As the census authorities have repeatedly stated, the data regarding the number of living children under 1 year of age are also utterly incomplete and inaccurate, owing to the fact that the number so returned is too small in practically all localities, partly owing to the practically universal tendency to report children in the later months of the first year as 1 year old. For these reasons it would be utterly futile to attempt the compilation of any figures of the infant mortality rate in the United States at large for purposes of comparison with the returns for other countries presented in the preceding tables. In the Registrar-General's annual reports for some years past, returns of this character from thirty-two countries have been presented, but the United States has been the one great country in the world for which no figures were given.

In the light of these conditions it might at first seem practically impossible to obtain even an approximate idea of the status of infant mortality in this country, but such is not the case, the number of registration States and the magnitude of their combined population being sufficient to afford a fairly accurate index of the conditions in Continental United States as a whole. The registration systems of these States greatly differ in point of comprehensiveness and reliability, but by common consent that of Massachusetts is regarded as of foremost importance, and the complete record of infant deaths in that State for the last half-century undoubtedly affords by far the best available standard of measurement and comparison in a study of infant mortality in this country. In the Twenty-eighth Annual Report of the State Board of Health of Massachusetts, published in 1897, there appeared a comprehensive study of "The Vital Statistics of Massachusetts-A Forty Years' Summary," which was prepared by Dr. Samuel W. Abbott, secretary of the Board of Health. This chapter of more than 100 pages (pp. 711-829) begins with a graphic tracing of the "Marriage, Birth, and Death Rates and Infantile Death-rate, Massachusetts, 40 Years, 1856-95," and contains a complete tabulation of "Infant Mortality, Massachusetts, 1856-95, Forty Years," presenting the annual figures for each of those years.

For some reason not explained in the text, the birth statistics in that tabulation begin with July 1, 1856, and end with June 30, 1895, whereas the deaths under 1 year in the same table are taken from the calendar-year records, thus making the birth and death rate figures materially differ from those in the twenty-year infant mortality record presented in recent Massachusetts registration reports of births, marriages, and deaths. It has seemed desirable to eliminate this discrepancy, and the following tabulation of infant mortality in Massachusetts for the fifty years ending Dec. 31, 1905, has therefore been compiled in part from Dr. Abbott's table of infant mortality (p. 750), in so far as deaths under 1 from 1856 to 1895, inclusive, are concerned, and partly from his tabulation of marriages, births, and deaths from 1842 to 1895 (pp. 721-722), the supplemental figures for births and infant deaths in the calendar years 1896 to 1905, inclusive, being taken from the twenty-year table of infant mortality in the current (sixty-fifth) Massachusetts Report of Births, Marriages, and Deaths (p. 205). The composite tabulation of infant mortality herewith presented is thus made complete for the fifty calendar years ending with 1905, and, in order to permit of comparison with the statistics of foreign countries presented in the previous tables accompanying this article, is supplemented with a column containing the annual birth-rates in Massachusetts for the last fifty calendar years, as presented in the Sixty-fifth Massachusetts Report of Births, Marriages, and Deaths (pp. 141-142). Still-births have been excluded in all cases.

A COMPLETE RECORD OF BIRTHS, DEATHS UNDER AGE 1, THE INFANTILE DEATH-RATE PER 1,000 BIRths, and the BIRTH-RATE IN MASSACHUSETTS FOR EACH OF THE FIFTY YEARS 1856-1905, INCLUSIVE, EXCLUDING STILL-BIRTHS.

Thanks to the early establishment of the registration system of Massachusetts, recognized the world over for many years as the most reliable index of American vital statistics, the preceding table unquestionably affords by far the most comprehensive and most authoritative tracing of infant mortality in at least one section of this country which is now obtainable from any or all sources. The pronounced annual fluctuations in both birth and death rates are somewhat misleading, however, and the appended tabulation of births and infant deaths by five-year periods not only puts the case much more comprehensibly, but also reduces the Massachusetts tabulation to the basis followed in the preceding foreign tabulations, and thus makes possible a comparison by five-year periods. Thus arranged, the statement of births and infant deaths in Massachusetts for the last half-century is as follows:

TABLE VII.

BIRTHS, BIRTH-RATES PER 1,000 POPULATION, AND DEATHS UNDER 1 YEAR AND THEIR RATE PER 1,000 BIRTHS IN MASSACHUSETTS BY FIVE-YEAR PERIODS for the FIFTY YEARS 1856-1905, INCLUSIVE-STILL-BIRTHS EXCLUDED IN BOTH CASES.

When the summary for the last twenty-five years, in the last line of this table, is compared with the corresponding figures for foreign countries presented in Table III of this paper, one is immediately impressed with the surprising uniformity of the infant mortality rate the world around, which has already been alluded to. In the thirty-one foreign countries, in widely remote parts of the world, dealt with in Table III, the general average of deaths under 1 year to each 1,000 births in the twenty-five years ending with 1905 was 154: in the same period the infant death-rate in Massachusetts was 153.7. In the twenty European countries whose returns are presented in Table III-Austria, Hungary, and Russia, with their abnormally high death-rates, included-the average infant death-rate for the last twenty-five year period was 162 as compared with the Massachusetts rate of 153.7, and for the five five-year periods involved the European infant death-rates were, in order, 163, 162, 169, 162, and 153, as compared with death-rates of 160.1, 160.6, 161.2, 153.3, and 138.1 in Massachusetts. In fact, the correspondence between the infant death-rates of Europe and its leading countries and those of Massachusetts is so strikingly close that it can only be appreciated by means of a tabular statement, such, for instance, as the following:

TABLE VIII.

A COMPARISON OF THE INFANT MORTALITY RATES PER 1,000 BIRTHS OF THE WORLD AT LARGE AND LEADING EUROPEAN COUNTRIES WITH THOSE OF MASSACHUSETTS BY FIVE-YEAR PERIODS, 1881-1905, INCLUsive.