DISCUSSION 133 



man criticized my conclusion that host cells were necessary for the 

 transport o£ antigen on the ground that grafts in diffusion chambers 

 were undernourished and might in consequence not Hberate much 

 antigen. I set out to look for a system which was not open to this 

 criticism and found it in the mammahan foetus (Woodruff, M. F. A. 

 [1958]. Proc. roy. Soc. B, 148, 68). 



I should like to comment also on a brief communication in a recent 

 British Medical Journal (Hackett, E. and Beech, M. [1961]. Brit. med. J., 

 2, 1123) in which somebody has revived the old idea of trying to treat 

 chorionepitheHoma in women by trying to immunize the patient with 

 paternal tissues. This particular trial was not very successful, but it 

 does seem to me that the idea merits further investigation. 



Russell: I should like to summarize briefly some experiments on the 

 antigenicity of mouse trophoblastic tissue which Dr. Richard L. 

 Simmons and I have been doing. I think they bear directly on the 

 problems which Dr. Ha^ek has raised. 



We believe that there is clear evidence that during pregnancy the 

 mother is quite competent to deal with foreign antigens, derived from 

 adult tissues of the same genetic origin as its foetus, when they are 

 presented as ordinary grafts, say, skin grafts (Medawar, P. B. and 

 Sparrow, E. [1956]. J. Eudocr., 14, 240; Woodruff, M. F. A. [1958]. Proc. 

 roy. Soc. B, 148, 68). It is also clear that foetal tissues contain histo- 

 compatibility antigens from quite an early stage (Terasaki, P. I. [1959]. 

 J. Embryol. exp. MorphoL, 7, 409), certainly as early as 12 to 14 days of 

 gestation in the mouse and possibly earlier (Haskova, V. [1959]. In 

 Biological Problems of Grafting, p. 95. Oxford: Blackwell). We 

 reasoned, in view of this, that the facts inight best be explained by the 

 "barrier theory" of immunological segregation of mother and foetus. 



Theoretically the requirement is for an anatomical "buffer zone" 

 between mother and foetus which prevents effective contact between 

 foetal antigens and the reactive cells of the mother, a buffer which 

 must not itself release foetal antigens. 



Our experiments have made use throughout of placentas from F^ 

 matings between CBA and A-Hne mice (the paternal strain being A). 

 Placental transplants were made to adult CBA recipients. In this way 

 any reaction against such grafts could not be ascribed to maternal tissue 

 fragments which had been transferred with the placenta. 



