REGENERATION AND TISSUE EQUILIBRIUM 277 



These observations agree with those of Morgan, who previously noted that 

 if, in tadpoles, a tail is cut off and the cut-off tail of another larva is grafted 

 onto the wound, regeneration does not occur on the cut surfaces, although 

 both the stump and the grafted tail have the power to regenerate. 



Similar results were obtained in anuran amphibia by Graper. Transplanta- 

 tion of extremity buds on stumps of limbs succeeded, but regeneration was 

 prevented thereby only if the orientation of the cut surfaces of host and 

 transplant to each other was correct. If the two surfaces were not adequate, 

 the transplant either changed in such a way that it became secondarily ad- 

 justed to the host and was transformed into the right kind of extremity, or, 

 if this did not take place, there was a regeneration of the original limb, not- 

 withstanding the presence of the graft. Of special interest is the fact that in 

 case of a disharmonious character of the cut surfaces a regenerative growth 

 occurred, which did not need to be restricted to the cut surfaces but which 

 took place even at some distance in the transplanted limb. We have already 

 referred to similar results when we discussed regeneration in primitive adult 

 invertebrates, where likewise an outgrowth may take place at some distance 

 from the place of union of the two pieces, a contact effect apparently having 

 been propagated from the directly affected area to nearby parts. Therefore, 

 according to Graper and Weiss, a satisfactory axial orientation between trans- 

 plant and host is essential if regeneration is to be suppressed. An arm can 

 inhibit the regeneration of a posterior extremity, provided the axes in host 

 and transplant have an analogous orientation. If the transplantation occurred 

 not directly at the point where a part of the limb had been cut off, but at some 

 distance from it, in the direction towards the head in the branchial region, the 

 tendency to regeneration was greater, but in principle the same competitive 

 struggle took place between the prospective or early regenerate and the trans- 

 plant, and in certain cases both pieces, regenerate and transplant, coalesced. 

 The transplant, even if it did not succeed in suppressing the regeneration, was 

 able in some instances to make it less perfect. 



In many other experiments, also, especially those of Harrison, success in 

 the grafting of extremities in amphibian larvae depended largely upon the 

 fulfillment of the condition that the transplant satisfy the tendency of the host 

 to form a certain type of extremity ; unless the transplant conformed to this 

 condition, the reaction of the host tissue was unfavorable to a permanent union. 

 In these cases we have, it seems, to deal with specific interactions between 

 host and transplant at the point of contact. But homoiogenous tissue of a 

 different kind, such as transplanted living skin, may also exert an inhibiting 

 effect on the regeneration of extremities. Thus Harrison and Detwiler found 

 in embryos of Amblystoma that the regeneration of limbs which had been 

 excised, can, to some extent, be inhibited if the wound is covered with 

 homoiogenous skin, and it can be entirely prevented if the wound and the size 

 of the skin subsequently grafted onto the wound are very extensive. 



However, there are several investigations which make it very probable that 

 in addition to these specific contact actions, also purely mechanical, non- 

 specific factors may play a part in preventing regeneration. Thus Schaxel 



