34 THE VOYAGE OF H.M.S. CHAI.LENGER. 



forms with narrow ambulacral areas and long slender serrated spines, while Phijllacanihus 

 [this subdivision, as I have limited it, is equivalent to Leiocidaris and Rhahdacidaris of 

 Desor] would include species with broad ambulacral areas, having the poriferous zones 

 joined by a furrow more or less distinct ; while Cidaris proper would be restricted to 

 species in which the pores of the poriferous zone are not so connected. But thus 

 far no characters derived from the many species described, either fossil or recent, can be 

 given to define these sub-genera with any accuracy. The genus Cidaris has since the 

 Triassic period been represented uninterruptedly by a large number of species, and as 

 far as the radioles are known, while some of the types seem characteristic of the 

 Jurassic or Cretaceous, yet from what we know of the extraordinary variations in the 

 spines among the recent species, they hardly furnish a safe guide for any subdivision. 



In fact, the species of the genus Cidaris, like those of many other genera, present 

 soon after their appearance an extensive series of variations, showing an extraordinary 

 degree of plasticity, which has gradually diminished somewhat as we pass from the 

 Jurassic to the Cretaceous, the Tertiary, and finally the recent species. Among the 

 latter we still find all the principal types of radioles represented, with the exception of 

 the round-headed acorn-shaped radioles, like those of Cidaris glandifera, which appeared 

 with the Triassic Cidaris and died out during the Cretaceous. This is thus far the 

 only type of radioles of Cidaris of which no analogue exists among the species still 

 living at the present day. 



The family Cidaridse is liy no means so strictly circumscribed as it would seem 

 from an examination of the living forms alone, and when we come to intercalate such 

 forms as Acrosalenia, Pseudocidaris, Hemicidaris, and Pseudosalenia, wdth Tetracidaris 

 and Diplocidaris, we find afiinities developing among the genera allied to the Cidaridas, 

 pointing on the one side to the Cidaridse proper which preceded them in time, and to the 

 Pseudodiadematidae on the other side, which have such an extraordinary development in 

 the Cretaceous formation, the Cidaridae types developing into the small groups of Salenidae 

 and of Cidaridaj proper which have persisted to the present day ; and the Pseudodiade- 

 matidse type gradually disappearing and being represented at the present time only by 

 the Phymosoma group, and not developing into the Diadematidae proper, which are 

 evidently the successors of the Perischoechiuidae or the Echinothuridse of the Chalk ; 

 though the structure of the abactinal and actinal systems of some of the Palaeechinidae, as 

 wiU be seen hereafter, points to a far closer affinity between them and the Cidaridae than 

 has been hitherto acknowledged. In the descriptions of the young Goniocidaridae we 

 cannot fail to be struck with their similarity to the St Cassian Cidaridae, and the tests 

 of the diminutive Cidaridae of the Trias show at a glance embryonic features, such as 

 the young of all Cidaridae have, which were at that epoch characteristic of the whole 

 group of Cidaricte. 



The small Cidaridfe of the St Cassian are the most perfect emliryonie Cidaridae imagin- 



