224 KANSAS UNIVERSITY SCIENCE BULLETIN. 



Table TVb. — Association Tests. The average association times taken 

 from the total individual results, and the distribution of extremely 

 long and short times. 



Subj. R. M. Br. E. H. Do. 



After rest. Per cent correct, 71.9 80.7 67. 47. 40.6 54.4 



Average time 5931 2765 7426 4706 4449 5966 



No. of cases 74 59 34 15 24 34 



No. > 10,000* 13 2 5 2 3 8 



No. < 3,000* 23 41 5 4 7 11 



After fatigue. Per cent correct, 47.1 70.2 54.2 51.2 50.9 35.4 



Average time 9488 3159 6151 5595 7172 5442 



No. of cases 42 52 30 22 35 23 



No. > 10,000* 9 2 7 4 6 4 



No. < 3,000.7 7 32 8 13 8 11 



DISCUSSION OF RESULTS. 



The consideration of results that deal in any way with 

 physical fatigue is certain to raise the question as to how much 

 physical fatigue, general or special, has been produced in a 

 given individual by the method used. This it must be con- 

 fessed is a difficult question to answer in any case, and par- 

 ticularly so in the present set of experiments where mental 

 tests were to be given which must not be disturbed more than 

 is necessary. For this reason, and also because of the diffi- 

 culty of analyzing the chemical products of fatigue, the at- 

 tempt to determine the physical fatigue by measuring these 

 products was abandoned. Furthermore, it is doubtful whether 

 a very definite evaluation of the degree of fatigue could be 

 secured in this way. The only remaining possibility seemed to 

 be to test the physical capacity before and after the work 

 period, and compare the results with those obtained before 

 and after an equal rest interval. For this purpose a large 

 dynamometer was used. This instrument was similar to the 

 ordinary hand dynamometer, only much larger, and was 

 fastened to the floor by means of a chain attached to one 

 end of the spring. A wooden handle was attached to the 

 other end of the spring. The subject exerted himself in as 

 nearly as possible the same position as was required at the 

 beginning of the lift with the wall machine. The result was 

 disappointing. The subject, either because he shifted his 

 position slightly, or because for a single pull he exerted more 

 effort, was able to exert as much or more force on the dyna- 

 mometer after he had worked at the wall machine. For ex- 

 ample, Ca. was able to lift 100 kilograms before the work 



