On Nutrition and Sex-determination in Man. 263 



(3) On the other hand, the labouring classes, upon whom the 

 bulk of the muscular work falls, and who form so large a pro- 

 portion of the industrial population, are seriously underfed. The 

 average energy value of the diet in the fourteen families selected 

 for study being no less than 23 °/o below standard, while the 

 average deficiency in proteids amounts to as much as 29°/o- The 

 enquiry, it is true, has shown that the money available for the 

 purchase of food is not always spent in the most economical way, 

 but the fact remains that unless an unreasonably stringent diet 

 be adopted, the means to purchase a sufficient supply of nourishing 

 food are not possessed by the labourers and their families. 



Rown tree's investigations were carried out upon the city of 

 York, but, as he himself points out (loc. cit. p. 299), there is no 

 reason for considering that the conditions obtaining in London 

 differ to any appreciable extent from those in the more northern 

 city. Indeed, in either case, the proportion of the population 

 living in poverty, and therefore seriously underfed, was about 

 30 °/ . It will be assumed, therefore, that the conclusions arrived 

 at by Rowntree for York apply also to London. 



In the official Census of the County of London for 1901, the 

 metropolis is divided into 29 boroughs. Excluding the City of 

 London, with its peculiar conditions, we are left with 28 boroughs, 

 having a total population of just over 4,500,000. For each of 

 these boroughs the proportion of male to female infants under 

 one year has been reckoned. The servant-keeping capacity (Census, 

 p. 154) has been taken as the guide to the relative wealth of the 

 different boroughs, it being assumed that when a family is able 

 to do so it will keep a servant. In Table I the 28 boroughs are 

 arranged in order of wealth (on the servant-keeping capacity 

 basis), whilst at the same time the relative proportion of male 

 and female infants is given. At first sight there would appear 

 to be no relation whatever between the two columns. This is 

 probably because we are dealing with a long series and with small 

 differences. Accordingly the 28 boroughs have been arranged in 

 three groups : A, with a percentage of indoor servants less than 

 15 (per 100 families); B, with a percentage of over 15 but less 

 than 30 ; and C, with a percentage of over 30. The relative pro- 

 portion of male and female infants then works out to — 



99 5 (/s per 100 $s: for group A. 

 100-7 „ „ „ „ „ „ B. 

 102 2 „ „ „ „ „ „ C. 



Hence in group A where there is a larger proportion of people 

 living in poverty, and therefore under less favourable nutritive 

 conditions, the relative number of $ infants is greater than for 

 group B, and still greater than for group G. On the other hand 



