The question op fatigue prom the economic standpoint. 257 



Type-selling by hand. — 'Piece-hands.' Average number of 'ens' 

 over period of ten full working days {February 1916). 



Notes. — There are two mid-spell breaks of ten minutes : — 



(a) From 9 to 9.10. 



(6) Round 3.30, when tea is taken. For the periods 9 to 10 and 3.15 to 4.15 

 the output is averaged up to the full hour. There is no break in the work from 

 3 to 3.15. 



The average curve of the output for all the individuals engaged on 

 these type-setting processes follows very closely the curves which were 

 given last year for soldering and labelling tins, and which were then 

 suggested as the normal curve for all work requiring concentration and 

 attention. 



Here again the two spells show a similar level of output and a 

 similar curve. On the machines the afternoon output is 2 per cent. 

 higher, in the hand-work it is 2 per cent, lower than the morning output. 

 In both spells, with one exception, the output is at a maximum in the 

 second hour and falls off in the third and fourth. In the afternoon the 

 fall in the fourth hour of the spell (and the last of the day) is particu- 

 larly marked. The one exception to the rule of a maximum in the 

 second hour occurs in the afternoon spell of the type-setting by hand, 

 when the maximum is in the third hour (from 3.15 to 4.15). 



If we may venture on an explanation of the above facts, the usual 

 rise in output between the first and second hours of a spell would seem 

 to be due to the worker getting practised, the fall occurring after the 

 second hour to be due to fatigue. As for the exception in the time of 

 the maximum output, the explanation probably lies in the cup of tea 

 and the break of ten minutes given to the piece-hands at 3.30. The 

 effect of the similar break at 9 a.m. in the case of machine operators 

 as well as piece-hands no doubt adds its weight to that of practice in 

 producing the morning maximum in the 9 to 10 hour. 



The above table also records the average output of each individual 

 separately. As might be expected in industry where so many different 

 factors contribute to the result, individuals show some wide deviations 

 from the average curve of output for the day.' 



' The extent of these deviations from the curve can only be measured 

 clearly if the hourly output of each individual be expressed as a percentage of 

 his average hourly rate. Otherwise individual deviations in the level of output 

 will interfere and affect the deviation. 



1916 s 



