Page:Popular Science Monthly Volume 41.djvu/494

478 families. This is an exceedingly vital question, and much light will be thrown upon it under future statistical investigations.

In discussing the number of families and the composition thereof, it is interesting always to learn the relation of persons to dwellings. The following table gives the total number of dwellings and persons to a dwelling, by geographical divisions, under the census of 1890:

An examination of the foregoing table proves that the number of persons to a dwelling is constantly decreasing, although slightly, thus indicating increased comfort on the part of the population as a whole. In 1850 there were 5·94 persons to each dwelling in the country, while in 1890 the average was 5·45. In the West, however, this statement is reversed, for in 1850 the number of persons to a dwelling was 4·27, and in 1890 it had increased to 5·05. This, as in the case of the increased size of family, shows the effects of the new settlements.

A dwelling, for census purposes, means any building or place of abode in which any person was living at the time the census was taken, whether the abode was a room above a warehouse or factory, a loft above a stable, a wigwam on the outskirts of a settlement, a hotel, a boarding or lodging house, a large tenement-house, or the dwelling-house ordinarily considered as such. On this basis the number of dwellings in 1890 had increased 28·22 per cent over the number in 1880. In 1890 there were 11,483,318 dwellings and 12,690,152 families, there thus being 10·51 per cent more families than dwellings, while in 1880 the excess was 11·06 per cent, and in 1850 it was 7·02 per cent.

It should be remembered, in making any comparison between dwellings and persons from 1850 to 1890, that in 1860 and 1870 the total number of dwellings included both occupied and unoccupied dwellings, while in 1850, 1880, and 1890 the total number of occupied dwellings only was reported. Again, in 1850 and 1860 the number of dwellings stated was for the free population only, the dwellings of the slave population in those censuses not being returned. Any figures, therefore, for 1850 to 1870, inclusive, do not afford a very fair basis of comparison. The table just given should be used in the light of these remarks.