The size of the mortality sex ratio from pneumonia depends upon several factors which exert varying pressure with age, since it was shown for the London boroughs that the relative male risk was not high or low throughout life but displayed considerable variations.

In infancy social conditions have little effect upon the ratio, and there was no association with the degree of urbanization. In London the correlations with the indices of density became insignificant when the infant death-rate from pneumonia was kept constant. The male infant risk increases with the prevalence of the disease, and significant correlations were obtained when the indices of overcrowding were kept constant.

In childhood the size of the mortality sex ratio is not related to urbanization. The conditions operating for a high male mortality in childhood varies from those in infancy, since, for the eighty-three county boroughs, there was no correlation between the two ratios.

For the age group 15–65, in which the male pneumonia rate was twice that for females, there was a distinct association with degree of urbanization. In the London boroughs the mortality sex ratio was unrelated to social and economic conditions when the usual indices, the proportion of the population living more than two to a room, and the proportion of males in social class V were used. That environmental conditions had little effect on the ratio is also suggested by the experience of the mining communities which showed little difference in the mortality sex ratio between the county boroughs and urban districts. The rural mining areas were in agreement with the urban when allowance was made for the agricultural workers in these districts. In London there was a small significant correlation between the sex ratio and the pneumonia mortality sex ratio, indicating that the residential or industrial character of the borough was associated with the mortality ratio. Some support for this is given by the county boroughs where high ratios were found in the industrial towns and low ratios in the residential towns. Some occupations involve a direct risk of pneumonia, i.e. furnacemen, rollermen, etc., and others have a more indirect risk, e.g. inn-keepers, hotel-keepers, etc. Although the prognosis of pneumonia is affected by previous indulgence in alcohol, the effect of this factor can only be a matter of speculation. The mortality sex ratio from pneumonia in the three occupations in which a higher proportion of heavy drinkers are found than in the general population, inn-keepers, hotel-keepers, etc., barmen, etc., makers of alcoholic drinks, all of which have a high standardized mortality ratio from pneumonia, form an interesting progression. These values were 247, 394, and 1200, but the number of deaths among wives in the last two groups, 5 and 2, was so small that the ratios are really of no value. Yet there is, perhaps, an indication that as the opportunity for the wives to indulge in heavy drinking decreases with each class the mortality sex ratio from pneumonia increases. This factor of alcoholic indulgence probably operates in most occupations, and since the proportion of heavy drinkers is larger among men than among women, it would be necessary for only a small proportion of men to carry a heavy risk of pneumonia mortality to effect a large increase in the mortality sex ratio. It seems that the factors responsible for the relatively high male mortality from pneumonia, in this age group, were an occupational risk and the habits of the males.

In old age, as with the young age groups, there was no association with degree of urbanization. The mortality sex ratios of the London boroughs showed evidence of a geographical distribution. This was apparently a local characteristic, since the ratio of the county boroughs did not show any tendency to vary geographically.