68 



The gastrulation, which proceeds in the usual way (PI. XVII. Figs. 1 2) 

 begins at the age of about 18 hours. The blastoporus is sometimes pushed 

 out towards the side (PI. XVII. Figs. 2 3); this is, however, hardly any- 

 thing but an individual variation; generally it is in the middle of the 

 hind end of the body, and in this place a trace of it is generally found 

 in the more advanced stages (PI. XVII. Figs.5 6; PI. XV 1 1 1. Fig. 7). The 

 gastrula? are not bound to the surface, but swim freely in the water, with 

 the usual rotating movement, the aboral end always being turned up- 

 wards, as a necessary consequence of the fact that the vacuoles containing 

 the light, probably fatty substance are in the main confined to this part 

 of the body, and it remains so during the whole course of the develop- 

 ment. It seems beyond doubt that the substance contained in these 

 vacuoles forms the nourishment on which the embryo subsists until the 

 young sea-urchin has got its mouth and is able to feed actively; the 

 embryo itself is perfectly unable to feed, the gastrula mouth 

 closing very early, sometimes already at the age of only 30 hours (PI. XVII. 

 Fig. 7) and no larval mouth being formed. 



In embryos 30 hours old the archenteron begins to difl'erenliale (PI. XVII, 

 Figs. 3 4), its upper end widens, and gradually the widening pushes itself 

 downwards, the wall of the archenteron making a fold on one side. The 

 pouch thus formed, representing the rudiment of the hydrocoel, is still 

 in open connection with the archenteron, but in embryos of 42 hours it 

 has been separated from it (PI. XVII. Fig.6; PI. XVIII. Fig. 7). The forma- 

 tion of the enterocoel cannot be made out in the material available. Jud- 

 ging from this material it would appear to originate as schizocoel spaces 

 in, not as pouches from the entoderm; but I do not venture to ascertain 

 as a fact that it does originate in this exceptional way. A rich and 

 most perfectly preserved material would be needed for giving full evidence 

 of such a surprising statement, and my material cannot be said to be of 

 such eminent quality. Also the question of the formation of the hydro- 

 pore must remain unsettled; in fact, I have been unable to find the 

 slightest trace of it, and it would appear that also in this regard the 

 development of this species is exceptional among Echinoderms. (Comp. 

 below, sub Peronella Lesiieuri). 



As already stated the blastoporus may close completely already at lin- 

 age of 30 hours (PI. XVII. Fig. 7), the archenteron completely separating 

 from the ectoderm. In most cases, however, the separation is not com- 

 plete, a fine strand remaining as a connection between the entoderm and 

 the ectoderm (PI. XVII. Figs. 5 6). Within the cavity of the archenteron 

 a distinctly limited mass of a substance is generally seen which stains very 

 strongly in hematoxylin or safranin; it may perhaps be an agglutinated 



