250 THE MOSAIC STAGE OF DIFFERENTIATION 



take place independently and in the absence of normal morpho- 

 logical differentiation. This may be seen in those experiments in 

 which a portion of the embryonic area of the blastoderm of the 

 chick is made to undergo development on the chorio-allantois of 

 another egg ;^ or in embryos of Cephalopods, the normal develop- 

 ment of which has been impeded by toxic agents. ^ 



The same conclusion emerges from the results obtained by 

 culturing in vitro various rudiments of the chick embryo, such as 

 those of the eyes, fore-limbs, and ears. In these cases, histological 

 differentiation may reach a high degree of perfection, while there 

 may be little or no approach to the morphological differentiation 

 of normal anatomy.^ (See also Chap, xi, p. 375.) 



A pretty example of abnormal morphogenesis is seen in the 

 differentiation of reconstitution-masses of dissociated sponge cells 

 which contain an excess of collar-cell tissue. In this case, partial 

 spheres consisting of a single layer of collar-cells are produced with 

 the collars directed outwards instead of inwards, as in the normal 

 gastral lining.* (See p. 281.) 



Purely morphological differentiation, then, seems to be in large 

 part conditioned by physical and mechanical factors of available 

 space, material, and pressure. Histological differentiation is in 

 large part independent of these factors. While these two kinds of 

 differentiation are sufficiently distinct during the later stages (i.e. 

 after their initial determination) for the one to take place without 

 the other, the question next arises as to what relation these two 

 kinds of processes bear to one another at the start. 



The first visible important steps in differentiation are concerned 

 with the form-changes which result in gastrulation and neurulation. 

 These may be held to constitute a phase of morphological differ- 

 entiation, which, in development, is thus seen to precede histo- 

 logical differentiation. The question therefore arises as to whether 

 the latter is dependent on the former in the initial stages. If it were, 

 we should have another instance of the supposed effects of "dy- 

 namic determination ", referred to on p. 163. The problem therefore 

 presents itself as to whether it is possible to prevent a certain region 



^ Murray and Huxley, 1925; Hoadley, 1924, 1925, 1926 A. 



- Ranzi, 1928. ^ Strangeways and Fell, 1926; Fell, 1928. 



* Huxley, 191 1; de Beer, 1922. 



