organisers: inducers of differentiation 139 



has never been acted upon by an invaginated organiser. Or we may 

 adopt the method known as explantation, in which the pieces of 

 blastulae, after being enclosed in epidermal jackets, are grown in 

 vitro in suitable media. DiiTerentiation of neural tube and of noto- 

 chord can be obtained in this way also, from tissue which has never 

 been acted upon by an invaginated organiser^ (see fig. 18, p. 49). 



There is therefore some determinative agency at work in ad- 

 dition to the invaginated organiser. The labile determinations thus 

 induced are presumably due to the transmission of stimuli from 

 the organiser before gastrulation, in relation to the main axes of the 

 tggy in a manner which will be considered below in connexion with 

 gradient-fields^ (see p. 310). 



In any case, it is clear that the labile determination of the 

 blastula stands in some relation to the bilateral symmetry imposed 

 upon the tgg at the moment of fertilisation. 



The action of the organiser, then, must be considered as taking 

 place in two phases. First, working as part of the gradient-field, 

 the organiser may be figuratively said to sketch out the presumptive 

 regions in pencil, and then, after invagination, the organiser goes 

 over the same lines with indelible ink. At the same time, the organ- 

 iser is capable of roughing out the sketch straightway in ink, with- 

 out any previous pencil work, as in those experiments in which 

 the organiser is grafted into the flank of another embryo. Neural 

 folds can arise from the pencilling alone, and from the inking alone, 

 and this duplicity of methods whereby neural folds can be formed 

 is another example of the principle of " double assurance ". 



But there is another point to notice here. When an organiser 

 is grafted into the flank of another embryo, the host-tissues are 



^ Bautzmann, 19296,0; Holtfreter, 1929 A, b. 



^ These examples have been mentioned in order to show that determination 

 and differentiation can take place in the absence of an invaginated organiser. But 

 several of these experiments introduce a new complication, since the tissue which 

 is differentiated in interplantation and in explantation frequently is of a nature 

 quite different from the presumptive fate of the region from which the piece was 

 taken. Presumptive neural tube material, for instance, has been found to differ- 

 entiate into notochord, muscle, mesenchyme^ and glandular epithelium (Kusche, 

 1929; Holtfreter, 1931A; Erdmann, 1931); presumptive epidermis can give rise 

 to neural plate, especially, for some unknown reason, when interplanted into 

 the coelomic cavity. Pieces of tissue from any part of the blastula have been 

 seen to differentiate into notochord and muscle (Bautzmann, 1929B) (see 

 P- 317)- 



