I. NEURULATION 111 



environment, ectodermal grafts will produce only ectodermal 

 tissues, and no muscle tissue, notochord, or gut any more 

 (Mangold, 1923). This shows that chemodifferentiation has 

 progressed in the animal material, the parts of which were still 

 equivalent at the beginning of gastrulation. The cells have 

 been fixed in one definite direction; they have become de- 

 termined to one definite differentiation. 



How does this determination take place? Experiments by 

 Spemann (1918) have given a clear answer to this question. It 

 appears that the material of the dorsal part of the marginal 

 zone is provided with special properties. We have seen that 

 during gastrulation this material is rolled in over the dorsal 

 lip of the blastopore, and subsequently it forms the roof 

 of the archenteron, from which the notochord and somites 

 originate. Spemann removed part of this material from 

 an early gastrula, and grafted it into the ventral side of an- 

 other germ of the same age. In this region the primordium 

 of a complete embryo was formed, consisting of neural tube, 

 notochord, somites, and gut. Spemann's pupil Hilde Mangold 

 (1924) repeated this experiment, but combined two different 

 species, in a so-called heteroplastic transplantation. The graft 

 was taken from the dorsal lip of the blastopore of Triton 

 cristatus, and it was grafted into the ventral side of T. taeniatus. 

 It became clear in this case that only part of the secondary 

 embryo originated from the graft itself. The latter was, indeed, 

 rolled in under the influence of its special topogenetic potency, 

 and it produced notochord and somites. The neural primordium 

 of the secondary embryo, however, did not arise from the graft, 

 but entirely or for the greater part from the overlying ectoderm 

 of the host. The secondary gut primordium also developed from 

 host material (Fig. 40). This means that the graft enforced 

 a definite course of development on the cells of the host. It 

 ''induced" them to undergo a cert^.ln differentiation. This, 

 however, did not lead to a chaotic aggregate of tissues, but 

 the various organs produced by the co-operation of graft and 

 host showed normal topographical interrelations so that to- 

 gether they constituted an organised embryo. In very favour- 

 able cases, this may go so far that the anterior end of the 



