;86 



NATURE 



[January 17, 19 18 



results are concordant with those of Macdonald- J 

 and others. From these experiments it appears that, I 

 for any one person, the relation between H, the total 

 energ-y transformed (measured in thermal units), 

 and W, the external work done (also measured in 

 thermal units), is adequately expressed by the 

 equation H=aW + b, where a is a constant and 

 h a variable parameter, varying with the speed of 

 work performance. In the case of a professional 

 cyclist, upon whom Benedict and Cathcart per- 

 formed a large number of experiments, a was 

 approximately equal to 3*3, while h increased from 

 2-4 to 5-2 as the rate at which the pedals were 

 rotated increased from 68-72 to 108-112 revs, 

 per min. When unpractised persons used the 

 ergometer the value of a increased, but the avail- 

 able data were not sufficient to permit of the para- 

 meters being determined with any accuracy. 



From these results we may infer that (i) the 

 incremental efficiency of muscular work may be 

 as high as 30 per cent, in favourable circum- 

 stances, and (ii) the total cost of work perform- 

 ance depends upon its rate. We can scarcely, 

 however, venture to generalise the arithmetical 

 results by using them to calculate the needs of 

 those doing other kinds of work. 



(2) This method was largely used by Zuntz and 



Schumburg^ in 



their well-known studv of the 



requirements of marching soldiers, and has also 

 been employed by Amar* in investigating the 

 energy transformations of industrial workers. 

 Many physiologists, including Atwater and 

 Benedict, Voit, Rubner, and Tigerstedt, have 

 carefully determined the heat output of persons 

 at rest,' obtaining reasonably concordant results, 

 so that the energy transformations of workers 

 can be contrasted with those of sedentary persons. 

 From Amar's experiments it appears that a 

 metal filer plying his tool at the rate of 70 strokes 

 per minute (a skilled operative, aged thirty-eight 

 years, weighing 74 kilograms) would transform 

 or liberate 3656 Calories daily if he worked at the 

 rate mentioned for eight hours, slept for eight 

 hours, and "rested " the remaining eight hours. 

 The figure just given is reached on the assumption 

 that the heat output during sleep is i Calorie per 

 kilogram of body-weight an hour ; during non- 

 working but waking hours, i'25 Calories— assump- 

 tions in accord with the means of other experi- 

 ments. Allowing a margin of 12 per cent, to 

 cover unavoidable waste in the preparation of food 

 and non-assimilation of portions of the ingredients 

 consumed, this daily transformation is covered 

 by a diet having an energy value of 4155 Calories 

 as purchased. Little significance attaches to an 

 isolated series of observations, and it is to be 

 hoped that the method will be more widely em- 

 ployed in that organised physiological research 

 into industrial conditions which is an urgent need 

 of the time. 



(3) This process has been widely adopted, the 

 largest individual collections of statistics being 



(a) those recently compiled and analysed by the- 

 Welfare and Health Section of the Ministry of 

 Munitions, and relating to more than 18,000 

 munition workers ° ; (b) the studies issued from 

 the Nutrition Laboratory of the United States 

 Department of Agriculture, which cover more 

 than 13,000 persons, of whom, however, only a 

 small minority were industrial workers 6; (c) the 

 Solvay Institute's analysis of the food consump- 

 tion in more than 1000 Belgian industrial 

 families"; (i) English urban working-class and. 

 agricultural budgets analysed by the Board of 

 Trade some years ago.^ 



In the following table mean values computed 

 from the above-mentioned material (omitting the 

 American data, which may not be strictly com- 

 parable with those describing European conditions) 

 are collected : — 



" Proc. Roy. Soc, B, 1917, vol. Ixxxix., p. 394. 



* " Studien zu einer Physiologic des Marsches." (Berli; 



* " Le moteiir humain " (P.iris, TQ14), pp. 527 ct stg. 



1901.) 



* The average for the whole 1944 families (wages ranging from less than. 

 251, to more than 40^.) is : — 



Protein Carbohydrate Fat Calories 



98-8 593"2 83-7 3615 



The figures in this table, excepting those for 

 Belgium, refer to food as purchased. The Belgian 

 investigators have expressed their results in terms 

 of food absorbed by the digestive organs; the 

 deduced averages are accordingly enclosed in 

 brackets, not being directly comparable with the 

 others. The unbracketed figure for Calories is that 

 obtained on the assumption that a discount of 

 12 per cent, should be allowed between purchased 

 and assimilated values, and is (if the assumption 

 be admitted) comparable with the remaining 

 average energy values. 



These statistics must be interpreted with 

 caution. Two assumptions are made in computing 

 nearly all averages of the kind, and a third is 

 often involved also. The assumptions in question 

 are (a) that published analytical results showing 

 the composition of foodstuffs are generally ap- 

 plicable to the qualities used by the persons whose 



5 Summarised in Dr. Leonard Hill's " Memorandum on Workers' Food ' 

 (Health of Munition Workers Committee, No. 19, Cd. 8798). 



•i Contained in successive Bulletins of the U.S.A. Department of Agri- 

 culture. 



" Slos.se and Waxweiler, " Enquete sur le Regime alimentaire de 1065- 

 ouvr ers beiges." (Brussels, 1910.) 



8 Board of Trade, 1903, Cd. 1761, p. 210; 1913, Cd. 6955, p. 300. 



NO. 



2516, VOL. 100] 



