16 SALENIA VAEISPINA. 



it from a morphological point of view when compared to the single large anal 

 plate of young Echini. Yet, while it may perhaps not he an embryonic fea- 

 ture of the Echini which runs back through the Echinoid series of the earlier 

 palaeozoic times, it may yet be one of those cases of the sudden reappearance 

 of an ancient structual feature after a long period of time, as is also perhaps 

 the five-valved actinal opening of Palseostoma. I am inclined to look upon 

 the suranal plate of Salenia as strictly homologous with the central plate of 

 the dorsal surface of Starfishes, which, while it recalls the Crinoidal affinities 

 of the Echini, yet has not played the important part in the development of 

 the Echinoid series which it did in the Starfishes or Crinoids. 



As far as the plates of the anal system are concerned, the embryonic t\ pe 

 of many plates arranged in more or less irregular concentric rows round the 

 anal opening, such as we find it in the oldest palaeozoic Echini, has remained 

 remarkably persistent throughout the group to the present day. Even in 

 the Spatangoids ami Clypeastroids, in which the anal system has become 

 disconnected from the apical system, the same general embryonic type has 

 been retained in nearly all the principal groups, with the exception of a few 

 types, such as Echinocidaris, and some of the Clypeastroids and Spatangoids, 

 in which the number of anal plates has become reduced, and they form 

 a pyramid over the anal system ; a structural feature, however, which we 

 should remember is already found in some of the earliest known Crinoids. 

 From what 1 have shown of the mode of breaking up of the anal system* in 

 some very large specimens of Arbacia, the pyramidal anal covering may have 

 been the earliest type of plates of the anal system in Crinoids, and the split- 

 ting up of the plates of the anal system, at first few in number, may be a 

 feature characteristic of the more recent Echini. Such a splitting up actu- 

 ally takes place in the growth of the plates of the anal system of Salenia. 

 In vi'iy young stages of S. Pattersoni we have a suranal plate and five anal 

 plate- covering the anal system; while in old specimens the anal system is 

 covered by a number of smaller plates, due to the splitting up in part of the 

 original plates at their apex, and in part to the formation of new plates 

 round the anal opening with the increase in size of the anal system. 



Bow far we are justified in considering the anal plates of the recent Des- 



mosticha as homologous with the dorso-central plate of young Starfishes and 



of Comatula, seems somewhat doubtful, P. II. Carpenter! has well discussed 



some of the difficulties to be met in adopting the view Loven and myself have 



* Challenger Echini, p. 56. t Quart. Journ. Mic. Sue, XVIII. 3,">7. 



