Ill XENO-INDUCTIONS 445 



experimentally induced acrencephalon, symmetry does nf)t seem to appear 

 spontaneously, but if the explant is conveniently shaped, it can elicit a sym- 

 metrical fore-brain. Thereby, it approaches the influence of the normal prechordal 

 plate, although the derivatives of the latter control symmetrization more strictly, 

 thanks to their orderly morphochoresis. For, by its kinematic behavior, the pre- 

 chordal plate behaves typically as a part of the middle layer. 



How many distinct agents of induction have been shown to exist in the xeno- 

 inductors? This evaluation can only be made by reference to the qualitative, 

 morphogenetic effect registred; this does not necessarily imply — -we shall come 

 back to this point — biochemical identity. It is a striking result of these investigations 

 that the number of agents distinguishable on this basis is remarkably small. An 

 early discovery, with amphibians, was that a neural tube induced its like in com- 

 petent ectoblast (Mangold and Spemann, 1927), that notochord, somitic meso- 

 blast and pronephros primordium all respectively induce their like (Holtfreter, 

 ig33a). These results seemed to suggest the possibility of organo-specific xeno- 

 inductions, brain by brain, eye by eye, lens by lens, kidney by kidney, but this 

 hypothesis had to be abandoned. The embryonic origin of the organs used does not 

 afford a valuable clue. If such mesoderm derivatives as kidney, bone-marrow, 

 spleen (see p. 446, note 2) are mostly notogenic, the acrogenic activity of the liver 

 does not correspond to any developmental relation, and moreover the inductive 

 effect of one same organ may vary if taken from different species. After the elimi- 

 nation of unadequate ideas, the distinction between an acrogenic and a notogenic 

 agent has become generally accepted, a concept in excellent agreement with the 

 experiments on the major developmental inductors. In addition, some frac- 

 tionations performed on material rich in the latter agent succeeded in isolating 

 samples which induces electively either deutomerit structures or the (chordo-) 

 mesoblast, or the neural tube. Here the correlation with normal development 

 must be considered with caution, for direct work on the embryos has not yet per- 

 mitted such dissociations. On the contrary, observational as well as experimental 

 data speak for a close relation, if not for a complete identity, between the 

 inducing factors acting inside the middle layer and the ones evocating the 

 neural plate. In the instances where opposite results were obtained with the 

 xeno-inductors, we may have been dealing with some kind of experimental 

 artefact, which could be explained in several ways. It could well be, for example, 

 that the deutogenic fraction would be a variety of notogenic agent completely 

 absorbed by a relatively small territory of ectoblast, in which its action would be 

 maximal. For the selective induction of (chordo)mesoblast, the explanation could 

 be that the agent is completely incorporated in the deeper layer of proliferating 

 ectoblast and is not transmitted to its surface layer. In the case of exclusive neural 

 plate induction, no inwards proliferation would be provoked, and the epithelial 

 layer would, therefore, be directly neuralized. Of course, these ideas are guess- 

 work, only proposed to show that the multiplicity of specific agents is not neces- 

 sarily transposable to the normal mechanisms. 



For the relation between the notogenic and the acrogenic agent, there are 

 several instances where the former can be converted into the latter by an appro- 

 priate denaturation. The opposite conversion has not been obtained by physical 



Lileralure p. ^H;; 



