DOCKERAY : PHYSICAL FATIGUE AND MENTAL EFFICIENCY. 229 



were not counted until the entire series was finished, a certain 

 amount of space on the sheet might represent to the subject 

 a certain amount of work done. This knowledge, coupled with 

 the awareness that a certain amount of time had elapsed, 

 might spur the subject to greater activity in case he were fall- 

 ing behind his usual amount. As the accuracy of the work was 

 not determined until the entire series was finished, no more 

 knowledge was possible in this case than the other subjects 

 possessed regarding their accuracy. 



Accuracy is distinctly improved in every instance after rest 

 except in the first test. The recorded introspections show that 

 the subject felt greatly fatigued from the first test of the first 

 day, which the eighteen minutes of rest did not seem to re- 

 lieve. This may account for the discrepancy of the first day's 

 results. The lack of knowledge of the relation of the space 

 covered to the amount of work done should also be considered 

 a factor in this discrepancy. After lifting five kilograms ten 

 minutes accuracy is improved nearly as much as after rest, 

 but the variations are greater. After twenty minutes' lifting 

 there is a notable decrease in accuracy with a relatively small 

 variation from the average. 



As was stated earlier, Do. could not have lifted the weight 

 for twenty minutes at the beginning of the series. At least he 

 thought he could not. The ten-minute periods seemed decid- 

 edly fatiguing even in the last test, though they never com- 

 pared with the twenty-minute periods, which were extremely 

 exhausting. 



SOUNDER TEST. 



This test differs from the addition and multiplication tests 

 in an important respect. Speed as a variable factor in the 

 subject's work is eliminated by the sounders being operated 

 at a regular rate, which the subject must follow. This results 

 in an incentive for the subject to keep sufficiently alert to 

 discriminate the sounds as they are given in order to prevent 

 absolute confusion. The accuracy of the discrimination alone, 

 rather than the speed and accuracy of the association proc- 

 esses, is all that is measured. The speed of the subject's 

 discrimination and reaction may be recorded by this method 

 for a short series, but for a period of sixteen minutes it would 

 be impracticable, as it would necessitate running the kymo- 

 graph rapidly in order to record distinctly small units of time. 



