462 



GERMINAL ORGANIZATION 



INDUCTION PHENOMENA 



tion of separate neurarcuals. This contention should perhaps be checked, since it looks much 

 more plausible that the primary metamerisation (that of the somites) is responsible for the 

 appearance of separate vertebrae. 



This point being reserved, the inductive correlation between the spinal cord and vertebral 

 column has been proved for fish, amphibians, and birds, and we shall see that it is also 

 valid for mammals. The possible role of the neural crest has, however, not been excluded. 

 This same correlation is even valid for the regeneration of the Urodele tail (Holtzer, 1956). 

 In order to take place, this process needs the presence of the neural axis; the cord can be 

 shown to induce the new vertebral arches, but only by its ventral motor part. The notochord 

 forms only if it attains the blastema, while myoblasts arise either from the former muscles 

 or from blastema cells. 



< 1 mm 



Fig. 95. In vitro combinations of somites and spinal cord to check the importance of prox- 

 imity, (a) A row of 6 somites starting from the spinal cord fragment, used as inductor. 

 (b) A row placed tangential to the cord, at i mm distance. In both cases, a continuous sheet 

 of mesenchyme is first produced. The (b) controls remain completely negative, unless a 

 fortuitous displacement accidently provokes contact, (c, d) Two positive results of (a). The 

 induction takes place earlier and is stronger in the mesenchyme adjacent to the cord, but 

 islands of chondrogenesis appear at a distance, and not necessarily at the place where a 

 somite was placed, (c, d) from Grobstein and Holtzer, 1955. 



Axial chondrogenesis has been studied in mammalian embryos by means oiin vitro culture 

 of the anterior region of mouse embryos (less than 25 somites) either alone or with pieces of 

 spinal cord; these induce chondrogenesis, and once more the ventral parts are the most 

 efficient (Grobstein and Parker, 1954). In a further analysis, Grobstein and Holtzer (1955) 

 checked the necessity of contact for induction to occur. The result of this experiment (Fig. 

 95) is interesting but not decisive. On one hand, chondrogenesis only occurs if at least one 

 somite has been placed near the piece of cord; on the other hand, if this neighboring mesen- 

 chyme forms a piece of cartilage, the same impulse toward differentiation may be trans- 

 mitted through the sheet of stellate cells to some condensation existing at a distance of i mm 

 or so. It seems to us as if the inductive stimulus from the cord is not transmissible without 

 contact, but a center of recently-induced chondrification, i.e. early precartilage, would let 

 diffuse an assimilative stimulus at a distance. Observations of Milaire (1956, p. 347) con- 

 cerning the limb bud suggest that alkaline phosphatase could well be present in the sheet 

 and help in the transmission of this stimulus. 



