38 DISCUSSION 



may, however, be preserved for several days when dried, without 

 electrolytes, in the frozen state. 



Woodruff: If you mixed the two, it would resolve the question of the 

 enzyme destruction. 



Davies: There does not appear to be any such enzymic degradation 

 in the ascitic fluid, because this can be incubated overnight, and you 

 don't find any drop in the haemagglutination inhibition titre, but this 

 would not tell us whether antigenic potency had been affected. 



BilUngham: Since A and B blood group antigens are present on 

 human epidermal cells (Coombs, R. R. A., Bedford, D. and Pouillard, 

 L. M. [1956]. Lancet. I, 461), the hair of mice and men might con- 

 ceivably furnish a convenient source of blood group substances if 

 means can be devised to extract and purify them. 



Medawar: One must bear in mind that the form in which these anti- 

 gens are present in ascitic fluids and in cell membranes is profoundly 

 different: the one is probably glycoHpid and the other certainly muco- 

 polysaccharide; this may entail very great differences in methods of 

 extraction. 



Davies: Yes, but we do not know that this is true of H-2 or any other 

 kind of antigen; it has only been proved for blood group substances. 



Simonsen : Has anyone here tried to isolate the antigen from the liver ? 

 I recently heard rumours from America, where they claim that about 

 60 per cent of the total transplantation antigen that can be extracted 

 from the mouse is concentrated in the Hver. 



Medawar: The Herzenbergs in America have made membrane 

 fractions from liver which are certainly serologically active ; the figure 

 of 60 per cent is new to me. 



Mitchison: May I return to the red cell again? The picture which 

 seems to be emerging of the red cell here is one of a cell which has the 

 antigens, but is non-immunogenic, and is therefore dismissed. I was 

 wondering, on the other hand, about its tolerance-conferring properties. 

 For instance, if a mouse tolerates a transfusion of incompatible red cells, 

 is its response to a skin graft thereby altered ? In my experience, mice 

 don't eliminate incompatible red cells except in certain strain com- 

 binations. Pursuing the same line of thought : since this is transplan- 

 tation antigen in some non-immunogenic form, what about the recom- 

 bination experiment ? Shouldn't red cells be mixed in with other tissue 



