DISCUSSION 215 



buted in the unfertilized egg to begin with— that is, differentiation 

 occurs because the genetic material of one cell does not have the same 

 kinds, or amounts, of substrates to work with as the genetic material of 

 another cell. I wonder why a similar mechanism couldn't be involved 

 in the establishment of antibody-forming clones ? 



Simonsen: I would have no objection to that at all. 



Loutit: I don't know whether I have not fully understood your hypo- 

 thesis, as demonstrated in your Fig, 2. If you invoke mutation, which 

 I always think of as random, why do you get the particular pattern 

 which you illustrated ? If it is random there should be n patterns which 

 are possible, yet it seems to me that only that pattern explains your 

 results. It looked to me as though something directed this process 

 rather than random selection. 



Simonsen : I wouldn't like you to take the last diagram too literally. 

 All I meant to illustrate was the idea that the different loci (I am speaking 

 now in the strict mutational sense) determining the spontaneous affmity 

 to different antigens may have different mutabihty. And if so, it would 

 lead to a heterogeneous population of the kind I have been talking 

 about. 



G. Klein: I agree with Prof. Woodruff that this is more attractive 

 than the predestination theory. I would agree also with Dr. Silvers that 

 this is probably more akin to differentiation than mutation, but Dr. 

 Silvers implied that differentiation is due to cytoplasmic segregation. 

 In fact, however, differentiation is due to something about which we 

 know nothing. So I would suggest that we use some more non-com- 

 mittal term than cytoplasmic segregation, or mutation; we could just 

 call it variation. 



Silvers: I would go along with that. 



Simonsen: I agree too. 



