64 Effect of Alcohol on Psycho-Physiological Functions. 



is found to be 471, 474, and 466 a, respectively, with a total average of 

 470 <t. For the alcohol days the comparable averages are 466, 456, and 

 408 a, respectively, with a total average of 443 a. On the average, 

 therefore, the word reaction following the taking of the control mix- 

 ture is 27 a longer than that after the alcohol dose. The mean varia- 

 tion following the alcohol is slightly larger, being 38 a as compared 

 with 31 o-. 



In terms of differences (see section n of the table) we have in the 

 right-hand column average differences for word reactions for the normal 

 days of —10, +15, and +37 <r, with an average of +14 <r, and for the 

 alcohol days +6, +1, and — 52 a, with an average of — 15 <r. In other 

 words, on two of the normal days (July 1 and 3), there was an average 

 tendency in the periods which followed the taking of the control mix- 

 ture for the reaction time to be longer than in the first period of the day, 

 while on the first normal day, June 29, the reaction time at the begin- 

 ning was rather long, 481 <r, in comparison with the other first-period 

 values in the series. The tendency shown on the last two normal days 

 for the reaction time to be longer after the taking of the dose is also 

 evident on the alcohol days June 30 and July 2, but not to so great an 

 extent. On July 4 the average reaction for the first period, 460 a, is 

 somewhat longer than would be expected, considering those for the pre- 

 vious two days. This was followed by other reaction averages for 

 periods 3 to 6, which are remarkably low, these being 391, 401, 405, and 

 375 <r. The effect of these unusually rapid reactions following the rather 

 long ones of period 1 is to give us an average difference between the first 

 period and the succeeding periods of —52 <r. The average alcohol 

 difference for all alcohol days, -15 <r, is thus greatly influenced by the 

 result of July 4. If we calculate the average effect of alcohol by the 

 usual method, i. e., by subtracting the average normal difference from 

 the average alcohol difference, we have, as a result, — 29 cr, which, in 

 percentiles, shows an average decrease in the word reaction of 6.3 per 

 cent. It is further clear that the result would be in the same direction 

 and almost of equal magnitude if we omitted the first normal day and 

 the last alcohol day. The mean variation between the individual reac- 

 tions is, on the average, increased after the alcohol to the extent of 14 

 per cent. 



The results considered by periods show that there is a definite length- 

 ening in the reaction time following alcohol in period 2 as contrasted 

 with the same period on the normal days following the control mixture, 

 the average reaction time in period 2 for the alcohol days being 469 cr, 

 and for the normal days 452 <r. The differences on the three alcohol 

 days are +8, +13, and +10 a, respectively, with an average of +10 <r, 

 as contrasted to -21, +6, and +2 <r, with an average of -4.3 a for the 

 three normal days. The effect of alcohol is therefore (+10) - ( — 4.3) 

 or + 14 a, which in percentiles is +3.1 per cent. If we should omit the 



