Although the available statistics are meager, an attempt was made to approximate roughly the total number of illegitimate births in the United States each year. It was impracticable to arrive at such an estimate by the method based on the total number of live births in the United States, since incomplete birth registration makes it impossible to obtain, for the country as a whole, statistics having any degree of accuracy. Hence, the estimate was based on the number of single, widowed, and divorced women of child-bearing age. In the United States in 1915 the estimated number of single, widowed, and divorced white women 15 to 44 years of age was 8,769,000. In the 16 States for which figures of illegitimate births were obtained the rate per 1,000 single, widowed, and divorced white women of childbearing age may be estimated as at least 3.7.2 Applying this ratio to the estimated population given above gives 32,400 as an estimated number of illegitimate white births in the United States each year. It must, of course, be remembered that this figure is an estimate based on only a part of the country; it is improbable, however, that the true figure is below it. Indeed it may safely be considered as a minimum estimate because of incompleteness of birth registration and erroneous registration of illegitimate as legitimate births.

Because of the recognized inadequacy of birth registration in a considerable part of the United States, another estimate was made based on data from States included in the birth-registration area in 1915. By the method described above the rate for white unmarried

1 The number of single, widowed, and divorced white women 15 to 44 years of age on July 1, 1915, was estimated by projecting the annual increase in this group between 1900 and 1910, using the arithmetical method. Strictly speaking, the figures for " white women" refer to women of all races other than Negro. The figure 8,769,000 includes approximately 0.2 per cent of Indians, Chinese, and Japanese.

The rate per 1,000 white and Negro women in the group specified, exclusive of the Negro women in two States, was found to be 4.3. (See Table V, p. 21.) The rate per 1,000 white women in this group can be found by assuming, in the absence, for most of the States, of illegitimate births classified by color, that the ratio between the white and Negro illegitimate birth rates was the same as that in the area of good birth registration, Maryland and Philadelphia combined, where illegitimate births are classified by color. (Maryland was not in the birth registration area in 1915 but was admitted in 1916.) Applying this ratio to the respective white and Negro populations in the group specified results in the figure of 3.7 for the white unmarried female population of child

women of childbearing age in these States may be estimated as 4. This rate, applied to the number of single, widowed, and divorced white women of childbearing age in the United States, gives a total of 35,100 illegitimate white births. It must be borne in mind that States in the birth registration area have a disproportionate urban population among which the illegitimate birth rate is high. Nevertheless, this figure may be regarded as an understatement because of errors and omissions of registration of illegitimate births even in States included in the birth-registration area.2

The rate per 1,000 white and Negro women in the group specified was 4.6; the correction has been made on the same basis as described in note 2, p. 26.

A careful study of records of social agencies in Boston added one-eighth to the number of illegitimate births registered as such in the city. Similar estimates for Negroes would be subject to a much greater percentage of error than are estimates for whites, first, because of inadequate registration in areas where the Negro population is largest, and secondly, because of special conditions affecting Negroes. Such an estimate, if made, should of course be accompanied by a full discussion of the history of the problem and of the difficulties involved in setting up new standards of family morality in place of those existing under slavery conditions.

INFANT MORTALITY.

Infant mortality statistics are an index not merely of the number of infants who fail to survive their first year, but of the conditions surrounding infancy and early childhood. The significance of the infant mortality rate is not alone in the sacrifice of infant life, but perhaps of even greater social import are the impaired physical development, hardship, and social handicap likely to be the lot of many who survive. The unfavorable conditions surrounding children born out of wedlock are clearly shown by the fact that the infant mortality among them is invariably found to be far in excess of the mortality among infants born in wedlock. Most European countries recognize the significance of the relation between illegitimate birth and infant mortality and furnish comparative statistics on the deaths of infants of illegitimate and of legitimate birth. Table VIII gives for 13 European countries the infant mortality rates among children of illegitimate and of legitimate birth, and the relative differences between the mortality rates among infants born in wedlock and those born out of wedlock.

TABLE VIII.—Average infant mortality rates for legitimate and illegitimate births in specified countries of Europe, 1910 to 1914.

a Derived from statistics in Annuaire International de Statistique. II. Mouvement de la Population (Europe). L'Office Permanent de l'Institut International de Statistique, La Haye, 1917, pp. 158-160. Annual average for the quinquennial period 1910-1914 in all countries with the following exceptions: Austria, France, Scotland, and Sweden, 1909-1913; Belgium, 1908-1912.

Rate in column 1 divided by rate in column 2.

e Mackenzie, W. Leslie, M. D.: Scottish Mothers and Children, being Vol. III of Report on the Physical Welfare of Mothers and Children. The Carnegie United Kingdom Trust, East Port, Dunfermline, 1917. p. 133. Annual average 1909-1913.

The infant mortality rate among infants born out of wedlock is without exception higher than among other infants, ranging from 1.3 to 2.3 times that found among infants of legitimate birth. The four countries having the lowest infant mortality among children of legitimate birth-Norway, Sweden, Denmark, and Switzerlandalso had the lowest mortality among infants born out of wedlock. Nevertheless, in these countries the illegitimate infant mortality rates were from 1.7 to 2 times as high as the rates among infants of legitimate birth. The countries showing the lowest relative difference between illegitimate and legitimate infant mortality rates-Austria, Belgium, and three States of the German Empire-had extremely high infant mortality among infants of both legitimate and illegitimate birth.

In Norway the infant mortality rate among infants born in wedlock during the quinquennial period 1910-1914 was only 62, whereas the corresponding rate for infants born out of wedlock was 122. As an argument for the passage of the Norwegian law relating to "children whose parents have not married each other," the Government made an exhaustive study of the infant mortality among these children. The conditions imposed by the law passed in 1915 are expected to effect radical changes in this respect, through insuring for the children born out of wedlock maternal care and adequate support either by the parents or by the State. In an article on this measure the author of the Norwegian law says:1

This demand for the increasing of a father's duties to his illegitimate child was considerably strengthened by the result of the official investigation into the economic and social position of unmarried mothers and their children. Those investigations * * * presented a dark picture of the existing conditions, especially in regard to stillbirths and infant mortality. It was stated that the stillbirths in the years 1891-1900 amongst the illegitimate children were 164 to 165 compared with 100 stillbirths among the legitimate children. The district doctors stated the reasons for this to be-besides venereal disease in several towns-unsatisfactory obstetrical help, concealment of birth, and the mother's needy position during pregnancy. ** * ** A still graver impression is produced by the investigation of the mortality of children born out of wedlock. It is stated that the mortality of illegitimate children in proportion to that of the legitimate ones is in the first month of life 195 to 100; in the second month of life 239 to 100; in the third month of life 274 to 100-nearly three times as great.

The annual report of the registrar-general of England and Wales shows for the year 19152 a mortality rate among infants born in wedlock of 105 per 1,000 births, and a mortality rate of 203 per 1,000

1 Castberg, J.: "The children's rights laws and maternity insurance in Norway." Journal of the Society of Comparative Legislation. (New Series.) Vol. XVI, Pt. 2 (1916). pp. 290-291. London.

2 Seventy-eighth (1915), Seventy-ninth (1916), and Eightieth (1917) Annual Reports of the Registrar-General of Births, Deaths, and Marriages in England and Wales. London,

births for infants of illegitimate birth. During the year 1916 the infant mortality among babies of legitimate birth was reduced to 87, while the rate among babies of illegitimate birth was 183. In 1917 the rates showed an increase to 90 for children of legitimate birth and to 201 for those born out of wedlock. The registrar-general has persistently called attention to the meaning of these comparative infant mortality rates, and there is evidence that public opinion in England has been aroused to a realization of the necessity for correcting the existing conditions by providing through legislative action and constructive social measures better protection, support, and care for these infants.

In the Report on the Physical Welfare of Mothers and Children in Scotland Dr. W. Leslie Mackenzie, after discussing the difficulties surrounding unmarried mothers, says in regard to their children:1

In many respects it is less difficult to make provision for the mothers than for the annual crop of 8,600 children. The mothers, however they are provided for, can at least fight for their lives and often attain to a passable living. The newborn infant can do nothing for himself. He hangs on the service of others. Within hours of his birth he may be taken from his mother's breast and put among those whose skill is often no substitute even for an unskilled mother's care. He may pass from hand to hand and from place to place. Besides the risks he has encountered in coming to birth, he encounters a thousand others that fall only to the children of the unmarried. If, starting from the same line, he loses in the race with his legitimate fellows, it is from no fault of his own. He has not sinned, but he comes short of the glory. That is the tragedy of the unmarried mother's child.

2

In New South Wales the infant mortality rate among children born out of wedlock in 1916 was 145.9, as against 63.9 for children. of legitimate birth. For the years 1906-1915 the rate for the former was 170.4, and for the latter 68.1. In his report for 1917,3 the president of the State Children Relief Board discusses the desirability of providing special homes for babies with mothers, both for the sake of the mother's training in infant care and of her moral development and for the safeguarding of the life of the child during the first critical period of infancy. In the State of Victoria the discrepancy between the mortality rate of infants born out of wedlock and of those of legitimate birth is even greater than in New South Wales. The report of the Government statistician sums up the situation thus:4

1 Mackenzie, W. Leslie, M. D. Scottish Mothers and Children, being Vol. III of Report on the Physical Welfare of Mothers and Children. The Carnegie United Kingdom Trust, East Port, Dunfermline, 1917. p. 131.

2 State Children Relief Board: Report of the President, Alfred William Green, for the Year Ended 5 April, 1917. Sydney, 1918. p. 26.

3 State Children Rellef Board: Report of the President, Alfred William Green, for the Year Ended 5 April, 1917. Sydney, 1918. p. 24.