414 SCIENCE PROGRESS 



ridea) of the extremely heterogeneous and diverse assemblage 

 of Cystids. Aristocystis has probably undergone some sim- 

 plification and reduction, especially in the matter of the absence 

 of arms and ambulacra. Yet it still shows primitive character- 

 istics in the repetition of similar parts (which is capable of 

 inducing exuberant variability), e.g. in the numerous plates with- 

 out definite orientation which form the sac-like body, with an 

 undifferentiated and indefinite porous structure of the stereom. 

 Even the Cambrian Trochocystites among the Carpoidea shows 

 some specialisation in the ring of large plates encircling the 

 planoconvex calyx, which is otherwise composed of an indefinite 

 number of small plates (larger, however, on one side than on 

 the other). Here a primitive bilateralism has been retained 

 and emphasised, and although the Carpoidea are stalked forms, 

 yet radial symmetry has not resulted from the adoption of 

 a stem for a sessile habit. Indeed, since the short, pointed 

 stem is hollow and plated it is not improbable that it was 

 adapted for a balancing or directive function rather than for 

 the purpose of fixation. 



As soon as a stalked habit had become general a tendency 

 towards pentamerism must have rapidly made itself felt, for 

 as I have already explained (p. 401) with regard to the prevalent 

 pentamerism in flowers, it can be derived for geometrical 

 reasons from the exigency of combining economy of mate- 

 rial with the attainment of the greatest amount of enclosed 

 space. 



A progressive reduction and specialisation can be traced in 

 the Cystids ; among the Diploporita, Glyptosphaerites shows 

 a distinct advance in type upon Aristocystis in the pentamerism 

 of its ambulacral grooves. Its polygonal plates, however, are 

 still numerous and irregular and the diplopores are irregularly 

 arranged. This is also the case in the earlier Dichoporita as 

 exemplified by Echinosphaerites. In this group the pores are 

 definitely arranged in pore-rhombs, and it contains transitional 

 forms leading up to the Blastoids and Crinoids, both of which 

 afford clear instances of a rapid reduction from the indefinite 

 number of plates of the early Cystids to a strict pentamerism 

 of the calycal plates. The Blastoids have been shown b}?- 

 jaekel ^ to be derived from the Dichoporite division of the 

 C^^stids by an ever-increasing regularity in the orientation of 



' Shim fnesgeschuhie dcr Pel >naiozoe?i,^\.. i. 1899. 



