582 ON INCREASE IN COMPLEXITY [pt. iii 



although that of the rest of the body was varying considerably 

 according to the food eaten. 



3 -8. Axial Gradients 



It is now time to turn to another aspect of the analysis of embryonic 

 development, the theory of axial gradients. A convenient transition 

 is afforded by Bellamy's discovery that the dorsal lip of the blasto- 

 pore and the animal pole in the frog's egg are regions of "high 

 protoplasmic activity". According to the theory of axial gradients, 

 this would mean that the metabolic rate at those points — obviously 

 of great importance as being the seat of the organiser — was higher 

 than anywhere else in the embryo at that moment. It is plain that 

 this is a matter of much interest, and it is therefore necessary to 

 examine in some detail the theory of metabolic, axial or physio- 

 logical gradients as a whole. Much of the evidence on which it is 

 based is to be found in the four books of Child, and Abeloos and 

 Ranzi have written valuable reviews of the subject. 



The fundamental conception lying at the base of these views is 

 that of axiate pattern. Child emphasised in all his work the idea 

 of polarity and of gradients of activity between poles. He proposed 

 that we should think of the embryo or the animal as existing in a 

 three-dimensional graph or co-ordinate system, and being con- 

 stituted in a kind of pattern of axes of symmetry. Each axis would 

 pass from one pole to another, but along its length the protoplasmic 

 activity would not be constant; on the contrary, it would be very 

 high at one pole and very low at the other, dwindling away at a 

 definite and measurable gradient. Such an axis may or may not 

 at its origin in time have a visible morphological outward sign of 

 its existence. It may or may not last as long as the differential 

 growth to which it gives rise. In fact, ontogenesis from this point 

 of view is the clothing of the original protoplasmic axiate pattern 

 with a corresponding morphological axiate pattern. The anatomical 

 gross differentiation of parts with reference to a given axis is pre- 

 ceded, as it were, by the appearance of a gradient of physiological 

 activity along this axis. By a gradient of physiological activity Child 

 meant a series of quantitative differences between the properties of 

 the cells, following a definite orientation with reference to the 

 eventual pattern of the animal. Child may be said to have trans- 

 ferred those co-ordinate diagrams which d'Arcy Thompson used to 



