WYVILLE THOMSON, ON SYNAPTA INH^RENS. 139 



the ring and tentacles by contracting suddenly into a short 

 flask. 



Nearly opposite the Polian vesicle, on the other side of the 

 ring, the madrcporic tubercle gradually increases in size and 

 distinctness. It appears at first as a greyish granular hood, 

 capping, apparently, a very short branch from the ring, Avhich 

 takes a sort of curve in the hood, and to which the hood acts, 

 I should think, as a valve (PI. V, fig. 3, /) . A delicate canal 

 goes through the valve, and I have frequently observed a 

 current of Avater passing in through it; sometimes I could 

 not detect any current, but 1 never saw one passing out. The 

 madreporic tubercle is at first imbedded in the sarcode of the 

 outer wall, and I believe the canal then communicates directly 

 with the water outside. Afterwards the tubercle becomes 

 separated, apparently by the liquefaction of the sarcode round 

 it, and hangs in the body-cavity. In the most advanced in- 

 dividuals Avhicli I have examined, there is no appearance of 

 any calcareous deposit in the madreporic tubercle. The am- 

 bulaeral vessels contain a perfectly colourless fluid, slightly 

 opalescent, with minute molecules held in suspension ; large, 

 structureless granules likewise float in the liquid, and with 

 them bodies of a very peculiar and definite form. These are 

 globular masses of transparent granules, and have stifi* cirri 

 standing out from their surface. They are large, and not 

 very numerous. The interior of the vessels seems to be cili- 

 ated ; at least, one would imagine so from the motion of par- 

 ticles within, and the cirrated granules course round the ring 

 and into its csecal processes with great rapidity, sometimes 

 passing through a valve into a tentacle, whirling round its 

 cavity till the valve opens to let them out, popping into the 

 Polian vesicle, which is sometimes half stufted with them, 

 apparently in a state of hopeless congestion, till a rapid con- 

 traction of the vesicle drives them all out again into the 

 general circulation. The five delicate, almost rudimentary am- 

 bulacral vessels passing down from the lower surface of the ring 

 Avithin the longitudinal muscles, are now very distinct. 



Immediately above the origin of each of the ambulacral 

 vessels, above the junction of the simple ends of each pair of 

 spiculse of the oesophageal ring, a pale yellow crescentic gra- 

 nular nervous ganglion (PI. VI, fig. 1, h) gradually acquires form 

 and consistency. In an individual a month old (PL I, fig. 2), 

 the ganglia are very distinct, but no trace of nervous filaments 

 can be detected proceeding from them. No great change 

 occurs in the structure of the alimentary canal. The oeso- 

 phagus becomes more defined and muscular, and the whole 

 canal increases in length, at last forming a partial convolution 



