PROBLEMS OF MEASUREMENT OF MUTATION RATES 45 



say that it is associated with another genotype? Can you say, with 

 these few samples, whether it is the other allele or anything else in the 

 genotype, which makes for the change? It's just equivalent, isn't it? 



Atwood: It is for the few cases I have been discussing. 



Cotterman: But we have this. Dr. Stern, in quite a few individuals 

 in the family, and over two or three generations. This has been ob- 

 served in so many pedigrees that it would be unlikely that the modifi- 

 cations were due to many other genes than the B allele itself. 



Lederberg: That's fine. Is there any example of an interaction at all? 

 What Kim called the A2 allele is simply expressing itself in only 70 

 per cent of the cells in which it appears, regardless of what the other 

 genes are doing independently. 



Atwood: That is what this indicates. We have no particular reason 

 to think it is expressing itself differently in AiAs's than in AsB's. It 

 seems to express itself in just the same way. 



Lederberg: The Ai covers up the Ao. It's only when the Ai is 

 dropped out that — 



Atwood: Exactly, which would bring you down to about the level 

 of 6 X 10-^. Then, you get a slightly lower level, because the A3 isn't 

 completely expressed. 



Goldstein: One of the problems that arises from what you have been 

 telling us, it seems to me, which relates to the question of age de- 

 pendence or non-age dependence, is the question whether the changes 

 that you have been discussing arise from the stem cell line or whether 

 they arise during the clone multiplication from stem cells. 



Atwood: That is certainly correct. 



Goldstein: It seems to me that it should be possible to regard each 

 clone of final erythrocytes as an independent population, with an 

 average life of 120 days. If one made repeated determinations, say, at 

 six-month intervals in the same individual, one should have, in essence, 

 a variance test. Well, one should expect, then, high variance between 

 such determinations if these were arising primarily during the clonal 

 development of erythrocytes and not primarily if they were in the 

 stem cell line. Would that be a possible thing to do? Do you have 

 any data on the consistency of repeated determinations in a single 

 individual over a long period of time? 



Atwood: It is stable over three years. 



Now, your question as to the variance introduced by mutations 

 occurring in the intervening clone between the stem line and the 

 peripheral erythrocyte has an answer that depends on the number of 

 such clones whose products are in the circulation at the same instant. 



