210 MUTATIONS 



Benzer: You did find differences from one pair of twins to another 

 pair of twins? 



Glass: I had only one pair of twins to work with. I tried the same 

 test on a group of students in a class and got all sorts of results, 

 with no apparent consistency at all. But these subjects were not care- 

 fully diagnosed as to coffee-drinking and noncoffee-drinking indi- 

 viduals, so the interpretation is loose. In so far as the two monozygotic 

 twins were concerned, they were not used to drinking coffee, and they 

 reacted exactly alike over a period of about two to two-and-a-half 

 hours, after ingesting eight grains as a dose. 



Goldstein: Eight grains is approximately half a gram. 



Glass: Yes. 



Benzer: Is there any age dependence? 



Goldstein: I don't have data on age dependence, because our sub- 

 jects are all within the age of the early twenties. I would like to com- 

 ment, though, in respect to what Dr. Glass said, that on the whole it 

 has been very difficult, in the experience of psychologists and pharma- 

 cologists, to develop objective tests which will reveal the mild psychic 

 stimulation obtained by people from drugs of the type of caffeine and 

 amphetamine, except under conditions of prolonged sleep deprivation 

 and fatigue. In that case there is ordinarily a decrement in the ability to 

 perform various tasks, which can be partially restored by caffeine or 

 amphetamine. A sleep deprivation procedure would not be useful for 

 screening large numbers of people. We are working at developing a 

 more practical test. 



The criterion that we have used so far to distinguish reactors from 

 nonreactors is a reliable one but clumsy. It is based on establishing 

 standard conditions of caffeine deprivation from noon on a particular 

 day until bedtime. Then a standard dose of caffeine is administered in 

 decaffeinated coffee, in a double-blind, placebo-controlled design. We 

 then ask the next morning whether the individual did or did not have 

 difficulty falling asleep. This seems very crude, but with sufficient num- 

 bers of individuals and with a proper experimental design, you get 

 c}uite statistically significant differences between caffeine and placebo. 

 I might add that we were able to develop a method for adding caffeine 

 to decaffeinated coffee so there was no perceptible difference in taste or 

 appearance. 



A reactor is classified as an individual wdio, for a long period of 

 administrations, night after night, in a random design, in every case 

 responds to caffeine by a disturbance in his ability to fall asleep, and 

 in every case is not so affected by placebo. The nonreactors are those 



