PHALLUSIA MAMILLATA 399 



lies on this side. If this were really a typical gastrula it would 

 be necessary to suppose that the original polarity had been 

 turned so that what was at one time the transverse axis had 

 become the dorso-ventral axis. All the evidence available shows 

 that such a change of polarity never occurs. But there is also 

 direct evidence that these apparently typical gastrulae are not 

 really such. The typical gastrulae are bilaterally symmetrical, 

 these ^ gastrulae are not, but invariably lack the substances and 

 cells of the missing half; the typical cup-shaped gastrula gives 

 rise to a typical elongated gastrula and this becomes a typical 

 larva, the | gastrula does not; in some cases in which the cells 

 have been prevented from growing in over the injured side the 

 gastrula derived from | of an egg may be seen clearly to be ^ of 

 a typical gastrula, but there are fewer of these cases in Phallus' a 

 than in Cynthia, owing perhaps to the fact that the protoplasm 

 of the egg is more labile and the blastomeres grow in over the 

 njured side to a greater extent in the former than in the latter. 



What has been said of the cup-shaped ^ gastrula is true also 

 of the elongate ^ gastrula; so far as I have observed it is never 

 typical and entire, though at first sight it may appear to be so. 

 However a careful study of fixed and stained specimens shows 

 that these gastrulae are not bilaterally symmetrical but that 

 they lack the muscle cells and other parts of the missing half. 

 Usually these forms are well rounded and are covered with ecto- 

 derm cells on the side next the injured blastomere, but in some 

 cases the endoderm remains uncovered on the injured side, as 

 in fig. 3, and such forms are plainly | of a typical gastrula. 



All that has been said of the partial character of the gastrulae 

 derived from ^ blastomeres, applies with still greater force to those 

 derived from anterior or posterior f blastomeres, or from i blasto- 

 meres. Such embryos are never entire, so far as I have observed, 

 and they are even less typical than are | gastrulae. 



3. Larvae from partial eggs of Phallusia. The crucial test of 

 the potency of single blastomeres of the egg of Phallusia is to be 

 found in the kind of larvae which develop from such blastomeres, 

 for it is possible that regulation might be incomplete in cleavage 

 and gastrula stages, but complete in the larva. For this reason 



