ioo PROBLEMS OF RELATIVE GROWTH 



females during the remainder of life (k = about 0-95 instead 

 of 072). 



In the parasitized females, k for length is throughout just 

 below that of normal specimens, whereas k for breadth descends 

 well below the normals between 10-5 mm. and 13-5 mm. 

 carapace-length, then gradually ascending almost to the level 

 of the normals. This latter fact would seem to indicate that 

 in females the parasite (as in Sacculina) acts as would a pre- 

 cociously functioning ovary, but that the value for normal 

 chela-breadth in the larger normal specimens represents a final 

 partition-coefficient towards which parasitized as well as nor- 

 mals tend to approach. It is further clear that breadth-growth 

 is more sensitive than length-growth to ovarian influence and 

 to the parasite's pseudo-ovarian influence, and reacts to them 

 both earlier and to a greater degree. The failure of the chela 

 of parasitized males to respond in breadth-growth to the 

 influence of the parasite as much as the normal female chelae 

 to the influence of the ovary may be due to a specific difference 

 in male and female chela-tissue, but it is more probably due 

 to the fact that the degree of feminization effected by the 

 parasite is variable. As breadth-growth is more sensitive than 

 length-growth to feminizing influences, this variability will be 

 reflected more markedly in the curves for breadth-growth. 



Finally, the work of Hecht (1916) on the growth of Teleost 

 fish is of some interest here. He finds in a number of species 

 of markedly differing body-form that the relation of the maxi- 

 mum width of the body to total length remains fairly constant 

 (to about ± 16 per cent.) whereas the relation for depth is 

 highly variable (by over ± 55 per cent.). The form of fish 

 appears thus to be determined much more by changes in the 

 growth-coefficient for depth than by changes in that for width. 



§ 7. Gradients in Growth-rate of Epidermal 



Structures 



Another extremely interesting case of growth-gradients is 

 provided by the work of Juhn, Faulkner and Gustavsen (1931). 

 Working on the domestic fowl, they find in the male (normal 

 and castrated) definite regional gradients in the rate of growth 

 of feathers which are regenerating after plucking. The most 

 marked of these regional gradients is in the breast, and has 

 its high point posteriorly (Fig. 56). 



In addition, there is in normal and castrated males a 

 dorso-ventral gradient in breast, back, and saddle, with high 



