18 MUTATIONS 



rocytes. These might represent mutant stem cells accumulated in the 

 blood-forming system. 



The method depends on agglutination, and this is a reaction that 

 does not go to completion, so in order to get at the frequency of cells 

 that are not agglutinable by an anti-A or anti-B reagent, you can use 

 an iterative process in which you co-precipitate the agglutinable cells 

 with other agglutinable cells that are repeatedly added. To tell the 

 difference between the added cells and the original cells in which you 

 are trying to find the inagglutinable frequency, the original cells are 

 labeled with chromium 51, and the added unlabeled cells act as carrier. 

 We successively agglutinate and separate, until the Cr^^ activity in the 

 mixture becomes constant (1) . 



When it has become constant we are not removing any more of the 

 agglutinable cells by addition of further carrier cells in the presence 

 of the agglutinin, so we can say that these cells are, phenotypically at 

 least, not A or B, as the case may be. It should be noted that this 

 phenotypic criterion is operationally valid even if some variant cells 

 can absorb agglutinin but do not agglutinate. 



Figure 3 shows a reconstruction experiment. When you add some 

 known inagglutinable cells, O cells in this instance, they stay in the 

 system as the naturally occurring cells are supposed to stay in it. 

 The system is anti-A with AiB carrier cells, and a trace of labeled 

 cells added. Each one of these points is a stage, representing an 

 agglutination and separation. AiB carrier cells are added to the 

 supernatant at each stage. As we go through these first seven stages, 

 nothing happens to the cells. They stay in the system ; in fact, they 

 seem to increase a little bit. 



Now, we add some labeled AiB cells to increase the Cr^^ activity 

 about seventyfold and keep on with the same stages. These labeled 

 AiB that were added are cleaned out rather quickly, as expected, and 

 we come to essentially the same level. 



This paradoxical increase in the cells is real; it comes from the 

 peculiar fact that the volume an agglutinate displaces contains spaces 

 into which the inagglutinable cells cannot find their way; so that every 

 time we discard the agglutinated mass, it actually contains fewer of the 

 added cells than does an equivalent volume of the supernatant. The 

 supernatant becomes a little more concentrated at each stage. This 

 is corrected for. It may be different with different reagents, so we 

 cannot always expect to come out with an exact level in an experiment 

 unless we have a reconstruction experiment with the same systems, but 

 very often we do come out with an exact level. 



