PROBLEMS OF MEASUREMENT OF MUTATION RATES 



19 





^ iOOO- 





O 



< 



in 



100- 



PHASEOLUS ANTI-A A, B CARRIER 



LABELED A, B CELLS 



LABELED CELLS 



I 



1 



T 1 1 1 1 1 1 1 1 1 1 1— 



0123456789 10 M 

 STAGE 



Figure 3. Reconstruction experiment with in- 

 agglutinable O cells and agglutinable AB cells. 

 The point on the left is the initial reaction mix- 

 ture in which only the cells are labeled. The 

 following six points are supernatants of suc- 

 cessive agglutinations with unlabeled AB car- 

 rier cells. The high point is the reaction mixture 

 of stage 7 after the addition of labeled AB cells. 

 The remaining five points are supernatants as 

 before. (From Atwood and Scheinberg, Science 

 129: 963-64, 1959, by permission.) 



This was done on various people, anyone I could get to volunteer, 

 and we found that A individuals had non-A cells in the amount of 

 something like 1 in 1000. There was a great deal of individual vari- 

 ability. 



About two years after that, it became possible to do the same experi- 

 ments with anti-B, and we obtained the first results from which you 

 could attempt to make sense. We also have an anti-H which, for the 

 purpose of this discussion, I will call anti-0. It agglutinates cells as 



