51 



lacral plaies. In Lankester's „Treatise on Zoology" III (1900) Gregory has made 

 a most interesting attempt at a more natural classification. Tha Palæechinoids are 

 no more regarded as a separate subclass equivalent to all other Echinids, but they 

 are united with the regular Echinids and, together with the Cidaridcv , classed 

 as Regiilaria Evdobranchiata. I think this a very important step towards a 

 natural classification, though I do not quite agree with Grkgoky in the arrange- 

 ment of the families. It seems to me rather bold to place the Cidaridce and Lepi- 

 docentridæ in one order. But that the Archœocidaridœ are closely related to the 

 Cidaridce seems, indeed, very probable, the only important dilTerence, as far as we 

 know, being the pluriseriate arrangement of the interambulacral plates in the 

 Archœocidaridœ — and this character even is not quite exclusive, there being four 

 series of interambulacral plates in Tetracidaris, which can by no means l)e classed 

 with the Archœocidaridœ. I think it right then to follow Gregory in uniting the 

 Archœocidaridœ and the Cidaridœ into one order, Entobranchiata^), against another 

 order: Ectobranchiata, comprising the other regular Echinoidea, except, perhaps, 

 some of the Palœechinoidea. Gregory is scarcely right in assuming that none of 

 the Palœechinoidea had outer gills; but as I have no very extensive knowledge 

 (and only literary) of the Palæechinids I shall not enter into a more detailed dis- 

 cussion of this problem. 



The family Echinothuridœ Wyv. Thomson is adopted by Ag.\ssiz in the 

 ,,Challenger"-Echini, though it is pointed out that its difference from the Diade- 

 niatidœ is very slight, „some of the species of Echinothuridæ here described show(ing) 

 that some of the important characters upon which this family is distinguished 

 from the Diadenmtidœ may become gradually obliterated". In some Echinothurids 

 „the lapping of the plates is reduced to a minimum, if it exists at all", while on 

 the other hand in Asiropyga the lapping of the plates is very distinct, which genus 

 therefore must be considered „a genus either belonging to the Echinothuridæ or at 

 any rale possessing some of the most characteristic features of both the Diadema- 

 tidæ and Echinothuridæ". (Op. cit. p. 72.) These facts pointed out by Agassiz cer- 

 tainly show that the lapping of the plates is no exclusive character of the Echino- 

 thuridœ, but they do not at all show the family to be untenable. The structure of 

 the ambulacra and the continuation of the ambulacral plates on the buccal mem- 

 brane are characters so distinct that there can be no doubt at all of the Echino- 

 thuridœ forming a very distinct family. Also the structure of the actinal spines 

 wliich end in a distinct hoof or a thick bag of skin seems to be a very exclusive 

 character; it is to be remembered, however, that we do not know, how the actinal 

 spines of Kamptosoma terminate. With regard to pedicellariæ and spicules no 

 characters of importance can be pointed out as distinguishing the Echinothuridce 

 from the Diadematidœ. Duncan and Lambert (and recently Meissner) place the 



') The name Holostomata cannot be used, as also the Echinothurids are holostoraatous. 



T 



