FIELDS AND GRADIENTS 297 



physiological activity throughout the piece ; the extent of this in- 

 crease will vary with the age and position of the fragment. Thirdly, 

 the operation of cutting also results in an increase of activity : this 

 is intense close to the cut, and then appears to grade away rapidly. 

 Fourthly, external conditions influence the activity both of the old 

 tissues and still more of any new tissue proliferated at the cut 

 surface. This question has been discussed at some length by Child. ^ 

 Comparable results have been shown to occur in the regeneration 

 of fragments of certain plant tissues, such as seakale roots." 



(viii) So far as the facts are relevant here, we may sum them up in 

 the form of the following rule : The frequency or absence of regenera- 

 tion, and the type of structure regenerated appear to depend (a) on 

 the level of the cut surface within the original gradient-field, and (b) 

 upon the form and steepness of the gradient eventually established 

 between the proliferating tissues at the cut surface and the rest of 

 the piece. 



As Child has epigrammatically put it, when a new apical region 

 is regenerated, it arises not because of the activities of the rest of 

 the fragment, but in spite of them. 



As a corollary of these various rules with regard to the establish- 

 ment of polarity, at the autonomy and subsequent dominance of 

 the apical region, the facts concerning the varying number of 

 structures produced by a given piece of tissue may be satisfactorily 

 explained. A given length of Tubularia stem normally possesses 

 but a single hydranth, whereas regeneration experiments show that . 

 it is capable of producing dozens. The limb-disc of a Urodele, if 

 cut up and the pieces grafted, can produce several fully developed 

 limbs : why in normal development does it only produce one ? In 

 the regeneration of a fragment of Corymorpha stem, sometimes one 

 new dominant region is produced, sometimes two, sometimes 

 several : why is this ? 



The reason that a given field normally gives rise only to one of 

 the structures characteristic for it is due to the inhibiting effect of 

 a dominant region, once initiated, upon the development of other 

 dominant regions. Normally the gradients within the field are such 

 as to give one region a start ; this becomes the dominant region and 

 inhibits the potentialities of other regions. This is well seen in 



^ Child, 1915 A. ^ Jones, 1925. 



