FIELDS AND GRADIENTS IN NORMAL ONTOGENY 313 



(p. 60). The rule appears to be of general application for the 

 developing egg. 



The overriding or abolition of the original polarity by external 

 agencies appears seldom to be obtainable with eggs ; but some re- 

 markable cases are known from Echinoderms. We have already re- 

 ferred (p. 83) to the fact that in developing fragments of Lytechinus 

 and Patiria eggs, which have been obtained by cutting before fer- 

 tilisation and subsequently inseminated, the first two cleavage 

 planes are always at right angles to the plane of the cut. 



Subsequent development demonstrates that even more radical 

 changes have been effected. When the gastrulation of the fragments 

 occurs, it invariably takes place at the centre of the cut surface, and 

 at right angles to it. The polarity of the developing egg-fragment 

 and the axis of the resultant larva is therefore determined in re- 

 lation to the cut, and not in relation to the original polarity of the 

 whole tgg.^ 



It is to be supposed that the operation intensely stimulates the 

 cut surface, and that the resultant increase of protoplasmic activity 

 grades away across the fragment. The activity-gradient thus pro- 

 duced must be able to override the original gradient within the 

 fragment. 



This is also stated to occur in the California species of Para- 

 centrotiis. However, in the European Paracentrotus lividiis, meri- 

 dional halves of the Qgg produced by isolation of the 1/2 blasto- 

 meres appear to retain the original polarity.^ It is to be noted that 

 in this case no raising of activity by cutting has occurred; the 

 separation also took place at a later stage. Thus the observations on 

 the two forms are not necessarily contradictory. 



In P. lividus also, marked deformation as a result of centrifuging, 

 however, is incapable of altering the original polarity. The point at 

 which gastrulation is initiated is always at the original vegetative 

 pole, as indicated by its relation to the subequatorial pigment- 

 band. Thus gastrulae are produced which may be extremely 

 elongated, flattened, or obliquely deformed in the animal-vegetative 

 direction^ (see also Chap, iv, p. 69). 



^ Taylor, Tennent, and Whitaker, 1925. 



2 Horstadius, 1928. 



3 Harvey, 1933. 



