1128 THE VOYAGE OF H.M.S. CHALLENGER. 



podoconus, and the peculiar shape of its nucleus. He also published excellent figures 

 of some interesting new species. 



0. Biitschli, 1882, in his valuable paper entitled: " Beitrage zur Kentniss der 

 Radiolarien-Skelette, insbesondere der der Cyrtida" (Zeitsehr. fiir wiss. Zool., vol. 

 xxxvi. p. 485) made an attempt at a natural classification of the Cyrtida, which he de- 

 rived from the S p y r o i d e a or Zygocyrtida. As already mentioned above, we cannot 

 accept this essay as the foundation of a true natural system, since the affinities of 

 the Cyrtellaria (and of the Nassellaria as a whole) are far more complicated and 

 difficult than Biitschli supposed. His views were supported by accurate observations 

 only on the structure of the fossil Cyrtoidea of Barbados ; these, however, represent 

 the minority only of the genera, and many interesting and important forms (mainly of 

 true " Monocyrtida ") remained unknown to Biitschli. A great part, however, of his 

 observations are very useful, and his remarks on comparative morphology are very 

 suggestive. 



The Cyrtoidea may be divided into families and subfamilies according to three 

 diff'erent principles, viz., (l) the number of joints into which the shell is divided by 

 transverse strictures ; (2) the number of radial apophyses which arise from the shell ; 

 (3) the shape of the basal mouth, which is either open or closed by a lattice-plate. 

 At present every attempt of classification in this large group must be more or less 

 artificial, since the afl&nities of the numerous smaller and larger groups are extremely 

 complicated, and the ontogeny, the only sure guide in this phylogenetical labyrinth, is 

 perfectly unknown. It seems therefore the most convenient to employ for our 

 artificial classification, first, the number of shell-joints, second, the radial structure, and 

 third, the shape of the mouth. 



A. The number of joints into which the shell is divided by transverse constrictions, 

 serves here for the distinction of four primary groups or suborders of the Cyrtoidea, 

 viz., (1) Monocyrtida with one joint ; . (2) Dicyrtida with two joints ; (3) Tricyrtida 

 with three joints ; and (4) Stichocyrtida with four or more joints. In my Prodromus 

 (1881, p. 426) I divided the latter group into Tetracyrtida (with four joints), and 

 Stichocyrtida (with five or more joints); but these two groups may be united, since the 

 fourth and all the succeeding joints are of rather indifi'erent shape and of little morpho- 

 logical value. The three first joints, however, are usually very diff'erent and possess a 

 high morphological importance, so that we distinguish the first joint as cephalis, the 

 second as thorax, and the third as abdomen. The uppermost transverse constriction, 

 which separates the two first joints, cephalis and thorax, is the collar stricture and is 

 usually caused by an internal fenestrated septum, the cortinar septum. The second con- 

 striction, which separates the second and third joints (thorax and abdomen) is called the 

 lumbar constriction. The followmg constrictions (in the Stichocyrtida) are indifferent 

 and of little morphological interest, and require therefore no peculiar designation. 



