65 



at that lime, they were preserved, no attempt beiiifj marie to have them 

 metaniorpliose compieleiy. Wiiether the iarva will tal<e so long time to 

 develop under belter conditions (more food, more coni^einal temperature?) 

 must be left undecided. 



The first larval stage (IM. XI, Fig. 1) shows the typical form, olTering 

 no marked special features. The body is short, the arms rather long. 

 There is a very distinct suboral cavity. Pigment very scarce; a few 

 liigment cells, mainly in Ihe poinl of the anterolateral arms. The skeleton 



Fig. 24. .Skeleloii of larvii of Ileliocidaris (uberculala; .\. of young larva, sifle view: B. of 

 fully fornu'il larva, froTil view. --^/,. Letters as in Tig. 2(1. 



(Fig. 21 A) is very complicate. The body skeleton forms a basket structure, 

 but not in the usual simple way; there are two recurrent rods, so that two 

 large meshes are I'ormeel in each side of the body. In the end of the body 

 there are three transverse connecting rods, all strongly thorny along the 

 hind edge; also the body rod and the recurrent rods are rather strongly 

 thorny. The postoral rod is fenestrated, with many thorns along the outer, 

 fewer sometimes none - along the inner side. 



The fully formed larva (PI. XI. Fig. 2) is provided with four vibiatile 

 lobes, but there are no vibratile epaulets. The anal area is somewhat 

 concave. .\t the anterior edge there is a pair of small ciliated protuber- 

 ances. The vibratile band is arranged along the dorsal side so as to form a 

 pair of small lobes. The arms are fairly long, narrow; tin- postero-lateral 



