Ill X E N O - 1 N D U C T I O N S 443 



Definite fractions will perhaps be isolated which would specifically invoke deuto-, 

 cormo- or ourogenesis. The induction of mesoblast and of neural tissue would 

 also be controlled by diflferent factors'. Would this experimental dissociation really 

 mean qualitatively different inductors? Is it justifiable to transpose such a notion 

 to normal development? These are two more questions to be solved by future 

 research. 



From the standpoint of the biochemist, it appears that these recent investigations 

 are in agreement with the idea that the active inducing agent existing in xeno- 

 inductors is a protein, but that in frequent instances it is bound with a nucleic acid. 



A similar conclusion has been attained in a rather special case of xeno-induc- 

 tion, discovered by Montalcini, Meyer and Hamburger (1954). The hyperplasia 

 of spinal and sympathetic ganglia caused in chicken embryos by extracts of mouse 

 sarcoma was first found to depend on a ribonucleoprotein contained in the ultra- 

 microsomes of a tumor. Further analysis demonstrated that the activity was 

 bound to the protein moiety (Cohen and Montalcini, 1956, 1957; see also Ham- 

 burger, 1956; Levi-Montalcini, 1956) which is also present in snake venom, and 

 even in mouse salivary glands. 



(/) Interpretative remarks 



In relating this experimental analysis of a most intriguing phenomenon, our in- 

 terest has been mainly orientated by the search for the active substance (s) which 

 could be sufficiently ubiquitous and biologically active to account for the numerous 

 facts observed. These extensive investigations have however revealed much more 

 than just the biochemical aspects of the problem. As a supplementary reward, 

 they give a new insight into the transformation evoked in the induced ectoblast 

 when the reaction consists either in notogenesis or in acrogenesis. They suggest, 

 in certain cases at least, a plurality of agents, not only corresponding to these two 

 main types of reaction, but also causing either deutogenesis or tritogenesis, or 

 inducing the chordomesoblast separately from the neural organ, and reciprocally. 

 They rise the question of whether these different agents, their authenticity being 

 supposed granted, are convertible in one or both directions, or whether they simply 

 coexist and are able to be selectively inhibited or excluded by various treatments. 

 Let us consider first the morphological processes at play in the notogenic and 

 acrogenic reactions, then the problems of plurality of agents and of their possible 

 conversion, and finally their biochemical nature. 



Notogenesis, either complete or under its more regional aspects of deuto-, 

 trite-, cormo-, ourogenesis, is often obtained by implanting normal or foreign 

 inductors, usually via the sandwich method. It is time to stress the fundamental 

 difference between these processes and the induction normally exerted by the 

 roof of the archenteron. The latter process presupposes that the conditions neces- 

 sary to build a notochord and somites exist before this stage. They are the endow- 

 ment of the dorsomarginal zone, and we have tried to explain how they have 

 been acquired (p. 343). Notogenic induction implies that the ectoblast is provided 



* Niu (1956, p. 162) reports that he used "fixed" adult tissues and various preparations of 

 nucleoproteins and nucleic acids. He often obtained differentiation in neural tissues and 

 pigment cells, but he failed so far to observe the appearance of myoblasts. 



LiieiatuTc j). 483 



