&ENERA. AND GROUPS OF THE ECHINOIDEA. 29 



clubbed, and large at the blunt end ; fusiform, sharp, or blunt ; 

 cylindrical and blunt, with spinules, thorns, and granules or 

 disk-shaped espaasions, or cup-shaped at the end, with spines 

 radiating ; or long, flat, expanding, furrowed and spined. 



Tentacles disciferous actinally, and gradually becoming non- 

 prehensile and branchial. Pedicellari^ large, tridactyle, gemmi- 

 form, and small and blunt-headed. 



Sexes separate or not ; young usually undergoing metamor- 

 phoses, rarely not, and then found upon the parent. 



Fossil. Permian: Europe. Trias: Europe, Asia *. Jurassic to 

 Post-pliocene : England and Europe, N. Africa, Asia. Cretaceous 

 to Tertiary : JN". America, Egypt, and W. Africa. Tertiary : 

 Australia. 



Hecent. World-"wide. 



Every writer upon the classification of the Echinoidea since Desor has 

 complained of the unsatisfactory attempts of some of the most distin- 

 guished authorides to subdivide the genus Cidaris. The subdivisions 

 gradually became subgenera ; and of late these have received generic 

 importance. The divisions were made upon very unimportant external 

 characters ; and subsequent research has proved that these structures, the 

 variations of which led them to be considered of good diagnostic value, 

 are of no physiological importance!. The presence or absence of perfo- 

 ration and crenulation of the primary tubercles, and the connection of the 

 pores of pairs by a groove, or their disconnection by a granule or swelling, 

 have really been the main features relied upon to establish divisions, sub- 

 genera, and finally genera. Common observation will satisfy anybody 

 that crenulation is not invariable upon the same test in many instances ; and 

 it is a mistake to believe that it is a structure which relates to the strength 

 of the spine-muscles. These are attached to the edge of the scrobicule, 

 and are inserted just below the milled ring of the spine ; and they have 

 nothing to do with the crenulation. This appears to add to the attach- 

 ment surface of the membrane of connective tissue which closes the ball- 

 and-socket joint of the spine and tubercle. 



The grooving, or the reverse condition, between the pores of a pair is of 

 no physiological importance whatever ; and it is frequently impossible to 

 decide whether the pores are connected by a groove or not. 



Any classification in which these characters are used is artificial. 



The number of interradial coronal plates is of physiological importance ; 



* Eocidaris = Cidaris has been found in the Salt Eange, Eritish India, 

 t Duncan, 1888, Ann. & Mag. Nat. Hist, vol. i. p. 124. 



