208 MUTATIONS 



think one should say, in only two human subjects, so it does not take 

 into account the possibility of significant variations in the metabolic 

 pattern from person to person. 



As far as I am aware, we don't know anything about the mutagenic 

 activity of the metabolites of caffeine, so we don't know how to 

 estimate, really, whether falloff in the blood level is to be taken as a 

 diminution of mutagenic activity, or whether, as the caffeine level 

 falls and the metabolites come up, we may even be getting more 

 mutagenic effect. This is something that should certainly be investi- 

 gated. I will assume for the moment that only caffeine itself is muta- 

 genic. 



What we then, obviously, have to calculate is the mean levels to 

 which individuals who are ingesting given amounts of coffee or tea are 



TABLE 6 



Coffee-drinking habits of 167 medical students at Stanford. 



Cups per day % of sample 



23 



1 17 

 2-4 37 

 5 or more 23 



(Unpublished data of Goldstein and Warren.) 



exposed. If you start with the population of the United States and 

 the average consumption per capita, and correct for the number of 

 individuals under age 14, you come out with 121 million people of 

 coffee-drinking age, with an average consumption of about 23 pounds, 

 or 10 kg. The average caffeine content of coffee is about 1 per cent. 

 This would give us an exposure of about 100 grams of caffeine per 

 year per capita or approximately 300 mg a day. That amounts to 2 to 

 3 cups of coffee in a day. 



That this is a reasonable result we can judge from information 

 about actual coffee-drinking habits. Table 6 presents a small sampling 

 of such data from a study we did on Stanford medical students. This 

 group may not be representative of the population as a whole,* but 

 what we find is that about one-fifth of the sample doesn't drink any 



* These data are borne out surprising well by a nationwide survey conducted 

 by Elmo Roper in 1947-48 (51). 



