SECT. 3] AND ORGANISATION 563 



together, or when the stomodaeum unites with the gut. Here the 

 concrescence is one of surface, but it may also take place between 

 the margins of the layers, as in the union of the embryonic plate 

 with the trophoblast in some mammals. Finally, a layer may spHt 

 into two by lamination, which is what happens in the inner wall 

 of the pineal vesicle in lizards. 



Aggregates of cells in mass formation may also be subject to quick 

 and profound changes of shape. The outgrowth of the limb-buds 

 affords an example of simple increase of size and weight, but more 

 complicated processes may take place also, such as the elaboration 

 of internal cavities or special internal structures. The blastocoele 

 cavity offers itself as an obvious instance of the former alteration, 

 but another one would be the origin of blood-vessel cavities from 

 soUd ropes of cells. Internal re-arrangement is also seen in the 

 formation of the concentric corpuscles in the thymus and the spleen. 

 Division of masses occurs plainly in the production of somites, i.e. 

 the metameric segmentation of mesoderm and neural crest. Fusion 

 of masses occurs when two gangUa unite and attachment of 

 masses is seen when the sclerotome and the notochord come 

 together. 



Another version of the same catalogue is seen in the table taken 

 from Kellicott. Both of the tables include practically all known 

 morphogenetic processes. But the point to which I wish to direct 

 attention specially here is the possibilities which such classifications 

 hold out of laying bare the physico-chemical processes which accom- 

 pany the differentiation phenomena. Up to the present time, no 

 chemical experiments consciously directed towards this end have 

 been done, but the body of knowledge which has accumulated about 

 the chemistry of the embryonic body as a whole, and which will be 

 reviewed in the succeeding sections, does make it possible to draw 

 certain conclusions in relation to morphogenesis. These tables of 

 KelHcott and Jenkinson are given here to serve rather as a back- 

 ground for what follows than as a summary of the facts. We have 

 in them an assembly of the various means by which differentiation 

 takes place. Each of those means must have, as everyone would 

 grant, a physico-chemical aspect. We may prefer to say that each 

 of them must have a physico-chemical cause and basis. The main 

 aim of the chemical embryology of the future will be to unravel 

 these causes and their inter-relations. 



36-3 



