32 MUTATIONS 



When you knock out the bone marrow with radiation it has to go 

 through a period of clonal growth to catch up again. This offers an 

 opportunity for selection in which the negative cells might win out. 

 That could exj^lain these radiation results, as well as the ones in man. 

 On the other hand, if we assume the same selection factor for cells 

 in heterozygous and homozygous A individuals, for example, then 

 mutation and selection would produce an even greater difference be- 

 tween the proportions in the heterozygote and homozygote than 

 would mutation alone. But we actually observe less difference. This 

 kind of selection would not give you an excess of the kind observed. 



You would think that, if selection is involved, the starting number 

 of the negative cells would be related rather directly to the amount 

 of increase with radiation, whereas, if mutation alone is involved, there 

 would be no relation between them. The starting number is there, and 

 with a certain dose of radiation you produce so many mutant lines. 

 This number would be quite independent of how many negative cells 

 are present at the start. More experiments are needed before this 

 point is clarified and they should be done all with the same kind of 

 carrier. With pigeon cell carrier, you have anywhere from 10 to 100 

 times as many negative cells as you do with human carrier. 



Neel: In the pigeons, are the differences between individuals before 

 treatment greater than have been observed in man? 



Atwood: They are, if you use pigeon cell carrier. When I learned 

 of the carrier effect it scared me because I thought, well, everything is 

 artefact now. We have to use blood bank sources of carrier, and we 

 don't know anything about them. We tried different carriers on the 

 same blood and never found any carrier effect with man at all. This 

 restores a measure of confidence. 



The carrier effect could be explained on the assumption that there is 

 a specific difference between the pigeon and human cells and that part 

 of the lectin is bi-specific or heteroligating. This would explain a phe- 

 nomenon that Scheinberg observed in following up the effect, namely, 

 that if he isolates a negative population of cells from a positive pigeon 

 by successive sedimentation with lectin, and washes the cells that are 

 not agglutinable by lectin, then as soon as he adds human Ai cells, 

 the pigeon erythrocytes accumulate a rosette of human erythrocytes. 

 This suggests that they were coated with antibody but not aggluti- 

 nated, and that the basis of the carrier effect is that, with pigeon cell 

 carrier, you reach a point where there are a few cells left that are 

 coated with a bi-specific lectin, the pigeon end of which is inside and 

 the human end outside, and cells coated in this way can attach only to 

 human cells and cannot stick to other pigeon cells. 



