Page:Popular Science Monthly Volume 69.djvu/517

Rh 1. The decline in the birth rate is not merely the result of an alteration in the ages of the population, or in the number or proportion of married women, or in the ages of these.

It is necessary at the outset to remove one possible explanation. What the Registrar-General gives us is the crude birth rate—that is to say, the exact proportion of births during the year to the total population, whether old or young, married or single. But in comparing these birth rates for different years, we have to remember that important changes may take place, even in a single decade, in (a) the proportion between children and adults; (b) the proportion between married and unmarried; and (c) the proportion between married women of the reproductive age and those above that age. These changes—due, it may be, to emigration or immigration, to economic or social developments, or to mere prolongation of the average life—are sufficient, in themselves, to produce a rise or a fall in the crude birth rate, without there having been any increase or decrease in human fertility. To give one striking instance, the crude birth rate of Ireland per 100,000 population fell from 2,384 in 1881, to 2,348 in 1901. But we happen to know that in the course of these twenty years the proportion of married women of reproductive age to the total population so far diminished that the slight fall in the crude birth rate really represented, not a decline, but a positive increase in fertility. If the Ireland of 1901 had contained a population made up, by ages, sexes and marital conditions, in the same proportion as that of 1881, the recorded births in 1901 would have appeared as a birth rate actually higher by 3 per cent, than that of 1881. We have, therefore, first to ask what are the corresponding figures for England and Wales, eliminating all the elements of variations of age, of postponement of marriage, and of positive refusal to marry.

Now, it so happens that this problem has lately been worked out by the statisticians in a way to remove all uncertainty. Dr. Arthur Newsholme and Dr. T. H. C. Stevenson on the one hand, and Mr. G. Udny Yule on the other, have performed the laborious task of 'correcting' the crude birth rates for differences of age, sex and marital conditions, as regards the census years from 1861 to 1901. The results show a