EFFICIENCY AND FATIGUE 263 



of these individuals are very short, a moderate pro- 

 portion of them are rather short, a large proportion 

 of them are of medium height, a moderate proportion 

 of them are rather tall, and a small proportion of 

 them are very tall. Similar frequencies of occurrence 

 are observed for their muscular strength, for their 

 keenness of eyesight, and for other physical attributes; 

 and similar frequencies also for their capacity for pro- 

 ducing munitions, supposing that they are all working 

 their best. But supposing that the workers are arti- 

 ficially limiting their output, then the frequency of 

 distribution of their output values would almost in- 

 variably fail to conform to the normal probability 

 curve. Take, for instance, the other group of women 

 who were sorting cartridges at a time rate. Their 

 average output was only a third that of the piece-rate 

 workers, but unfortunately the values are not directly 

 comparable, as the type of sorting by the time-workers 

 was rather more complex in character than that by 

 the piece-workers. Assuming, for the sake of argu- 

 ment, that the best of the time-workers had as great 

 an output as the best of the piece-workers (though we 

 may be quite sure that they had no such thing), I 

 found that their output values had the frequencies 

 shown on the left side of fig. 3. A few of the women 

 had an output of only 17 to 22, but the majority of 

 them had an output of 32 to 52, or a lower value than 

 that observed in any except 2 per cent of the piece- 

 workers. In fact, the majority of the time-workers 

 sorted cartridges only about half as fast as the majority 

 of the piece-workers, though a few conscientious ones 

 achieved a good output. 



Other and much more important instances of limi- 

 tation of output I have recently observed in one of the 

 largest of our shipyards. In this yard, before the 

 war, most of the workers were paid at a piece rate, 



