206 KANSAS UNIVERSITY SCIENCE BULLETIN. 



a problem as |g|, the operations were performed as fol- 

 lows: 8x4, record 2; (8 X 7) + 3 + (6 X 4) , record 3; 

 (8x3) + 8 + (5x4) + (6x7),record4; (6x3) + 9 + (5X7), 

 record 3; (5 X 3) -f- 6, record 21. The problems were not so 

 difficult, therefore, as in Thorndike's (37) tests, but the sub- 

 jects were able to perform more of them in a given length of 

 time. Many of my subjects, moreover, found it next to im- 

 possible to retain the total product until the entire problem 

 was finished. 



It was difficult to determine what should be considered the 

 unit operation for computing the mistakes. The units place 

 in the total product involves one multiplication, the tens place 

 involves two multiplications and one addition, and so on. 

 Furthermore, a mistake in the tens place might show itself in 

 the hundreds and not in the tens place at all. It was finally 

 decided to consider a problem wrong that was not entirely 

 correct. While this method allowed rather wide variations 

 that could not be recorded, it seemed to be as nearly accurate 

 as any other method that could be devised. 



In order to test the effect of the fatigue of a small group of 

 muscles as compared with the more general physical fatigue 

 of the whole body, or at least of a much larger group of 

 muscles, two types of ergographs were used. One consisted 

 of a small spring about six inches long suspended between two 

 vertical bars. Beneath the spring hung a brass band into 

 which the thumb of the left hand was inserted. The whole 

 ergograph was mounted on a small board raised to the level 

 of the arm-rest. The arm was strapped to the arm-rest, palm 

 up, with the fingers under the board supporting the ergo- 

 graph. A downward pressure with the thumb, therefore, was 

 accompanied by contraction of the flexors of the four fingers 

 as well as of the thumb. The principal fatigue, however, 

 seemed to be felt in the muscles controlling the thumb. The 

 contractions were recorded on the kymograph by means of a 

 marker attached to the spring. The more general fatigue 

 was secured by means of a heavy spiral spring mounted in an 

 iron tube about two feet long and three inches in diameter. 

 The tube was attached in the vertical position to a plank 

 base two feet square. A heavy rod was attached to the lower 

 end of the spring and projected about a foot above the top 



