyrocypris, Crossophorus, Codonocera and Gigantocypris 

 is undoubtedly at least partly justified. The two first-mentioned genera are, as is clearly shown 

 by the descriptions given by G. W. MUller and by those I have worked out below, decidedly 

 opposed to the genera enumerated after them in so many respects that they must necessarily 

 be separated systematically from the latter. 



This, however, does not prevent the division given by G. S. Brady and A. M. NORMAN 

 fi-om having its advantages. The sub-families Cypridininae and PJiilomedinae arc, it is true, 

 well differentiated from each other, but on the other hand they are considerabl)' more closely 



