RECENT KESEAECUES ON THE EADIOLARIA. 173 



tidans. It must be borne iu mind, however, that the skeletons of 

 such forms are amongst those most likely to be preserved. 



Classification. 



The best classification yet offered is that proposed by Haeekel, 

 which is as follows*. He divides the whole group of Kadiolarians 

 into two sections : — A. Monozoa or Monocyttaria, the simple 

 forms, and B. Poltzoa or Polycyttaria, the compound forms. 



His Monocyttaria are subdivided into two sections : — A, a. 

 Ectolithia, with the skeleton external to the capsule; and A, h. 

 Entolithia, with more or less of the skeleton within the capsule. 

 The Ectolithia are further subdivided into (1) the Collida (with a 

 skeleton of scattered spicula or none), (2) Acanthodesmidaf, (3) 

 Ci/rtida, (4) Ethmosplicerida, and (5) AuIospJiW7^ida. His Ento- 

 lithia he subdivides into ((3) Ccelodendrida, (7) Cladococcida, (8) 

 Acanthometrida, (9) Diploconida, (10) Ommatida, (11) Spongurida 

 (with a skeleton wholly or in part spongy), (12) Discida, (13) 

 Lithelida. His Polycyttaria he divides into (14) Sphcerozoida 

 (skeleton absent or in the form of scattered spicula), and (15) Ool- 

 losphcerida (skeleton a perforated shell, surrounding the capsule). 



I think it would be very convenient, and therefore desirable, to 

 endeavour if possible to unite together these fifteen different 

 groups into large aggregations. Moreover, since Professor 

 Haeekel' s admirable monograph appeared, some new forms, the 

 curious form Myxohracliia amongst others, have been discovered. 

 Hertwig has also strongly insisted upon the greater importance 

 of the nuclear vesicle (the vesicula intima) as a classificatory cha- 

 racter than any characters which can be derived from the skeleton. 

 The aggregation or non-aggregation of zooids into colonies seems 

 to me a comparatively unimportant distinction, especially as 

 individual zooids of the compound species are found (however 

 derived) also in a single and separate condition. I cannot there- 

 fore but think the division of Eadiolarians into two primary 

 groups, the one single, the other compound, as an unnatural se- 

 paration. As to the possession or non-possession of a nuclear 

 vesicle (Binnenblaschen or vesicula intima), I quite agree with 

 Hertwig that it would form a most important distinction ; and I 

 should propose to adopt it provisionally, fully bearing in mind 

 that it may be found hereafter to be very widely, if not all but 

 * ' Radiolarien,' p. 237. 



+ The groups of which no characters are here given are characterized later, 

 having been adopted by me from Prof. Haeekel. 



