MUTAGENS OF POTENTIAL SIGNIFICANCE 213 



Unfortunately, there was also some uncertainty about the exact 

 amount of caffeine found in this experiment. The analyses, after cor- 

 rection for tissue blank, showed at least 1 mg/kg and possibly as 

 much as 7 mg/kg in the testis. Technical changes should improve the 

 precision of future assays. The point here is not that there is a 

 particular concentration of caffeine in the testis, but that cafTeine un- 

 questionably enters the human testis. The caffeine there was identified 

 by its characteristic ultraviolet absorption spectrum, i.e., by the 

 difference in spectrum between the blank and experimental tissue. 



Does anybody want to stop me at this time on any of these points? 

 These are fairly factual matters. 



Motulsky: What was the frequency of reactors versus nonreactors? 



Goldstein: The clear, strong, and consistent reactors in our experi- 

 mental group run about 20 per cent of the group. 



Atwood: Did that group contain most of the heavy coffee drinkers? 



Goldstein: No, there is not a good, clear relationship between re- 

 activity and coffee-drinking habits; in fact, the heavy coffee drinkers 

 tend to be less reactive to a given dose of caffeine, which you can 

 interpret in one of two ways. It may mean that they have become 

 tolerant to the effect of caffeine. But it may also mean that be- 

 cause they are less sensitive, they have to be heavy drinkers in order 

 to get the same psychic stimulation that somebody else gets with a 

 lesser amount. 



McKusick: Is there a real bimodality here or does it shade off? 



Goldstein: We can't say that, because of the insensitivity of our 

 methods. Clear reactors, this 20 per cent group, are easy to pick out. 

 We could also pick out those people who never, under any circum- 

 stances in a series of repeated administrations, gave a positive reaction 

 to caffeine. That was another 20 per cent or so. That left the whole 

 middle group who gave variable responses, sometimes to caffeine, 

 sometimes to placebo, which may or may not have meant something. 

 We were unable to say. We also don't know yet whether a person who 

 fails to react to a 300 mg test dose would react to some larger dose. 



Freese: I would like to ask whether the metabolism of caffeine in 

 bacteria is known. 



Novick: Yes. E. coli bacteria do not metabolize caffeine in any 

 detectable way. 



Freese: It could be, e.g., that in humans caffeine is metabolized so 

 fast that it cannot act, while in bacteria, it stays around and there- 

 fore is mutagenic. 



Novick: I feel that many things could be. 



