n6 



PROBLEMS OF RELATIVE GROWTH 



enlarged form. And further, even if functional hypertrophy 

 of the legs were possible in a beetle, it is impossible to see how 

 it could come into play in regard to the antennae. 



The second point is this — that the gradient is concerned 

 with spatial position, not with morphological position. The 

 mandibles are spatially the most anterior portion of the body. 

 Although the antennae are morphologically anterior to them, 

 they are spatially posterior, and are affected not by a correlated 

 decrease, as appears to be the rule for organs anterior to a 

 growth-centre (p. 122), but by a correlated increase. 



Another case in which analysis of existing measurements 

 has permitted us to plot the growth-gradients along the body- 

 axis, giving us what may 

 D e^f be called the growth- 



profile of the animal as a 

 whole, is that of the 

 metamorphosing herring, 

 Clupea harengus (Hux- 

 ley, 193 ib ; data of Ford, 

 partly unpublished, 

 partly in E. Ford, 1930). 

 In this case the growth- 

 gradients are distinctly 

 complex. Their nature 

 will be seen from Fig. 64. 

 Through their operation, 

 the elongated larval her- 

 ring is converted, while 

 considerably increasing 

 in absolute size, to its 

 post-larval form. Doubtless if measurements had been made 

 vertebra by vertebra, instead of at a few points only along 

 the body, the graded effect would have been more obvious, 

 and interesting results would have emerged as to what happens 

 at the boundaries of regions marked by different growth- 

 coefficients. We may perhaps assume that each region would 

 possess its own growth-gradient. This seems clearly so in 

 regard to the head. 



Later work by Ford (1931A) indicates that changes similar 

 in principle but different in quantitative detail are at work 

 in the metamorphosis of two other species of the same genus, 

 the pilchard {Clupea pilchardus) and the sprat (C. sprattus) ; 

 but growth-profiles have not been constructed for these. In 



Fig. 64.- 



-Growth-profile of metamorphos- 

 ing herring. 



Above, outline of larval herring at onset of metamor- 

 phosis, showing regions measured (after Ford, 1930) ; 

 below, its growth-profile, dorsal and ventral : the values 

 are growth-coefficients (k), relative to body-length. 



