OTHER GRADIENT THEORIES 175 



It is important to note that in Planarians the morpho- 

 genetic effects associated with the gradient (induction of 

 pharynx and other organs by the dominant region) take place 

 early, but do not interfere with the persistence of physiological 

 effects of the gradient (e.g. regenerative capacity, growth- 

 intensity) throughout life. In a not dissimilar way, the realiza- 

 tion of normal morphological differentiation of a limb or tail 

 in Triton does not interfere with the persistence of the morpho- 

 genetic potency of the surrounding area to produce a new or 

 additional differentiation of the same type throughout life 

 (Guyenot and Ponse, 1930). 



The gradient, though persisting throughout life, might be 

 altered in shape. It is more natural on various grounds we 

 would expect that it was more likely to be flattened than 

 steepened with age. If it were flattened, we should expect 

 that the potency of differentiation at a given body-level would 

 be altered to a potency originally characteristic of a more 

 posterior body-level. This is what appears actually to occur 

 in serially heteromorphic regeneration in Crustacea and insects. 

 The regenerated heteromorphic appendage almost invariably 

 is of a type which normally belongs to a more posterior seg- 

 ment — e.g. antenna regenerated in place of eye-stalk (Palae- 

 mon), or fore-leg in place of antenna (stick-insects.) 1 



For these and various reasons we may regard it as probable 

 that the primary axial gradient of the egg and early embryo 

 will normally persist, although doubtless often in somewhat 

 altered form, in later periods and probably throughout life, 

 even when we have no ready means, such as antero-posterior 

 differentiation, heteromorphic regeneration, or head-frequency 

 in regeneration, of deducing its existence. 



Further, if it does persist, we may again regard it as prob- 

 able, especially in view of the facts in Planaria, that it will 

 exert some influence upon growth. This influence may be 

 direct or indirect, but in any case would be graded in its effect. 



Further, to reverse the approach, we may as a matter of 



speculation conclude that it is probable, when gradients 



specifically concerned with growth, like those in the male 



chelae or female abdomena of Crustacea, are found, that these 



1 Recent work by Przibram (Akad. Wiss., Vienna, 9. vii. 1931) and 

 Suster (ibid.) confirm this view. In Dixippus, antennae amputated 

 in the 1st instar regenerate as antennae, while in later instars they 

 regenerate leg-like organs. In Sphodromantis, the regenerate forms 

 an antenna at 25°, but a leg-like organ at lower temperatures. In 

 both cases the leg-like organ is produced when the axial gradient may 

 be presumed to be flattened. 



