rate and the death-rate have steadily fallen the marriage rate has remained fairly constant.

The English birth-rate in 1861 was 34.5, death-rate 21.5, infantile mortality 15 per cent, an excess of 13 births over deaths per 1,000, or 1,000 became 1,013 each

year. The birth-rate rose to 36 in 1876. In 1900 it was 25 and is now lower. The death-rate has fallen to about 14 and the infantile mortality rate is about 10 per cent. The population is increasing almost as fast as before fall of birth-rate.

The German birth-rate rose to 41 in 1876, and has since fallen to 31. The death-rate has declined from 29 to 17, and the infantile mortality rate from 21 per cent to 17 per cent.

In France the birth-rate in 1881-84 was 38.9, falling in 1901-96 to 21.1, but the death-rate in the same time fell from 37 to 19.6, so rate of increase is about the same.

Holland had a birth-rate in 1876 of 37 which has fallen to 29 but the death-rate has fallen from about 25 to 13 and infantile mortality from 18 per cent to less than 10 per cent. Since 1895 Holland has taught by royal decree methods of preventing conception. According to the official year book the proportion of men drawn from the army over 5 feet 7 inches has increased from 24.5 per cent to 47.5 per cent since 1865 and the proportion under 5 feet 212 inches has fallen from 25 per cent to less than per cent.

8

The Australian birth-rate fell from over 42 in 1860 to a little over 26 in 1910. The death-rate fell, however, from average of about 18 to a trifle over 10 and infantile mortality from 11 per cent to about 7 per cent.

The New Zealand birth-rate in 1855 was about 42; in 1910, about 26. The fertility rate has steadily declined from 337.2 per 1,000 married women in 1878 to 226.6 in 1906. But the death-rate has fallen likewise from about 17 in 1860 to a little over 9 in 1910, and the infantile mortality rate is now about 7 per cent. The result is that excess of births over deaths is about 16 per thousand,

enabling the population to double in 44 years and increase 4.8 fold in a century.

Canada showed a decrease in birth-rate up to 1895 since when it has increased but curiously the death-rate showed a similar change.

The British Registrar-General reports on 29 countries. In 18 the birth-rate has fallen but the death-rate has likewise fallen in practically equal proportion. In 4 the birthrate is stationary (Russia, Roumania, Jamaica and Ireland). In these the death-rates and infantile mortality are also stationary. Russia with a birth-rate of 50 has highest death-rate of Europe, 36, and highest infant mortality rate 26 per cent.

In 4 the birth-rate has risen (Bulgaria, Ceylon, Japan, Ontario). In each the infant mortality and death-rates have risen.1

These figures have attracted much attention, particularly in Europe, where their significance is widely discussed; most of all perhaps in France, where the population has become almost stationary. In analyzing the situation it becomes clear that this decline varies with the economic status of the different groups. The accompanyBirth Rate in France 2

1 DRYSDALE, C. V. The Small Family System, p. 51 ff.

2 SCHALLMAYER. Vererbung und Auslese beim Menschen, p. 125.

ing table shows this. The reader is warned not to assume that the difference of a few francs in taxes paid is the cause of the phonomenon.

Leroy-Beaulieu adds that in France in the better situated groups the birth-rate is relatively low and is much higher in the poorer groups. In Paris the Chief of the Statistical Bureau states that 34-52 children are born yearly to each thousand women (aged 15-50) of the richer classes while 95-108 children are born to each thousand women of the poor groups.

As further evidence, we may add a diagram 3 based on the returns from a number of European cities:

Birth Rate according to Degree of Wealth in Different Sections of European Cities

AVERAGE NUMBER OF ANNUAL BIRTHS PER 1000 WOMEN FROM 15

There is a very widespread belief that this is true in the United States also and there is much evidence thereof in spite of the deplorable condition of our vital statistics. The following figures are to the point:

Kuczynski estimates the births in the native population of Massachusetts at 63 to each 1,000 women of child-bearing years. The French figure is about 85, English, 104; Russian, 143. If French population is stationary, New England must be actually losing.1

3 BERTILLON, J. Quoted by Chatterton-Hill - Heredity and Selection in Sociology, p. 325.

4 KUCZYNSKI, R. Quarterly Journal of Economics. Nov., 1901, and Feb., 1902.

Based on census figures it has been shown that in certain selected districts 13.1 per cent of the native white women who had been married from 10 to 20 years had borne no children, while only 5.7 per cent of the women of foreign parentage were childless. One woman in eight of the native stock was childless, as compared to one in twenty of the foreign. Of the second generation in America, 6.3 of the women had no children. The white natives had borne a child every 5.3 years, foreign whites every 3.2 years. Women of English parentage had a child every 4.2 years, those of Polish parentage every 2.3 years. City communities show highest percentage of those having no children. The The average number of years married per child born was 3.8 in Rhode Island; 3.5 in Cleveland; 3.6 in rural Ohio; 4 in Minneapolis and 2.8 in rural Minnesota. While these figures may not apply to the United States at large, they indicate certain real differences and changes of great meaning."

"I recently examined 150 Irish families in northern Pennsylvania, who are now in the second American generation. These families were of the best classes that emigrated about the time of the famine of 1847. With two exceptions (where the fathers became drunkards here) they gave the children the best example; they all succeeded financially so that their children were well fed, well housed and educated: nearly every family was able to send some of its sons to college. In the first American generation there were 5.52 children as the average to each family. If there were a full progression at this rate, these fifty families should now be represented by 1,523 persons. The total, however, in the second generation will not reach

5 HILL, J. A. In Quarterly Publications of American Statistical Society, Dec., 1913, pp. 583-604.