in PRIMARY INDUCTION EXPERIMENTAL DATA 407 



case, growth and fiber production are found in relation with the myoblasts (Fig. 69, a, b, c). 

 This effect is exerted either by unsegmented or by segmented musculature (the last type 

 being formed if notochord is present). Even a small number of myoblasts has a definite 

 influence on the proliferation of the neuroblasts and the thickness of the wall (Fig. 69, g, h). 

 Combination between the influence of the mesenchyme and of the myoblasts explains the 

 resulting form of an atypical spinal cord and the graded distribution of the neuroblasts. 

 The same simple technique produces cases where notochord is located quite near the 

 neural tube, while the myoblasts are either scarce and dispersed or maintained at some 

 distance. These instances teach us that the notochord exerts practically no effect on the 

 growth of the neural wall (Fig. 69, i, k). The relative inertness of the notochordal material 

 was already noted by Holtfreter; this factor, combined with the positive action of the 

 somites, plays an obvious part in the typical structuration of the normal spinal cord 

 (Fig. 69, d, e,f). 



Takaya's recent contributions reveal a new step in the transformation of the 

 inducing system: the two derivatives of the somite, the myoblasts and the mesen- 

 chyme, have a quite different action. The dispersion and transformation into 

 stellar cells goes hand in hand with the loss of the growth-promoting induction 

 and with some effect of the permeability of the neuroepithelium. These ideas 

 allow us to reconsider the formation of the brain, and especially of its prechordal 

 part. If mesenchyme and myoblasts play such a conspicuous role in the modelHng 

 of the cord, are we not permitted to infer that similar conditions are active inside 

 the head? That the mesenchyme has the combined property of limiting the 

 proliferation of the neuroblasts and favorizing the expansion of the ependymal 

 cavity is especially attractive. The thinning of the dorsal brain wall and the 

 existence of the two membranae tectoriae suggest the participation of especially 

 active mesenchyme. Its mode of action might well be influenced by the neural 

 crest cells which are known to contribute to the leptomeninge; the association 

 of these meningeal components might be responsible for the formation of the 

 choroid plexus and the accumulation of fluid in the future ventricles'. On the 

 other hand, the basis of the brain is characterized by zones of intense neuro- 

 epithelial growth, in the wall of the rhombencephalon, the optic vesicles, and the 

 telencephalic lobes, for which the inducing role of the parachordal and pre- 

 chordal mesoblast has been recognized. Takaya's work must remind us that 

 these inducing anlagen are mesenchymous only in their appearance, for many 

 of their cells possess myogenic potency (tongue and eye musculature). We can 

 understand the general picture of brain morphogenesis by admitting that these 

 growth-promoting elements assume a definite pattern under the anterior half 

 of the neural plate, while later, the predominance of the complex cephalic 

 mesenchyme around several dorsal parts of the primary brain vesicle influences 

 their fate. In the basal region of the vesicle, the parachordal mesenchyme seems 

 to acquire two areas of special inducing activity, responsible for the auditory 

 placodes. Similarly, the prechordal mesenchyme can be imagined to concentrate 

 into two areas which induce the optic vesicles and two responsible for the telen- 

 cephalon. The intermediate spaces would be filled by an inert, growth-inhibiting 



^ At early stages, no mesenchyme is, however, present against the thin roof of the romben- 

 cephalon. It seems that the epiblast must play the role of inhibiting cell division and 

 making the epithelium impermeable to the ependymal fluid. 



Literature p. 483 



