BIHANG TILL K. SV. VET.-AKAD. HANDL. BAND 18. AFI). IV. N:0 1. 31 



earlier, in palpeozoic time, certain genera of the Perischoeehinoi- 

 dea, sucli as Melonites or Lepitlestes, exhibited a structure 

 of the anibnlacra in which we perhaps onght to see a prece- 

 dent of that now obtaining in the Echinothuridce. and vvhich 

 highly deser\es a closer study: a snbmesial large plate and a 

 rather irregnlar series of nuraerons smaller plates. On the 

 other hand, the opposite imbrieation of tlie two eoronal sy- 

 stems lias descended also to the genus Astropyga and others 

 among the Diadematidre, and survives in the recent Spatan- 

 gidcf, of wliich the lieteropodons anibnlacra, the position of the 

 spheridia, the perforated mamelons, the petals, and the variforni 

 spines suggest a distant affinity to that gronp. a fact of which 

 LiNNiEUS already had a presentiment.* 



In the great majority of the Ectobranchiates the disintegra- 

 tion of the peristome is, as we have seen, much more effioient 

 than in the Cidaridse and the EchinothuridtP, the membrane 

 being left almost naked, and when the calcareous substance 

 is redeposited there, it is under the form of corpuscles much 

 reduced in si7>e, irregnlar in shape, and devoid of pores and 

 pedicels, Pl. XII fig. 15H—160. These corpuscles may be dis- 

 tinguished as of two kinds: spicules more or less submerged 

 in the tliickness of the buccal membrane, and others, lamellar 

 Hakes rising wholly or in part above its surface and armed 

 with spinnles and forcipes.. Diiferent in shape as all these are 

 from any other eoronal element, their mode of distribution 

 betrays their derivation. In front of ever}' iixed ambulacrnm 

 they mark ont more or less distinctly an area representing"^ 

 the site of its dissolved plates, the wrecks, as it were, showing 

 the way of the retreating peristome. Their adoral ends ad- 

 join the pair of persisting primary plates. Other areas are 

 seen in front of the interradia exhibiting other marks; but 

 the ambulacral areas are more conspicnous and more advanced. 



The affinities of the recent Diadematidse to the Astheno- 

 soma are distant enough, less so. however, than those of any 

 other gronp. Accordingly the arrangement of spieules and 

 laminee in their bnccal membrane may be expected to have 

 some closer resemblance to that of the Asthenosoma than that 

 of any other form. So it is in reality. In the Diadema saxa- 

 tile L.,'- VI. XII, fig. 153, every primordial pair of ambula- 



' See my paper on the Echinoids described by Linn^us p. l;ll. 

 2 1. c. p. 185. 



