SUMMARIES AND CORRELATIONS. 



263 



subject than on the average of the group. Subject III, on the other- 

 hand, was affected by alcohol less than the average of the group in all 

 but one case. On this basis it would be possible to arrange our normal 

 subjects in a series according to their susceptibility to the depressing 

 effects of alcohol. The total position of each individual with respect 

 to the average of the group is shown in the extreme right-hand column. 

 The several values may be regarded as an index to the personal 

 equation of each subject with respect to the particular process which 

 is involved. The value at the extreme right thus becomes a sort of 

 alcohol coefficient of the individual. Of course these particular values 

 are entirely relative, i. e., relative to the rest of the group and to the 

 kind of measurements. But if the number of subjects and the kinds of 

 measurements were sufficiently numerous the deviation of the individ- 

 uals from the average would approximate a true coefficient. It seems 



TABLE 50. Variations from the average measurements shown by the individual subjects as 



calculated from the percentile effects of alcohol. 



to us that we are already justified in using our average percentile 

 effects of alcohol as a provisional standard for the estimation of the 

 susceptibility, not only of our present subjects, but as well of those 

 subjects that may serve in later tests. For such a comparative esti- 

 mate, however, it would be of great advantage if there were some process 

 whose measurement might be taken to represent the average without 

 the laborious and time-consuming measurements of our entire series 

 of tests. In the effort to discover if any of our values would qualify 

 for such a purpose we have plotted the various values of table 50 in 

 figure 32. 



Doubtless the first impression from figure 32 is that the several values 

 are quite irregular and unrelated. A more careful inspection, however, 

 will show that there is a fairly close correspondence between the curves 

 for similar kinds of measurement, especially for the reciprocal innerva- 

 tion of the finger and that for the velocity of the eye-movements. 

 Furthermore, both of these curves resemble more or less closely the 

 curve for the total results. These three curves are not identical. One 

 could scarcely expect that, even if they were curves of the same identical 

 process. But the eye-movement curve is sufficiently similar to the 



