THE QUESTION OP FATIGUE PROM THE ECONOMIC STANDPOINT. 315 



The divergence shown between men and boys in comparing the 

 ninth hour's accidents with the earher hourly average of accidents 

 (Table IX. b) seems to vary for the different industries. Thus, while 

 in shipbuilding all males have in the ninth hour of work 143 per 

 cent, the accidents they have in the first hour, boys only have 128 per 

 cent. But in auto manufacture and iron and steel foundries boys 

 have in the ninth hour 200 per cent, and 196 per cent, the accidents 

 they have in the first hour, while males of all ages have only 132 per 

 cent, and 153 per cent. It is only therefore in these industiies that 

 boys' accidents diverge appreciably from the normal by rising still 

 higher ; but the height of the rise is significant. 



The explanation of boys' and women's continued rise in accidents 

 throughout the spell would seem to be either a lesser power of antici- 

 patory excitement or a stronger fatigue overcoming this excitement. 

 The women's fewer accidents in the afternoon, however, would seem 

 to exclude the stronger fatigue explanation for them. Lesser antici- 

 patory excitement (therefore the only explanation left) may be some- 

 thing psycho-physical in the woman, but more likely it is due to the 

 fact that for most women the end of the factory spell does not mean 

 hope of a rest and food, but the certainty of more work in the prepara- 

 tion of food. 



3. In examining how far differences in the output and accident 

 time-distributions are attributable to differences in the factor ' Nature 

 of the Work,' reference must be made to Section La, which dealt 

 specially with that factor. There it was advanced that there were 

 certain characteristics in any single craft or process that were 

 peculiarly important in evoking affections akin to fatigue in the worker. 

 These characteristics were ' concentrativeness, ' 'persistence,' com- 

 plexity, uniformity, frequency of recurrence and regularity of recur- 

 rence. A good opportunity of studying single occupations or crafts 

 as apart from whole industries is afforded by our output figures in 

 particular. It is to these figures that we turn, therefore, for the 

 results of what we called the ' evocative ' characteristics. 



First it must be observed, however, that, owing to the statistical 

 necessity of securing a frequently and regularly repeated output (see 

 Section II.), the varieties of output that our tables exhibit cannot include 

 infrequent and irregular operations. The characteristics chiefly 

 differentiating the processes are therefore (a) concentrativeness and 

 persistence ; (fa) complexity ; and, above all, (c) the degree of 

 uniformity. 



(a) Movements concentrating or focussing their tax on local muscles 

 are found in the processes of covering chocolates by hand (Table II. 

 Col. 2) and in sandwiching and stencilling biscuits by hand (Table V. 

 Col. 4), and here in both cases we find the output closely following the 

 normal distribution as far as the morning spell is concerned ; but in the 

 afternoon spell the output decreases very slowly, and in stencilling 

 does not reach its maximum till the third hour. Moreover, in the case 

 of the chocolate-covering, the output in the afternoon is higher 

 generally than in the morning. 



This difference from the normal distribution is attributable to the 



