Ill INDUCTION GENERAL REMARKS 473 



of two different but genetically related inductors is a progressive, time related event. The 

 conversion process P to A explains the origin of the powerful prechordal plate, but it is 

 necessarily not limited to this structure. Several experiments suggest that the cephalic 

 part of the dorsal chordomesoblast or even the whole roof of the archenteron produces the 

 acrogenic agent, but at a decreasing rate from head to tail; therefrom the apparently 

 caudo-cephalic gradient Z' as a secondary feature. It could even be imagined that a head- 

 ward flow of ^ products gradually reinforces or feeds the inductive yield of the prechordal 

 anlage, but positive data for this idea are not yet available. Its only real justification is 

 that it would give a functional meaning to the nearly generalized P—>A transformation. In 

 any case, a score of experiments demonstrate the relatively labile state of the P-tendency 

 and its inclination to shift to the more stable ^-condition, while no pertinent example of 

 the opposite process has been observed. Only early centrifugation and Li-action could be 

 thought to cause this reversal; however, it is much more satisfactory to consider that these 

 treatments either inhibit, stop, or moderate the transformation P —* A,an interpretation sup- 

 ported by their depressive effect on the P-tendency itself. 



It is immediately apparent that these four theses have several points in common. The 

 difference between the proposals of Yamada and Nieuwkoop is mainly that the former 

 does not insist so strongly upon the succession of two neuralizing agents. On the other 

 hand, his conception could easily be made compatible with ours, since a dorso-ventral 

 decrease of potential is also at the basis of my interpretation. I also agree with his "caudo- 

 cephalic" gradient but emphasize in addition that it is established secondarily. However, 

 I do not refer to abstract mediators, but rather attribute acrogenesis to the secondarily 

 evolved prechordal plate, and notogenesis to the primarily present chordo-mesoblast. 



At present, the main difficulty is to decide whether or not the necessity of the two successive 

 steps of activation and transformation in early neuralization can be admitted. For the simple 

 formation of the neural axis, it can be easily agreed that secondary individuating influences 

 are exerted on the one hand by the prechordal plate and on the other hand by the derivatives 

 of the chordo-mesoblast. Concerning the cause of the initial change, I do not adhere to 

 the two-waves theory of induction, for the following reasons: (i) the appearance of the 

 neural plate is a continuous event; (2) the fold experiments can be interpreted by hypo- 

 thetizing a continuous production of the A-agent in the whole chordomesoblast, with graded 

 space-time variations; (3) the activation-transformation idea is contrary to the bulk of 

 evidence relative to the exclusive modification of P into A ; (4) the fact that in certain sand- 

 wiches acrencephalic structures appear earlier and farther from the inductor than noto- 

 genic ones may be explained by the probable smaller molecular size of the A-agent; (f^) 

 the necessity of a preliminary activation does not manifest itself in the Yamada's latest 

 experiments, where the P-agent seems to have been better isolated than ever before; and 

 (6) although the course of events in amphibians is favorable to the idea of an activation by 

 the foreward migration of the prechordal endo-mesoblast, the same process is not evident 

 in birds, where the presumptive prechordal plate, included in the head process, does not 

 come into contact with the truncal part of the presumptive neurectoblast. 



The unitary interpretation which I advocate is also in agreement with our knowledge 

 concerning the genesis of the dorso-marginal zone. It still allows us to consider the normal 

 course of events in terms of morphogenetic potential, a preliminary expression which 

 should be translated into biochemical terms when possible or advisable. We should also be 

 conscious that this term includes substances which are not intimately the same at all suc- 

 cessive stages. Before the blastoporal lip appears, differences between areas are, in my con- 

 ception, purely quantitative within a single field, the substrates of which so far escape 

 analysis. When gastndation is put into action, a qualitative difference emerges, little by little. 

 Morphogenetic potentials should be compared not only within the general field ordered 

 around the presumptive notochord, but also in the more restricted region devoted to the 

 acromerite (p. 452). For this stage, we have at least indirect evidence concerning the nature of 

 the inducing agents (see infra). With neurulation, the picture changes again. In the noto- 

 genic part, the middle layer distinctly shows three levels of potential, still able to merge 

 with each other (p. 352), while around the spinal cord and medulla the role of the mesen- 

 chyme complicates the situation (p. 406). In the prechordal part, relations of potentials 



Literature fi. 483 



