PROBLEMS OF MEASUREMENT OF MUTATION RATES 25 



fill work. This is the isotope dilution curve with anti-A and A carrier. 

 At this point, anti-B and B carrier regime is started. Whether anything 

 happens is a matter of conjecture. I have plotted it in such a way as 

 to emphasize that something might have happened; that is, you might 

 have a little bit of decrement at the beginning here with anti-B and B 

 carrier. 



After five stages of this, Ulex and carrier are added, and there is 

 a large removal of cells. If any B cells are present in this individual, 

 they can't be more than 10 per cent of the remaining non-A cells, 

 whereas the cells amount to nearly all of the remaining cells. You 

 can't say that no B cells are produced from A, but you can say that if 

 they are produced, they are produced much less frequently than 

 cells are produced from A. 



This is consistent with the notion that there are more ways of chang- 

 ing the gene into a form not making A than into a form that makes B. 



Goodgal: Can you add the Ulex first and then measure what is 

 left? 



Ativood: That can be done, and should have been. Actually, how- 

 ever, we did an experiment in which anti-B was added first. 



If we have a heterozygote, which we can designate as AO and we 

 are getting some A to B, then this would give you OB cells that 

 would be detected in here. But if we have going to B, it gives you 

 an AB cell, which is removed here (upper part of curve) and there- 

 fore is not detectable. 



In a futile attempt to look for to B, we ran some stages with 

 anti-B at the beginning, before removing cells with anti-A. This ap- 

 proach is very insensitive, though (Fig. 6) . 



This second individual had a lower proportion of non-A cells than 

 the first. This curve is with anti-A alone. The other curve has three 

 preliminary stages of anti-B which, at the same rate of removal 

 as in other experiments, should have brought any B cells down to a 

 vanishingly low proportion. No removal was detected. Then, the anti-A 

 is run, and you see that it is hard to decide whether the difference 

 between the two curves is real. No real difference is expected in any 

 case. 



Ito did some experiments with normal people who had the isoag- 

 glutinins as well, and got the same sort of equivocal results with them 

 as with the first case, so I should guess that essentially no mutations 

 that create B cells are occurring. 



Glass: We have some data that would bear out this conclusion, from 

 growing various kinds of tissue cells in culture and then testing them 



