100 CLEAVAGE AND DIFFERENTIATION 



1/4 blastomeres which lack a portion of the organiser will go no 

 further than the germ-layer stage (unless a foreign organiser is 

 grafted into them). Something similar to the conditions in Am- 

 phibia is found in Amphioxus, where it has been shown that the 

 blastomeres are totipotent at but not beyond the 2-cell stage. ^ This 

 restriction is due to the localisation of chemo-differentiated sub- 

 stances necessary for mesoderm formation in the ventral meridian, 

 and of other substances necessary for notochord and neural plate 

 formation in the dorsal meridian. The fertilised tgg is thus bilaterally 

 symmetrical with regard to those chemo-differentiated substances 

 it contains ; and, since the first cleavage always occurs in the plane 

 of bilateral symmetry, the 2-cell stage is therefore the latest at 

 which the blastomeres can contain all levels of the main axis, and 

 therefore all these various substances (see below, p. 123). 



As already mentioned, the plastic stage of development, in which 

 regulation is still possible, comes to an end in Amphibia at about 

 the stage of mid-gastrulation. A similar state of affairs, though the 

 precise moment has not been so accurately determined, appears to 

 hold good for other vertebrates; e.g. in fish (Funduhis) defect- 

 experiments on stages prior to the formation of the germ-ring 

 (i.e. early gastrulation) give rise to defects in the size of the resultant 

 embryos.'^ Other experiments have shown that qualitative irre- 

 versible differentiation begins only when the embryonic shield has 

 reached a distinct size — i.e. some time after the beginning of gastru- 

 lation. In birds, it is known from experiments (see Chap, vi, p. 161) 

 in which an organiser is grafted beneath another blastoderm and 

 there induces the formation of neural folds that irreversible deter- 

 mination has not yet set in after 22 hours' incubation, but as no 

 interchange experiments have been performed, it is not known at 

 what stage determination of the various regions is definitely fixed. 

 In this connexion it should be mentioned that isolation experiments 

 demonstrate the "competence" (Waddington, 1932) to differen- 

 tiate into various structures, but they give no information as to 

 whether the power to differentiate into any other structures has 

 been lost. In mammals, nothing has as yet been experimentally 

 determined with regard to these points. 



^ Conklin, 1924, 1933. 

 2 Lewis, 1912. 



