442 Prof. E. W. MacBride. Some Points in the [May 6, 



posterior end of the animal, whilst the other two are the rudiments of the 

 postero-lateral ciliated arms, which in this species attain an extraordinary 

 length. 



The interior cells of the morula are now seen to be the skeleton-forming 



o 



cells of the calcareous spicules which support these arms. The ccelom arises 

 as a vesicle from the apex of the archenteron, and shortly afterwards the 

 stomodaeuni makes its appearance in the centre of the anterior surface. The 

 tongue of cells projecting into the archenteron is gradually absorbed. 



Until 1905 no doubt existed in my mind that the course of development 

 above described was the normal one ; but in that year, in addition to the 

 ordinary methods of artificial fertilisation, the device was tried of placing 

 males and females together in half-gallon jars filled with pure sea-water and 

 allowing them to spawn naturally. A vigorous culture was obtained in this 

 way, and the development differed in some important particulars from the 

 type described above. 



Segmentation results not in the formation of a morula, but of a thick- 

 walled blastula (fig. Id), one side of which becomes specially thickened and 

 buds off mesenchyme cells. The embryo now escapes from the egg-membrane 

 and becomes a larva. Soon at the pole from which the mesenchyme was 

 produced a regular invagination takes place, with no trace of the tongue 

 of cells described above. 



Mesenchyme cells continue to be budded off from the apex of the growing 

 archenteron. At the same time a conspicuous feature makes its appearance 

 at the animal pole of the larva. This is a crest of vacuolated cells (fig. le), 

 which has a triangular shape, and seems to be an apparatus for the floating 

 of the larva, which lasts until the ciliated arms are developed. 



The coelom next makes its appearance as a thin-walled sac, greatly 

 resembling in shape the coelomic rudiment in the larva of Asterina gibbosa 

 (fig. 1/), and at the same time the rudiments of the postero-lateral arms appear. 

 The vacuolated crest diminishes slowly in size ; when it first appears its 

 height is as great as the entire length of the larva, but by the time the 

 stomodaBum appears it has disappeared. Previously to this the coelomic 

 rudiment has become separated from the gut as a single vesicle. This then 

 divides into right and left halves, and the stomodaeum joins the archenteron, 

 which now shows beginning of the constrictions which mark it off into 

 oesophagus, stomach, and intestine. 



Pausing for a moment to reflect on the meaning of the two types of develop- 

 ment described above, it is clear that the difference between them cannot be 

 due to the fact that in the one case the genital organ was dissected out, and 

 in the other case that the animal was allowed to spawn. In some cases larvae 



