MUTAGENS OF POTENTIAL SIGNIFICANCE 187 



divide (46). Chromosome breakage by various agents could be sur- 

 veyed in vitro with this technique. 



The second thing you could do is use an isotope dilution method in 

 blood to find out whether any agent is radiomimetic. I don't say it tells 

 you whether an agent is mutagenic, but at least it would tell you 

 whether it behaves like radiation. Radiation, obviously, causes too 

 large an increase in the minor populations to attribute just to specific 

 locus mutagenesis, but some of the drugs that are mutagens might also 

 show that large effect. 



Auerbach: I should like to say that if any screening test for the effects 

 of mutagens on man is to be used it has to be used on mammals. In 

 addition to the method suggested by Dr. Atwood, I should like to 

 mention another one, which is somewhat more laborious but has 

 already given promising results. This method was first used by the 

 Kleins (34) in Stockholm and by Dr. A. N. Mitchison (41) in Edin- 

 burgh. It consists of testing mouse tumor cells for mutations — perhaps 

 I should call them genetical changes — at a histocompatibility locus in 

 the mouse. The mice are used as test tubes in which populations of 

 tumor cells are grown. Tumors that have been induced in hybrids be- 

 tween two inbred lines with different histocompatibility alleles do not 

 grow in either parent strain unless one of the histocompatibility alleles 

 has been lost. What this is due to one does not know ; but when the loss 

 is specific, that is, when the altered tumor cells take in one parent 

 strain and not the other, we may presume that it is a genetical 

 change — mutation, or deletion, or crossing over. Dr. Dhalival (20) 

 used this method in Edinburgh and produced such changes by TEM 

 and X-rays. He has now gone back to Malaya, where he will go on 

 with the work. I think the main drawback of the method is that treat- 

 ing tumor cells in vitro would not produce mutations by just those 

 mutagens in which Dr. Goldstein is most interested, namely, those that 

 require replication for producing mutations. But I don't think it 

 should be too difficult to devise a method by which the tumor cells 

 could be treated under conditions that allow growth, for instance, by 

 giving caffeine to mice in which the cells can grow, and then screening 

 in noncompatible hosts. 



Goldstein: Does Dr. Glass think that tissue culture of human cells 

 might contribute to useful screening tests of any kind? 



Glass: Yes, I do. I think that that would be a very natural develop- 

 ment in the use of the technique. In fact, as I said a while ago, we 

 have already begun this with some of the chemical mutagens, beta- 

 propiolactone, and some of the diepoxides. 



