TRANSPLANTATION OF PIECES OF TISSUE 249 



as Hellmich points out, comparable to the totipotent cells which have been 

 found in sponges (archeocytes), hydrozoa (interstitial cells), vermes (neo- 

 blasts), and tunicates (amoebocytes). Corresponding to the relatively un- 

 differentiated character of such cells, their organismal differentials are pre- 

 sumably also as yet relatively little developed and they can therefore be 

 successfully heterotransplanted, while this is impossible in ontogenetically 

 further developed stages. 



However, differences in the degree of differentiation of the organismal 

 differentials do not depend merely on the presence or lack of certain un- 

 differentiated cells, which, under ordinary circumstances, remain more or 

 less dormant, but such differences must also exist in the ordinary tissues 

 composing an embryo or a larva in various types of animals. It is presumably 

 due to this inverse parallelism between the prospective potency of embryonal 

 tissues and the degree of specificity of their organismal differentials, that in 

 larvae of urodele amphibia extremities can be successfully homoiotransplanted 

 under conditions which make such a result impossible in anuran larvae ; 

 in contrast with what is found in urodeles, in anuran larvae homoiotrans- 

 planted extremities placed in close proximity to a developing extremity of 

 the host heal in only temporarily; they then cease to grow and undergo 

 shrinking and necrosis. But so far a systematic comparison of auto- and 

 homoiotransplantation of limbs has not, apparently, been undertaken, and 

 this would be necessary before more definite statements can be made as to 

 the development of organismal differentials in these embryos and as to the 

 parallelism between the degree of organ and tissue differentiation and the 

 fixity of the organismal differentials during different stages of embryonal 

 life. 



The stage of differentiation of the transplant seems to influence its fate in 

 a graded way. We have seen that fully differentiated extremities of amphibia 

 cannot be successfully transplanted into the skin or subcutaneous tissues. 

 If somewhat younger extremities are used, only the less differentiated parts 

 like perichondrium and other mesenchyme cells remain alive, become a part 

 of the host and continue to differentiate. Also, after transplantation into the 

 interior of the host the fully differentiated cells die and only the less dif- 

 ferentiated cells remain alive. However, when very early stages of extremity 

 buds are transplanted, although no further development takes place, necrosis 

 does not occur and the mesenchyme cells of the transplant dissociate from 

 one another and migrate into the host tissue (Hellmich) ; but in other cases, 

 if an early embryonal bud is able to maintain itself in the host, it is more 

 accessible to the organizer action of the surrounding host tissue than older 

 tissues, and accordingly, it is modified in its development by the host tissue, 

 whereas somewhat further differentiated transplants develop through self- 

 differentiation. Correspondingly, several authors have stated that in trans- 

 plantation of avian embryonal tissues the results are not favorable if very 

 early stages are used, and similarly, Goetsch has observed that if in Hydra 

 very early regenerative buds of the oral region are transplanted into the 

 lateral zone of other polyps, the transplants are resorbed, whereas older re- 



