102 PROBLEMS OF RELATIVE GROWTH 



limbs, etc., that increase in relative rate of growth in an 

 appendage or region was accompanied by a steepening of the 

 growth-gradient. This holds even for the type of growth- 

 gradient affecting feather-growth only, which we are here 

 considering. In females and feminized males, the absolute 

 rate of growth of, e.g., the breast-feathers is reduced, and the 

 feathers, growth-gradient in this region is at the same time 

 almost flattened out (Fig. 56D). 



Mr. Miller, of the Animal Breeding Research Department, 

 Edinburgh, has pointed out to me that similar gradients 

 affecting the growth of epidermal structures occur in mammals. 

 For instance, in (untrimmed) manes of horses, the length of 

 the hairs increases steadily from the ears to the middle of the 

 mane, then sinks steadily to the hind end of the mane. A similar 

 gradient in hair-length occurs in the human beard, with high 

 point medially. In such cases we are only dealing with total 

 amount of growth as given by the definitive length of the 

 hair : I do not think anything is known about the rate of 

 growth or of regeneration. In the fowl, Juhn and her collabora- 

 tors (1. c.) find the following interesting facts. In breast, and 

 also in the back-and-saddle region, there exists a distinct 

 antero-posterior gradient in definitive feather-length. In the 

 breast, this is correlated with the above-mentioned gradient 

 in regenerative growth-rate, but also with the fact that 

 regeneration continues for a longer time in the posterior 

 feathers. Presumably the same two factors are at work in 

 normal growth. In back and saddle (information in a letter 

 from Dr. Juhn) there is no gradient in regeneration-rate, but 

 only a gradient in the length of time for which regeneration 

 proceeds. We thus have two distinctive methods of growth, 

 both capable of gradation, affecting the size of epidermal 

 structures. 



§ 8. Conclusion 



The chief points brought out in this chapter are the follow- 

 ing. Differential growth of a limb or appendage or a well- 

 marked region of the body appears never to be brought about 

 by an equal distribution of excess growth-potential throughout 

 the organ or region. On the contrary, the growth-potential 

 of the organ or region is distributed in the form of a growth- 

 gradient, normally with a single high point or growth-centre, 

 from which growth-intensity grades downwards in both direc- 

 tions (or in one, if the growth-centre be terminal). In general, 



