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



Discoideae and the Clypeastroids, and compare them to the true Spatangoids, it seems 

 impossible any longer to lay stress upon the characters which have mainly led palae- 

 ontologists to adopt these two great primary divisions. As has abeady been pointed 

 out by Lov(^n, the variations introduced are greatest on the actinal surface, and greatest 

 in the posterior region of the test. In the Echinoidea the first trace of such a want 

 of symmetry on the actinal side is found in the Echinometradae, while on the abactinal 

 surface the position of the anal plate in certain genera of Echinids, especially of the anal 

 plate in the Salenidae, indicates very early the tendency to an asymmetrical development 

 which culminates in the Sj)atangoids of the present day. The next stage in this 

 asymmetrical development is due to the exclusion of the anal system from the 

 apical system, and the corresponding marked distinction at once existing among the 

 Discoideae in the arrangement of the plates of the odd posterior interambulacral area, and 

 the compact apical system thus formed, which becomes the basis of the subsequent 

 modifications it undergoes on the one side in the Clypeastroids, from the soldering of the 

 plates by the spreading of the madreporic body, and the exclusion of the genital openings 

 from the apical system, and their appearance in the apical part of the interambulacral 

 region ; and on the other either into the compact apical system of the recent Spatan- 

 goids or the disjunct apical system of the CoUyritidae. 



In the Clypeastroids the asymmetry of the plates is almost entirely limited to the 

 posterior interambulacral area, and the further extensive development of the coronal 

 plates of the test which is limited to the whole of the actinal surface, and not to a single 

 area as in the Spatangoids proper. 



Passing from the Discoideae to the Cassidulidae we find there a further modification of 

 the actinal surface from that of the former group extending to the actinostomic plates 

 which still exists at the present time ; and the first trace also of a more distinct petaloid 

 system than we find among the CoUyritidae. The asymmetry of this group is limited to 

 the actinal surface, and to the odd interambulacral area. If we further examine the 

 older Spatangoids we find that in all the more globular genera, such as Hcmiaster, there 

 is a marked uniformity in the size and number of the coronal plates ; though by no means 

 so great as that in the CoUyritidae or such genera as Galerites and the like, of which the 

 deep-sea genera belonging to the Pourtalesiae, such as Cystechinus and similar forms, 

 may be considered as the representatives; and thus we little by little pass from genera in 

 which the actinal plastron differs only slightly from the other interambulacra, Cassidulus, 

 Holaster, Pourtcdesia, Genicopatagus, and Hemiaster to the genera with a more marked 

 actinal plastron, such as Palceostoma, Agassizia, and Spatangus, and finally to Meoma, 

 Brissojms, Ecliinocardium, Brissus, &c. It by no means follows, however, that the 

 specialisation of the ambulacral petals has kept pace with this elongation of the actinal 

 plastron, as can readily be noticed on comparing such widely-differing genera as Echino- 

 cardium, Spatangus and Schizaster, which are all characterised by this highly-developed 



