A DISCUSSION OF AGE STATISTICS HY ALLYX A. YOUX(i A Thesis Submitted for the Degree of Doctor of Philosophy in the Department of Economics of the University of Wisconsin. A DISCUSSION OF AGE STATISTICS BY ALLYX ABBOTT YOUXG A Thesis Submitted for the Degree of Doctor of Philosophy University of Wisconsin i 902 [REPRINT OF BULLETIN 13 OF THE BUREAU OF THE CENSUS] WASHINGTON 1904 jUN 9 1905 XotD, DEPARTMENT OF COMMERCE AND LABOR BUREAU OF THE CENSUS S. N. D. NORTH, DIRECTOR BULLETIN 13 A DISCUSSION OF AGE STATISTICS WASHINGTON GOVERNMENT PRINTING OFFICE IQ04 CONTENTS. Page. Letter of transmittal - 5 Summary of results ", 8 The inquiry , 8, !1 Significance of age statistics 9-11 Errors in the reported ages 11-21 Unknown ages 14, 15 Age groups __ 15, 16 Children's ages 16-20 ( Jentenarians 20, 21 Age constitution of the population ' 21, 45 Median and average ages 21-25 Productive and nonproductive ages 25-30 The distribution of the population in 10-year age groups ; >o '■'•'■' The number and proportion of children iu the population 40—43 Other age groups 43 The proportions of the sexes in different agi groups.. 4:; 15 Appendix A. The adjustment of the returns 47-51 Appendix B. Bibliography of discussions of age statistics 53 Diagram 10 Map — Median age of the total population, for stati - and territories 24 (3) LETTER OF TRANSMITTAL DEPARTMENT OF COMMERCE AND LABOR, 111 REAXJ OF THE CeNSI S, Washington, D. ('.. November 1, 190J.. Sir: 1 have th(> honor to transmit herewith ( lensus Bulletin L3, entitled "A Discussion of Age Statistics." It has been prepared under the supervision of Professor Walter F. Willcox, of Cornell University, special agent of the Bureau of the Census, by Allyn A. Young, assistant professor of finance in Dartmouth ( lollege. Professor Young was employed for more than a year in the Census Office and much of his time was spent in studying the age statistics of this and other countries and various mathematical and statistical questions growing out of them. As long ago as 1870 the Census Office perceived the importance of the problems with which this bulletin deals, and in that year secured from the actuary of the Treasury Department a careful study of a part of the field. Since that time comparatively little consideration has been given to them in the United States, but in various European countries, as well as in India and Australasia, mathematicians and statisticians have given them attention and published the results. Professor Young writes after a critical analysis of their investigations, and in the present bulletin he lias explained and interpreted the age -tali-tie- of the Twelfth Census, making the treatment of the subject in this country, it is believed, fully abreast of the best discussions elsewhere. It is hoped that the bulletin will lie found to constitute a distinct advance in the interpretation of this difficult subject. This bulletin comprises four main divisions: (1) A description of the way in which the information as to the ages of the population was obtained, together with an account of the age tables that appear in the previously printed results of the Twelfth Census. (2) A short discussion of the general significance of age statistics. (3) An examination of the amount and nature of the errors in the reported ages. (•4) A discussion of some of the more important facts which the census reveals with reference to the age constitution of the population. The topics considered are the median and average ages of the population, the distribution of the population into productive and nonproductive age groups, the population in 10-year age periods, the number of children in the population, and the sex distribution of the population in various age groups. The age constitution of the population of continental United States, classified by sex, nativity, and race or color, is studied for the censuses of 1900, L890, and 1880. For the census of 1900 the discussion is extended to include the ages of the aggregate population of the states and territories, and of the aggregate popu- lation living in cities and in rural districts, classified by sex. An adjustment of the age returns for continental United States, classified by sex, race, and nativity, and a brief bibliography are printed as appendices. The practical significance of these inquiries is shown by the statement that one of our highest American authorities has sought by an analysis of the census age figures to show that in all recent censuses there had been serious errors in the enumeration of young children, amounting among those under five years of age to at least a million omissions both in 1890 and in 1900. This argument is traversed by Professor Young, who reaches the gratifying conclusion that" there is no conclusive evidence of material omissions either in 1890 or 1900. Very respectfully. Director. Hon. Victor H. Metoalf, Secretary of Commerce and Labor. (5) A DISCUSSION OF AGE STATISTICS. Bv Allvk A. Young, l'h. I>. SUMMARl OF RESULTS. J The ages of the population were ascertained more accurately at the census of 1900 than at any previous census of the United States. This improvement was largely due to the addition of an inquiry as tp date of birth to the former direct question as to age. The age returns of the more illiterate classes of the population are less accurate than those of the more intelligent classes. Moreover, the distribution of errors in the age returns is such as to suggest that the census information obtained about the more intelligent classes in districts in which there are large numbers of illiterates is less accurate as a rule than information obtained about the same class of persons in distL ; an innovation in the practice of the United States census. Prior to L850 the population was returned as within specified age groups. At the censuses of L850, 1860, 1870, ami 1880 the enumerators were instructed to ascertain the age at the lasl birthday previous to the taking of the census. In 1890 the inquiry was modified to "age at nearest birth- day." At the census of 1900 the older form of inquiry was restored primarily because it accords with popular usage, but also because it gives results comparable with those of the censuses of other countries. In this con- nection it should l>e noted that the census age inquiry shows, not how many persons tire at a certain age. hut how many are in definite age periods — that is, between n and n4- 1 years old. 1 The tabulations of the results of this inquiry, printed in Twelfth Census, Volume II, are as follows: Table XVI. — Ages of the aggregate population of the mainland of the United States, classified by sex, general nativity, and color. Table XV'II. — Ages of the Chinese, Japanese, ami Indian popula- tion of the United States, classified by sex. Table 1. — Ages of the aggregate population of the United states, classified by sex, general nativity, and color. Table 2. — Ages of the aggregate population, classified by sex, gen- eral nativity, and color, by states and territories. Table 3. — Ages by periods of years of the aggregate population, classified by sex, by states and territories. Table 4. — Ages by periods of years of the native white population of native parentage, classified by sex, by states and territories. Table 5. — Ages by periods of years of the native white population of foreign parentage, classified by sex, by states and territories. Table 0. — Ages by periods of years of the foreign white popula- tion, classified by sex, by states and territories. Table 7. — Ages by periods of years of the colored population, classified by sex, by states and territories. Table S. — Ages by periods of years of the negro population, classi- fied by sex, by states and territories. Table 9. — Ages bj periods of years of the aggregate population, classified by sex, general nativity, and color, for cities having 25,000 inhabitants or more. Tables 10-18. — Persons of school age, by sex, general nativity, and color, by states ami territories. Tables 19-21. — Persons of school age, by sex. general nativity, and color, for cities having 25,000 inhabitants or more. Table 22. — Males of militia age, by general nativity and color, by states and territories. Table 23. — Males of militia age, by general nativity and color, for cities having 25,000 inhabitants or more. ■With reference to the method of obtaining the ages of the population, European census practice is divided. In the majority of the countries, however, the date of birth is asked, while in the others, "age at last birthday" is the form of the question. In the Italian census of lsso the two inquiries were combined, as in the United States census of 1900. At the St. Petersburg ses- sion (1S72) of the International Statistical Congress, the following recommendation was passed: "Whenever the degree of education permits, ami especially in large cities, age should be indicated by the year ami month of birth. When age is expressed in years, completed years should be understood; for children less than a year old, age should h<- expressed by the number of completed months." (SeeCompte Rendu, Vol. II, page 425. ) In the state census of .Massachusetts, and, since 1S85, in that of Rhode Island, the age inquiry has been as to the "nearest birthday." Table 24. — Males of voting age, by general nativity and color, by states and territories. Table 25. — Males of voting age, by general nativity and color, for cities having 25,000 inhabitants or more. Table 26. — Persons of school, militia, and voting ages, by sex, gen- eral nativity, and color, by comities. Table 27. — Persons of school, militia, and voting ages, by sex, gen- eral nativity, and color, for places having 2,500 inhabitants or more. An analysis of the general age tallies presented in Volume II, containing a number of summary and ratio tables, is found on pages xxxv to Ixxviii of that volume. Volume III, Table 27, pages <'>*:! to 687, shows the num- bers of the aggregate population of the United States at each specified age. per 100,000 of known ages, by sex, color, general nativity, and parent nativity. Table 28, pages 689 to 695 of the same volume, shows the num- bers of the aggregate, the white, and the colored popu- lation in each age group, per 100,000 of known ages, by states and territories. The age tables contained in the Abstract of the Twelfth Census will be found especially useful for the purpose of supplementing those contained in the present discussion, as they are based on a different system of grouping. In addition to the tables which relate primarily to age, there are tables which give other classifications of certain age groups. In the tables relating primarily to citizenship and years in the United States, which are to be found in Volume 1, pages 907 to 1006, males of voting age are classified by general nativity, parentage, color, and literacy, and foreign born males of voting age by citizenship, country of birth, number of years in the United States, and ability to .-peak English, for states and territories and for cities having 25,000 inhabitants or more; and males of voting age by general nativ- ity, color, citizenship, and literacy, for counties. A more or less detailed classification by ages has been introduced into the general tables on conjugal condi- tion, school attendance, illiteracy, ability to speak English, and occupations. These tables are printed either in Volume II or in the special report on occupa- tions. In Volume III, pages xlvi and xlvii, there is a table showing, for the registration area and its sub- divisions, the white population in each of eight age groups, classified by birthplaces of mothers. In the same volume, pages 285 to 555, is a table which shows the number of children under 1 year of age and under 5 years of age. classified by sex, general nativity, and color, for each state and registration city, and for the groups, counties, and local divisions of registration states. SIGNIFICANCE OF AGE STATISTICS. For the purpose of a scientific study of the popula- tion the classification by age is only less important and fundamental than that based on sex. Not only does this classification afford information concerning the genera- tion which is at present doing the world's work, but it also makes it possible to form some conclusions about the generation next to enter on its estate. It is scarcely correct, how T ever, to speak of "generations' 1 in the sense in which the word is ordinarily used, because the population is being constantly recruited by births and depleted by deaths. Professor von Mayr has compared the population of a country to an endless rope made up of a vast number of individual threads. 1 If the rope be cut at any given point, the threads thus exposed to view will vary from those so short that they can scarcely be meas- ured to those of a maximum length. This is the view presented by the census age tables — a view which em- phasizes the ever-changing character of the population. The classification of a population by age makes it possible to measure its economic and military strength. A population containing a large proportion of aged persons and children may be less efficient than a smaller population in which a larger proportion are of produc- tive age. Moreover, the proportion of young children in the population is at least as valuable an index of social conditions as is the birth rate, and is of special utility in the United States, where accurate records of the births ate kept in few localities. The chief useful- ness of age statistics, however, is found in the increased value which they oive to the results of other census inquiries. The combination of the age and sex classifi- cations reveals the number of voters and the number of potential fighting men. It is impossible to compare different classes of the population accurately with ref- erence to .itch tacts as crime, pauperism, literacy, number engaged in productive occupations, mortality rates, etc.. without taking into account the differences in their age constitution; for example, the mere fact that death rates are higher in cities than in rural dis- tricts i-; without special significance until the differences in tin age constitutions of the two regions are taken into account. The age composition of a population on a given date may be represented graphically by a diagram sometimes i-tlled the age pyramid, but which might more accu- rately be called the age triangle. The example given on page 10 illustrates in this way the age composition of tlie population of the United States at the date of the Twelfth Census. The vertical line running through the apex of the triangle is divided into 100 equal parts, one for each year of life. From this line horizontal distances are measured representing, according to the indicated scale, the number of persons reported at the specified age. Distances to the left represent male-, to the right females, and the horizontal distance between the sides of the triangle at each point the total popula- tion of the specified age. Each complete square within the triangle thus represents 2,000,000 people and the entire area of the triangle represents the total popula- tion of the United States of known age, classified by sex and age. 'Statistik und Gesellschaftelehre, Vol. II, pagi 10 2 2 HUNDREDS OF THOUSANDS 10 The age constitution of the population is affected by three factors — births, deaths, and migration. If a country were so isolated as to be affected neither by immigration nor by emigration, if the number id' births in each year were the same as in the year preceding, if the number of deaths were uniform for each sui i sive year and for each year of aye. and if the conditions for male- and females were identical, the aye consti- tution of the population would be of the simplest and most regular form. If a figure constructed on the same principle as this diagram were drawn to r'epresenl such a population, it would take the form of an isos celes triangle. But a condition of this kind is purely imaginary, and is never realized in actual conditions. First and most fundamental in causing deviations from this theoretical distribution is the fact that neither the number of deaths nor the death rate is constant for all ages. The death rate is very high in the first year of life, decreases rapidly through about the first twelve years, then increases gradually until the period of old age is reached, when it increases with rapid accelera- tion. If this factor lie taken into account — theassump- tion still being that the population is affected to only a slight extent by migration, ami that the number of births in successive years is fairly constant — the graphic representation of the age constitution will be bell- 11 shaped, the sides of the figure changing from concave to convex in the period of advanced age. These arc the conditions represented in the ordinary life table, which, taking as a basis a number of persons beginning life at the same time, shows what would be the number <>f survivors at each age under prevailing conditions. Such a population would be called a " normal stationary population." The population of France approaches this condition more nearly than that of any other coun- try for which statistics are available. Rut in most countries the annual number of births is not a constant factor. If the birth rate is constant, an increasing population tends to perpetuate its increase through the continued augmentation of the number of births; and even with a diminishing birth rate the absolute number of births from an increasing popula- tion may be increasing. This condition, which would be shown in the graphic representation by a relatively broader base of the "age pyramid," is prevalent in most European countries, but more noticeably in the United States. Factors of less importance are the changes in birth and death rates, due to varying economic, social, anil climatic conditions. These give to the diagram an irregular form, but their effects on the age constitution are less noticeable in a large and heterogeneous popula- tion like that of the United States than in smaller populations. In addition to these primary factor- of birth and death rates, then 1 is one secondary factor, migration, winch has an important influence upon the distribution of ages. In the decade 1890 to 1900, 7". per cent of the immigrants to the United States were between the ages of 1") and 40 years, L6 per cent were less than 1.".. and only '.) per cent were more than 40 years old. In a country where the excess of immigration over emi- gration is as large as it is in the United State-, this preponderance among the immigrants of persons be- tween 15 and 4*) years of age leads to a marked widen- ing of the age pyramid at and immediately above these ages. This effect is of course strikingly apparent in the age constitution of our foreign born population. 1 Iuternal migration lias a corresponding effect on the age constitution of particular areas, and thus leads to marked differences between cities and rural districts in the distribution of ages. 'Twelfth Census. Vol. II, Plat,- 2. ERRORS IN THE REPORTED AGES. Age may be defined as the sum of time lived since birth. Modern census practice follows ordinary usage in stating the ages of children less than one year old in months and the ages of all other persons in years. A casual inspection of the tables in which the ages are given by single years will show that while the number reported decreases as the age increases, this decrease is far from constant; indeed, the number reported as of a given age is not always smaller than the number reported at the preceding age ' The facts are shown even more clearly in the diagram. It is not probable that in the age constitution of the actual pop- ulation the principle of continuity is so far violated that the age series could not be represented by a fairly smooth curve; indeed, experience has shown that the greater the accuracy of a census, arising from special care in the enumeration, or from the intelligence of the persons enumerated, the more nearly does the curve representing the age returns approach smoothness of form. Most of the irregularities in the curve repre- senting the ages actually reported take the form of angular deflections, corresponding to abnormal values of single terms. Consideration of the factors which influence the age constitution of the population gives no warrant for supposing that these irregularities corre spond to the facts. It is certain, of course, that changes in birth and death rates, as well as variations in the amount and direction of migration, must leave their marks in the form of certain irregularities; but these irregularities would in most cases be spread over a term of several years, and in the curve representing the age -cries they would take the form of flexures covering a period of several terms in the series. It follows that the irregularities in the age table- represent, for the most part, erroneous information. A knowledge of the nature and amount of the errors is a prerequisite to the intelligent use of census age statistics. Before entering upon a discussion of the particular forms which these errors take, it will be well to get some idea of the total amount of error. Comparisons with previous censuses, with reference to this point, will make it possible to ascertain the effect of the addi- tion of the inquiry as to date of birth to the inquiry as to age. 'Twelfth Census, Vol. II, Table xvi. 12 If tin- number returned at each year of age (except, of course, the number reported a^ "under 1") be subtracted from the number returned at the next lower year of age, the remainders will l>e the "first differ- ences" of the age series. If these firsl differences be added, with due regard to plus and minus signs, the .sum will, of course, be equal to the difference between the number reported as less than 1 year old and the number classified as " loo and over." But if the signs be disregarded in the addition, the sum of the differ- ences will be equal to the difference between the num- bers reported at the first and at the last age, plus twice the sum of the minus differences, it may l>e assumed that in a series representing the true ages of the popu- lation the minus differences are zero, or approximately zero, and the sum of all the differences would therefore be the same whether plus ami minus signs were regarded or not. The difference between the sum of the differ- ences in the reported and in the actual ages would be, therefore, approximately twice the sum of the minus differences, h follows thai (he sum of the minus dif- ferences may be properly regarded as a measure of the inaccuracy of the reported ages. It will be noticed that in this method no assumption is made about the true nature of the age scries, except that the minus differences are approximately zero. Even if this ap- proximation is not very close, the general validity of the method is not affected. For the foreign horn ele- ment of the population, however, this method can not be used, as the fact that there are comparatively few foreign born children among the population produces an unduly large number of minus differences. Table 1 shows these differences computed, as described, for the other classes of the population, at the Twelfth Census and the two censuses immediately preceding; it shows, also, for each of these population classes the total number of persons less than loO years of age and the per cent which the sum of the minus differences forms of the total number of persons less than 100 years of age. Table t.— SUM OF THE MINI'S DIFFERENCES IN THE AGE RETURNS OF THE AGGREGATE POPULATION AND OF THE TOTAL NATIVE, THE NATIVE WHITE J'.Y PARENTAGE, THE TOTAL COLORED, AND THE NEGRO POPULATION, AND PEE CENT THAT SUM FORMS OF PERSONS TO 99 YEARS OF AGE, CLASSIFIED BY SEN. FOB CONTINENTAL UNITED STATES: 1900, 1890, AND 1SS0. A.— NUMBER OF PERSONS TO 99 YEARS OF AGE. 1900 1890 1S80 B Males. Females. Both sexes. Males. Females. Both Males. . 37,102,733 62, 156, 104 ,1.767 25,517,411 24,634,356 65, 176,691 56,474,76! 10 B36.9 ■ : . .. 9,127, 180 8,782 630 :; :,0 6 2 I, (72,888 7,831,720 1,584, 17.5 32,400,688 27,870,153 20,064,050 ;. -.I-. 103 1,543,005 1 122, 126 Native white Native white— native parents ! >,764,84 34,269 ' 11,494,900 7,594,409 17,413,505 5,776,719 L6 ,6,440 5,718, 1H 36, 842, 699 IS. 609, 028 18,233,671 6,749 . - 3.386.904 3,362,848 B.— SUM OP MINUS DIFFERENCES IN REPORTED AGES. 11100 IS'.IO 1SS0 Both sexes. Males. Both sexes. Males. I'Vm.'iliv Until sexes. Males. 2,552,243 1,274,640 4,683, 2, 359. .595 'J. 184,674 4,133, i ,7 1,974,593 2,186,201 1 . 670, 9 10 810,987 627,525 216,247 956,780 912,977 B01, 180 419,689 326, 173 hit, mi 146,631 118.263 889,843 11 ,, ..:l 321,344 112,894 511,(l5h 196,045 2,458,985 1, 970, 051 503, 320 1,093 663 1,262, 128 1,019,113 251,896 1,554 864 1,053,393 282,907 1,664,318 789,676 923,770 1,264 l 18 571, 203 693, 704 C— PER CENT THAT THE SUM OF MINUS DIFFERENCES FORMS of PERSONS TO 99 YEARS OF AGE, OR COEFFICIENT OF ERROR. 1900 IS1I0 1SMI Both sexes. Females. Both sexes. Males. Both sexes. Males. Females. 1 3. 3 3 7,5 7. 1 - 1 8.2 7. 7 8.9 2.6 1.4 1.5 1. 1 10.5 10. 1 2.4 1.5 1.6 1. 1 9.7 9.6 2.7 1.5 1.6 1.4 11.2 11.2 .5.4 5.7 1.4 14.4 1 1. 1 13.2 5.9 6. 2 1.9 15.7 4.5 4.2 > IS, 7 17.0 20. 6 1 13 The absolute .sums of the differences are in themselves without special significance, on account of the varying sizes of the population classes for which they have been computed. The per cents indicate the relative degrees of inaccuracy in the age returns of the differ- ent classes, and may be called the "coefficients of error.'" Considering first the results for 1900, it appears that the smallest coefficient of error is found in the returns for the native white population of for- eign parents, while the largest is found in the returns for the negro population. In the returns for the latter class the coefficient for females is considerably larger than that for males, the excess being reflected in the results for the aggregate population. A comparison of the results for the three censuses shows that there has been a general gain in accuracy since 1880; for the aggregate population of both sexes the coefficient of error was 8.2 per cent in L880, T.."> per cent in 1890, and 3.4 per cent in 1900. The marked dim- inution in the coefficient of error in 1900 as compared with 1890 indicates that the inquiry as to date of birth had a very important influence on the accuracy of the results. For all classes of the population the returns for L900 indicate increased accuracy as compared with previous censuses. It will be noted that in 1S80 and 1S90 the age returns of females were, for all (lasses < if the popula- tion, more erroneous than those of males. The addition of the inquiry as to date of birth seems to have had a more marked effect on the age returns of females than on those of males. This is probably due to the fact that a larger proportion of women than of men were seen directly by the enumerators in their house-to-house visitation. The errors in the age returns of males are probably due, in larger measure than those of females, to ignorance of the true age of the person enumerated on tin' part of those who furnished the information to the enumerator. Naturally this kind of error could not be eliminated by any amount of care in the manner of asking the question. In the case of the native white population of native parents the coefficient of error in 1900 was less for the total population than for either the male or the female population. This seeming paradox is due to the neutralization of a certain amount of understatement of age on the part of one sex by over- statement on the part of the other. Table 2 shows the minus differences and per cents in L900 for the five main geographic divisions. Table 2.— SUM OF THE MINIS DIFFERENCES IX REPORTED AGES OF THE POPULATION AND PER CENT THAT SUM FORMS OF PERSONS TO 99 YEARS OF AGE, CLASSIFIED I'.V PARENT NATIVITY AND RACE, FOR MAIN GEOGRAPHIC DIVISIONS: 1900. CLASS OF POPULATION. PERSONS II TO 99 YEARS 01 IG1 [900 North Atlantic division. Native white Native white— na tive parents Native white — ior eign parents Negro 1 1,870, 165 9,891,395 5,978,770 382,635 South Atlantic division. 6, 095, 498 3,712,504 North South Central Central division, division. !07 11, 141,893 14,111,559 8 736,056 7,471,64* 490,971 4 166,696 Western division. 3,094,424 2.003,430 1.090.994 i THE MINIS DIFFERENCES IN 1:1 ■ OF THE POPULATION 1 North South Atlantic Atlantic division, division. North South Central Central division, division 260. 665 171,499 147,557 .'- 582 9,819 125,000 167.;' 15 92,276 32 182 West- ern di- vision. 57,929 13, i 15 16.331 2,292 PER CENT THAT BUM OF MINUS DIF- FERENCES FORMS OF PERSONS TO 99 YEARS OF AGE: 1900. North Atlan- tic di- vision. South North Atlan Cen- tic di-tral di- vision \ ision. South Cen- tral di- vision. 1.6 ,1 1.7 2.4 1.7 7.8 11. 1 1.1 1.0 1.2 l.S l.s 2.4 111. 3 West- ern di- vision. 1.9 '.!. '-' 1.5 7.7 The coefficients of error shown in Table 2 confirm the inference drawn from Table 1, that the reports of the ages of the colored population are especially untrust- worthy. In three of the five main geographic divisions the accuracy of the returns for the native white popu- lation of native parents is about the same as that of the returns for the native white population of foreign parents. In the South Central division, however, the returns for the native white population of foreign parents are distinctly less accurate than those for the native white population of native parents, while in the Western division this condition is reversed. The facts shown in Table 2 with reference to the geographic distribution of errors in the age returns are worthy of note. It will be observed that if the five geographic divisions were ranked in the order of decreasing size of the coefficient of error for the native white population of native and of foreign parents, and for the negro population, the order of rank would be South Atlantic, South Central, North Atlantic, West- ern, North Central, except that for the native white population of native parents the coefficient of error in the Western division would advance from fourth to second. For each class the greatest amount of inac- curacy is found in the South Atlantic division and the least amount in the North Central division. A comparison of the coefficients of error shown in Tables 1 and 2 with the census statistics of illiteracy shows a general correlation between the two sets of facts. 1 Illiteracy, like the errors in reported ages, is greatest among the negro population and least among the native white population of foreign parents. More- over, for the total population, illiteracy, like errors in age returns, is at its maximum in the South Atlantic- division and at its minimum in the North Central divi- sion. But this is as far as the correlation can be car- ried, for the geographic distribution of illiteracy is by 'For the per cent illiterate in the population of the main geo- graphic divisions, see Twelfth Census Abstract, Table 57. 14 no means so strikingly uniform for the different popu- lation classes as is the geographic distribution of the errors in the age returns. It would seem, therefore, that the presence of a large illiterate population in cer- tain parts of the country lias lowered the general accu- racy of the returns from those regions. Despairing of getting any but inaccurate returns from a large propor- tion of the persons in their districts, the enumerators in the more illiterate regions have been less painstaking in their efforts to secure accurate returns from the rest of the population. By far the most noticeable errors in the age returns are due to the tendency of the persons enumerated to give their ages in round numbers. Multiples of live — especially the even numbers — are the chief centers of concentration. This tendency is not of great impor- tance below the year 20, and the concentration on that year is, in the ease of males, offset by an easily explained concentration <>n the year 21. Table 3 fur- nishes a measure of the amount of this concentration on multiples of 5. It is assumed that in the actual population the number of persons aged 25, 30, 35, -to, 45, 5o. .".5. and •'>" years is approximately one-fifth of the total number between '_'■". and 62 years of age, inclu- sive an assumption which corresponds with the facts closely enough to give all necessary accuracy to the method. The per cent that the number reported at multiples of 5 forms of one-fifth of the total number in the age group 23 to t52 indicates the degree of con- centration on round numbers. Table 3.-MK\Sfl;l ol mi\. KNTKATION ON MULTIPLES OF 5 IN THE KEI'OKTED AGES OF ADULTS IX SPECIFIED CLASSES OF THE POPULATION OF CONTINENTAL UNITED STATES: 1900, 1890, AND 1880. CLASS OF )'< IP1 I ATION. TOTAL NUMBER OF PERSONS AGED , ;o, :;',, in, 15, 50, 55, ind 60 YEARS. ONE-FIFTH OF TOTAL NUMBER OF PERSONS A<-i:i> 23 TO 62 YEARS, INCLUSIVE. PER CENT THAT NUMBER RE- PORTED AT Mn.TIPI.ES OF 5 FORMS OF ONE-FIFTH OF TOTAL NUMBER AGED J3 TO t>2 YEARS, INCLUSIVE. 1900 1890 ISSII 1900 18SI0 1880 1900 lS'.MI ISSO ■ 1 6 tOl 7,079,822 6,026,970 6, 7112. 831 5,390,228 4,161,802 1 1 9. 8 131.3 144.8 5,061, 174 1,215,705 3,308,893 1,612, 172 1,105,605 % 1,587,382 1,490,345 715, 104 3, 529, 587 1 . 299, 775 ,-,C,n. self, 2,674,693 992,791 494, 318 iin ;: 126. 7 165.9 119.4 1 12 3 180.8 123 7 ] , 888, 58 1 1 186 343 !S0,l 1 hi i 080 162.4 223.7 The results presented in Table 3 show that of the three classes of the population considered the tendency to concentrate on round numbers is strongest in the col ored population, while the native white population is least affected by it. For all classes of the population there is a marked improvement in the returns for L900 as compared with those for 1890, and there was a corresponding improvement in the returns for 1890 as comparedwith thosefor 1880. A seeming anomaly lies in the fact that while the returns for 1890 are but little improved in general accuracy over those for L880, as is evidenced by Table 1. yet in the matter of concen- tration on multiples of 5 the improvement in accu- racy over the preceding census was quite as marked in 1890 as in 1900. The explanation is found in the fact that in 1890 a special effort was made to lessen this particular form of error. The instructions to enumerators on this point were full and explicit and evidently were responsible foravery material decrease in the concentration. 1 The further decrease in the amount of this inaccuracy which is shown in the census of 1900 is undoubtedly due to the addition of the inquiry as to date of birth. But it is evident that this particu- lar form of error was not reduced so materially as were other inaccuracies in the age returns. In those Euro- pean censuses in which the ages are obtained b\ ascer- 1 Instructions to Enumerators, Eleventh Census, reprinted in Wright and Hunt, History ami Growth of the United Stan- ( lensus, pages is? and L88. For the corresponding instructions for the Tenth Census, see Wrightand Hunt, page 171. taining the date of birth, it is found that there is a tendency to concentrate on those calendar years which are multiples of 5. ' : This tendency was undoubtedly operative in the United States in 1900, and on account of the census year being an even multiple of 5 the ten- dency to concentrate on certain ages and the tendency to concentrate on certain years of birth coincided— that is. so far as all persons bom in the first five months of the year are concerned. It seems probable that as high a degree of accuracy would have been obtained if the inquiry as to "age at last birthday" had been omitted. Unknown ages. Table 4 shows the proportions of the different classes of the population whose ages were returned as unknown. Table 4. — Number of persons of unknown ages in 100,000 of all ages, classified by st r,for continental United States: 1900 and 1S90. I '.Mill IS'.KI no> Both sexe Males, Females. Both sexes. Males. Females. i gregate s 328 197 2 9 323 192 Native white Foreign white . .. 212 244 641 138 186 567 210 270 537 275 593 i ii 192 480 The proportion of persons of unknown age is largest for the colored and smallest for the native white popu- ', i-on Vlayr, Statistik und Gesellsehaftslehre,Vol. II, page 75. 15 lation — a statement that holds true for each sex and each census. For all classes of the population this proportion is smaller for females than for males. This is undoubtedly due to the fact that a larger proportion of women than of men were seen personally by the enumerators. The excess of the proportion of un- known ages among the colored and the native white population in 1900 over the same proportion for 1890 indicates that the inquiry as to date of birth lessened the number of mere guesses that were accepted by the enumerators as statements of age. For the foreign white population, however, this proportion decreased in 1900 as compared with 1890. This decrease may be due to the. fact that on account of the excess of immi- gration in the decade L880 to L890 over the immigration for the ensuing decade, the average foreign born person in 1900 had been in this country longer and was, con- sequently, better qualified to understand the enumera- tors questions and to answer them intelligently. Age groups. — Combinations of the age classification by single j-ears with the results of other census inquir- ies are impracticable in most instances. In the major- ity of cases in which the census differentiates economic and social classes by age, recourse must be had to age groups. From the standpoint of accuracy a classifica- tion by groups is preferable to one by single years, and because of the excessive concentration of reported ages on years which are multiples of 5 a quinquennial grouping is especially accurate. This form of group has the added merit of convenience. In the census practice of all countries it is customary to make use of 5-year or of 10-year age groups. The usual form of quinquennial group is the one used in the United States census. In this method the live lowest ages of the table are placed in the first group, the next five ages in the second group, and so on to the end of the tabic. This form of grouping has been criticised by several statisticians on the ground that each group begins with a multiple of 5; that is, witli a year of concentration. 1 The critics recommend a quinquennial group in which the year of concentration will be in the middle of the group. It is for this reason that the English census defends the use of 10-year groups in which the multiple of 10 comes in the middle of the period, 2 and that the American census has followed the same prac- tice in its mortality statistics. The accuracy of an age group depends upon the ratio between the number which it includes and the number of persons whose true ages would be included within the same group. In other words, the choice of age groups will depend upon the manner in which concentration takes place. The most accurate form of age group, then, is one which includes with each year of concentra- 'Bertillon, Cours elementaire 3e Statistique, page 52, note; Mayo-Smith, Statistics and Sociology, pages 59 and 60; Holmes, in "The Federal Census," Publications of the American Economic Association, New Series, No. 2, pages 57 and 58. 2 Census of England and Wales, 1891, General Report, page 27. 4091— Bull. No. 13— 04-:— 3 tion the years from which this concentration is chiefly drawn. If the field of attraction of each age at which concentration occurs does not extend beyond the limits of one age group, the fact of concentration will not affect the accuracy of the groups. If the tendency to understate ages is materially greater than the tendency to overstate them, the form of group used in the United States census may be more accurate than the one in which the year of concentration is in the middle of the group. The fact that concen- tration is greater on multiples of 10 than on odd multi- ples of 5 makes it possible to determine definitely which form of group is the more accurate. The greater the concentration the greater, of course, will be the deple- tion in the years from which that concentration is drawn. If the quinquennial groups are so formed that the fields of attraction of the years of concentration extend beyond the limits of the several groups, a year which is a multiple of 10 will draw relatively more from the adjacent groups than will a year which is an odd multiple of 5. This will result in making the groups alternately too large and too small. The excess of the number of reported ages in the groups contain- ing years that are multiples of 10 over the number in groups containing years that are odd multiples of 5 may accordingly be used as a measure of the relative inaccuracy of a system of age grouping. Of two series of quinquennial age groupings of the same returns, the one in which the alternate excess and deficiency in the size of the successive groups is less marked will lie the more accurate. Table 5 illustrates the way in which this principle ruay be applied, and also shows the relations between the successive age groups in a "normal stationary population." Table 5.- Per cent thai innuh,,- ;» each specified age group form* of tin- iirillimelic mean of the numbers in the two adjacent groups. [Farr's English Life Table, No. 3, P, column.] AGE GEODP. Number in group. Arithmetic mean of num- bers in pre- ceding and following groups. Percent. 4,004.211'. 3, 584, 556 :: 166,213 3,372,234 3,242,793 3, 095, 021 2,939,916 2, 777, 260 2,604,856 2,119.396 2,216,195 1,981,004 1,701,562 1,371,499 99S, 238 622,519 311,393 115, 4S4 28, 996 4,445 3,735,214 3, 478, 395 3,354,503 3,283 i 8,091,354 2,936,140 2,772,886 2,598,328 'J, 11". 525 2,200,200 1,958,878 1,676,251 1,349,900 997,009 i, 4 ■■) • 369, 002 170, 194 59,965 96.0 99.1 100.5 100.3 100.1 100.1 100.2 L00.S 100.4 50 to i j ears 100.7 101.1 101.5 101.6 100.1 95.1 84.4 67.9 48.4 It will be noted that during the periods of youth and middle age, when the death rate is fairly constant, the number in each age group is only slightly greater than the arithmetic mean of the numbers in the two adjacent n; groups. It can not be supposed that in the numerical relations of the corresponding age groups of an actual population the regularity would be so great. However, uniform or systematic irregularities in the returns for different classes of the population may be safely charged to errors in the returns, rather than to actual peculiari- ties in the age constitution of the population. Table 6 shows the results obtained by applying the above process to the age reports of the aggregate pop- ulation of continental United States. Three methods of grouping have been used; in the first the census plan is followed, a year of concentration being placed at the beginning of each group; in the second the 3 T ear of concentration is in the middle of the group; and in the third it is placed at the end. For conven- ience these methods will he called Method I, Method II, and Method III, respectively. The expressions obtained for the relative numerical values of groups containing multiples of 10 and of groups containing odd multiples of 5 are placed in separate columns, and the average of each of these two classes of per cents is shown for each method of grouping. Only those mul- tiples of 5 from 20 to 65 are considered, because on multiples below 20 there is no perceptible concentra- tion, and at advanced ages the rapid decrease of the population has the effect (shown in Table 5) of produc- ing a marked diminution in the per cents. Table 6. — Per cent that the aggregate population in quinquennial age groups containing specified years of concentration forms of the arith- metic mean of the numbers in the two adjacent groups, by three methods of grouping, for continental United Stales: 1900. YEAI: OF CONCENTRATION. Method I. Method II. Method III. 20 104.1 101.3 101.3 96.1 93.4 97.4 101.1 100.0 110.5 106.6 99.1 104.0 93.0 90.9 93. 9 100.4 99.7 104. 2 109. 7 104.4 103.1 25 103 5 30 96.7 3ft 90.6 40 100.9 45 92.9 60 .. 103.9 56... 94.2 in 102. 65... 93.7 101.5 97.9 103.5 96.4 104.2 95.0 The difference between the average value of the per cents for the groups containing even multiples of 5 and for groups containing odd multiples of 5 is, by Method 1, 3.6; by Method II, 7.1; by Method III, 9.2. These differences represent the relative inaccuracies of the different methods of grouping, and indicate that Method I is preferable to either of the other two methods. Table 7 shows the corresponding differences for various classes of the population, classified by sex. Table 1. — Excess of the average per cut that die numbers in quin- quennial groups containing even multiples ofSforms of the arithmeti mean of the numbers in the two adjacent groups, over the correspond- ing per cent for groups containing odd multiples of 5, for the aggre- gate, native white, foreign white, and negro population, classified by sex, for continental United Slates: 1900. CLASS OF POPULATION. Method I. Method II. Method III. Aggregate population: 3.6 3.6 3.6 7.1 6.8 7.2 9.2 8 2 9.4 Native white population: 2.9 3.1 2.6 3.4 3.1 3.6 9.9 7.5 12.5 4.3 4.4 4.2 8.3 7.7 8.8 23.7 19.9 28.5 5 9 Foreign white population: 11.8 11 Negro population: 17.7 37.7 The sizes of the differences shown in Table 7 indicate the extent to which preference is given in the age reports to multiples of 10 rather than to odd multiples of 5. This preference is most marked with the negro population, and is more marked with the foreign white than with the native white population. It is more strongly marked for females than for males. But the most important fact brought out by Table 7, as by Table 6, is that Method I, in which the year of concen- tration is the lowest year in the group, is the most accurate method of grouping; while Method III, in which the year of special concentration is the highest year in the group, is the least accurate. This indicates that the tendency to understate ages is much stronger than the tendency to overstate them. The tendency to understatement of age is strongest in the negro popu- lation; stronger in the foreign white than in the native white population; and stronger with females than with males. It is undoubtedly true, although the preceding tables do not show it, that persons of advanced age in the United States, as elsewhere, are liable to overstate their ages. Further discussion of this point will be found in the section on centenarians. Children s ages. — An examination of the graphic rep- resentation of the reported ages does not reveal, for most of the ages below 20, an}- special form of error, beyond the tendency to prefer even numbers to odd ones, which is noticeable throughout the age tabic But it will be seen that there is an apparent deficiency in the number of children in their second year of life — reported in the census tables as " 1 year old/' Table 8 shows the number of children at each year of age 17 under 5, and the per cent which that number forms of the total number under 5, for the censuses of 1900, 1890, and 1880. Table 8. — Aggregate population at each year of age under o, • mil per cent that number at each age forms of the Ma! under 6, for con- tinental United States: 1900, 1890, and 1SS0. Under 1 year 1 year 2 years 3 years 4 years 1900 Number. 1,916,892 1,768,078 1,830,332 1,X24,312 1,831,014 Per cent. 20.9 19.3 19.9 19.9 20.0 1, 666, 734 1,077,008 1,729,817 1,631,988 1,629,146 Per cent. 20. ft 14.] 22.7 21.4 21.3 1880 Number. 1,447,983 1,256,956 1,427,086 1,381,274 1,401,217 Per cent. 20.9 18.2 20.6 20.0 20.3 At each of the three censuses considered fewer chil- dren were returned as 1 year old than at any other age under 5 — a deficiency which was most noticeable in 1890 and least noticeable in 1900. It should be remem- bered, however, that in 1890 the age inquiry related to "age at nearest birthday," and that since the ages of chil- dren under 1 were obtained by months, the class tabu- lated as " 1 year old" included, so far as the instructions were strictly followed by the enumerators, only those between 12 and 18 months, inclusive. We should expect that the number of children 1 year old would be noticeably less than the number under 1 year old, on account of the very heavy death rate in the first year of life; but there is no reason to believe that there are fewer children 1 year old than 2 years old. Table 9 shows the per cents which the number reported at each of the first three years of life make of the total under 5 in the censuses of certain countries in which — as in the United States — the age question has reference to the last birthday, and the ages of children under 1 are obtained in completed months. Table 9. — Per cent that population m each sp< cified year of age forms of the total population lea than 5 years nf age, for countries in which ages are obtained by asking "age at last birthday." ' COUNTRY. Date of census. PER CENT OF TOTAL UNDER 5 YEARS. Under 1 year. 1 year. 2 years. 1891 1890 1891 1896 1890 1891 1891 1891 1891 1891 1891 20.0 20.5 21.2 20. 6 19.3 21.0 19.8 21.4 20. 5 21.4 22 1 22. U 17.1 20.4 19.5 18.8 17.8 19.5 17.8 19.1 19.1 17.8 19.8 19.9 20 7 20 7 20 7 20.0 20.2 20.4 19.9 1891 20.1 1 Compiled from the census reports of the countries named. Iii each of the censuses shown in the table, with the exception of that of Denmark, the number in the sec- ond year of life was less than the number in either the first or the third year. Table 10 shows corresponding per cents for the cen- suses of those countries in which the information as to age is obtained by asking the date of birth. Table 10. — Per rent that population well as for those under 1. The marked deficiency in the number of children 1 year old must be attributed to this cause. The asking of the ages of children under I by months has lessened the overstatement in that year, so that the second class has gained less from the first than it has lost to the third. Additional light is thrown on this question by Table 11, which shows the per cents that the population id' different classes under I. 1. and -J years old make of the total population less than 5. Pable 11. — Per cent thai population specifit 'I full years. That is. the per cents shown in the table for 1890 have been obtained by the use of a relatively smaller divisor than was used in obtaining the per cents for 1880 and 1900. Hence the fact that the per cent which the population under 1 forms of the population "•under 5" is somewhat smaller for 1890 than for 1880 or 1900 indicates that the over- statement of the ages of children really in their first year of life was much more frequent in L890 than in either of the other censuses. This conclusion is con- firmed by an examination of the per cents which the number "I year old" 1 forms of the number under 5 for 1890. These per cents arc too large to represent the true proportion of children between li' and 18 months, inclusive, to children less than .">+ months old. Remembering that the number reported as 1 year old is usually too small rather than too large, it will be seen that some peculiar causes of error were at work in 1890. 'Ninth Census, Report on Population, pages xxix and \xx; Report on Vital Statistics, page 516. 2 Fair, Vital Statistics, page 208. It may be assumed that many persons disregarded the exact form of the age inquiry in 1890, and that to some extent the age returns of that census as of other cen- sus! s represent the completed years. This assumption is corroborated by the fact that the per cents which the population 1 year old form of the total under 5 in 1890 show less difference from the corresponding per cents for 1880 and 1900 in the case of the colored than in the case of the native white population. This indicates that the more illiterate population, or the enumerators working among them, gave less attention lo the exact form of the age question. This failure to return all ages at the nearest birthday would partly explain the excessive number returned as " 1 year old" in 1890, but it seems probable that there was another cause of error which also operated to swell that number. The instructions to enumerators in that census were to the effect that ages should be returned at the nearest birthday, but that the ages of children under 1 should be reported by single months. It is natural that some of the enumerators should have interpreted these in- structions as meaning that only T the ages of children less than 6 months old should lie reported by single months. This would decrease the number of chil- dren reported as in their first year of life, and would correspondingly increase the number reported as 1 year old. It is fair to conclude that the number of children under 1 is stated more accurately by the cen- sus of 1900 than by that of 1890. Moreover, there is good reason for believing that in this particular the census of 1900 is more accurate than that of 1880. The addition of the inquiry as to date of birth has increased the general accuracy of the age returns, and has probably lessened the overstatement of the ages of children under 1 as well as of children 1 year old. Evidence of this decreased overstatement is given in Table 11, where the per cent that the population 2 y r ears old forms of the total population under 5 is shown to be smaller in 1900 than in 1880. The re- turns of the population less than 1 year old by months throw some light on the question of overstatement. These returns were not tabulated at the census of 1880, and in 1890 and 1900 they were tabulated only by three- month periods. Table 12 shows the per cent that the population in each of these periods forms of the total under 1. for 1900 and L890. Table 12. — Per cent that papulation at specified months of age, forms of the total population less than 1 year of age, for continental United States: 1900 and i.svo. 11100 1890 iOS. Aggre- gate. Native whin-. Col- nivtl. Aggre- gate. Native white. Col- ored. 25.4 25. 9 24.8 23. '.1 25.1 25.5 25. 1 24 3 27.1 28. 1 23.3 21.2 22.6 28.1 29. 2 20.1 22.4 27. 9 29. 2 20.5 23.4 29.4 29.3 17.9 19 It will be noted that in Table 12 there is a marked deficiency of children in the last quarter of their first year of life in 1890 as compared with 1900. This fact supports the inference already made, that overstatement was more common in 1890 than in 1900. Children in the third quarter of their first year of life, however, seem to be relatively more numerous in 1890 than in 1900. This is undoubtedly due to the concentration of reported ages on "6 months." In the Massachu- setts census of 1895, in which ages were ascertained in the same way as in the Federal census of 1890, but tabu- lated for each month of the first year, more children were reported as 6 months old than in any other month, the number thus reported amounting to 11.1 per cent of the total population under 1. For those states in which the registration of births and deaths is fairly complete, it is possible to test the comparative accuracy of the returns of children under 1 year at different censuses. If the registrations of births and deaths were tabulated so as to show not only the total number of births in each month, but also the month of birth and the month of death of children dying before they have lived one year, it would be possible to estimate with reasonable accuracy from such records the number of children less than a year old living at a given date, such as the date for which the census speaks. But the published registration reports, although giving the number of births in each month, classify deaths and ages at time of death only by years. It is necessaiy, therefore, to assume that five-twelfths of the total number of deaths of children under 1 year in a given calendar year were before and seven-twelfths after June 1. This assumption does not quite accord with tin' facts, infant mortality being higher in summer than in winter. But the error thus introduced will be small, and for purposes of comparison may be disregarded. The proportion of the number of deaths of childi"en under 1 in the twelve months preceding the taking of the census (estimated in the manner just described) to 1,0(10 births in the same period may be called the "infantile death rate.'' Such a rate is, of course, valid only for purposes of comparison. We have as data, then, the number of births during the year preceding the taking of the census, the infantile death rate, and the number of survivors as returned by the census. We can then ascertain by a simple propor- tion what the approximate number of survivors would have been in a given census if the death rate had been what it was in the year preceding a later census, and if the degree of completeness of the census and the regis- tration statistics had been unchanged. Having thus adjusted the returns of the earlier census to correspond with the death rate at the later census, the next step is to obtain the ratio of this adjusted number of survivors to the number of births during the preceding vear. The same proportion of the number of births during the year preceding the taking of the later census will be an approximation to the number of children less than a year old that should have been returned by the later census if the registration records and the census were about as accurate as they were at the earlier census. The difference between this estimated number and the number enumerated will be a very fair measure of the deficiency <>r excess in the later census as compared witl: the earlier one. Table 13 shows such a comparison of the census and registration statistics of Rhode Island, Massachusetts, and Connecticut for L900, I890,and 1880. The census of 1880 is taken as the standard of compari- son. The point to be observed is that if the census and the registration of births were as accurate at one period as another, the "deficiencies** would be approximately zero. • Table 13.-C0MPARIS0N OF THE POPULATION LESS THAN 1 YEAB OF AGE WITH THE REGISTRATION RECORDS OF BIRTHS AND DEATHS FOR MASSACHUSETTS, CONNECTICUT, ANH LIIODE ISLAND: 1900, 1890, AND 1880. Births, year ending May 31 Deaths of children under 1 year of age Infantile death rate Survivors, by census Estimated survivors Apparent deficiency Per cent of deficiency MASSACHUSETTS. 190(1 71,936 10, 936 152 60, 492 '.I, Mj. 4. 054 6.3 1890 57,813 9.322 161 43, 043 51, 120 S.077 15.8 1SS0 42,053 6,411 152 37,586 CONNECTICUT. 1900 20, 881 3,114 149 19,774 19,111 -663 -3.5 1890 17, 367 2, 181 143 It, 469 16.020 1,551 9.7 lssn 13,906 1,927 189 12, S79 RHODE ISLAND. 1900 10,816 1,850 171 9,368 166 4.7 1890 8,399 1,422 169 6,890 658 768 10.0 ISSO 797 125 6,132 5, 824 Table 13 shows that if the accuracy of the registra- tion reports was constant, the census returns of children were much less complete in 1890 than in 1880 in each of the states considered. The results for 1900 show smaller relative deficiencies in Rhode Island and Mas- sachusetts, and a gain in Connecticut. This would in- dicate that the returns of children's ages in Massachu- setts and Rhode Island were more complete in 1900 than in 1890. but less complete than in 1SS0, while in Connecticut they were more complete in 1900 than in either 1880 or 1890. But these results may indicate improvements in the registration of births. In fact, the laws governing the registration of births in Massa- chusetts were changed at such a time as to affect the com- parisons between 1880 and 1890, and the laws in Rhode Island and Connecticut were changed between 1890 and 20 1900.' J n these changes we have the explanation of the excess of the apparent deficiency in Massachusetts for 1890 over the deficiencies for the, other states in the same census. Here too, undoubtedly, lies the explana- tion of the apparent deficiencies in Massachusetts and Rhode Island for 1900 as compared with 1880. It is significant that, notwithstanding improvements in the registration returns, the results for Connecticut show that the census of L900 was relatively more complete in its reports of children under L than the census of I ssi i. In general these results confirm the previous inference that the reports of the ages of young children were much more accurate in 1900 than in 1890 and probably more accurate than in 1880. It is thought by some that the apparent deficiencies in the number of children returned by the census are in large measure real deficiencies. 8 It is a matter of general census experience that the persons giving the information to the enumerators do not always realize, the necessity of including young children in the census, and that a few are accordingly omitted. 3 But such omissions are unimportant as compared with the errors resulting from overstatement of ages. There is no reason for believing that the omissions were relatively more numerous at one census than at another, while it is certain that the manner of obtaining the information as to age had an important influence upon the accuracy of the returns. Criiti -iiariiins. — If the returns were accurate there must have been 3,504 persons in the United States in 1 91 II i who had lived more than one hundred years. There is no doubt, however, that this number is an exaggera- tion, 4 although it is a relatively smaller number than has been returned at any other census of the United States. The. number of centenarians in 100,000 popu- lation of known ages at the last six censuses has been as follows: 1850, 11; 1860, 10; 1870, 9; 1880, 8; 1890, 6; 1900, 5. This regular diminution indicates, not that the longevity of the population has been decreasing, but that in this as in other particulars the accuracy of the age statistics of the census has been increasing. Table 14 shows the number of centenarians in lOO.oon population of all ages, classified by sex, race, and nativity, at the last three censuses. 1 Massachusetts Registration Report, 1880, appendix, page clxxvi; Connecticut Registration Report, 1900, page 5; Rhode Island Regis- tration Report, 1900, page 136. 2 Twelfth Census, Vol. Ill, page xl. 3 This subject has been treated by the present writer in an article on "The Enumeration of Children," Quarterly Publications of the American Statistical Association, March, 1901, Vol. VII, pages 227 to 254. *In European census practice it is quite common to check the accuracy of the returns of centenarians by special investigation of each case For a partial list of such investigations see G. von Mayr, Statistik und Gesellsehaftslehre, Vol. II, page 74. "In Prussia the number of persons [in a total population of 30,000,000] declared to be more than 100 years old in 1890 was 149, of whom more than one-half were discovered upon investigation to be of less age; and of these 8.8 per cent were found to be from 95 to 100; 14.3 between 90 and 95; and the rest not yet 90 years old." — Mayo-Smith, Statistics and Sociology, page 61. Even the uncorrected figures for Prussia amount to a proportion of only one centenarian in every 200,000 of the population — about one-thirteenth of the proportion in the United States in the same year. Table 14. — Population over 100 years of age in 100,000 of all known ages, by 8< ', rare, and natiriti/, for continental United Stales: 1900, 1S90, and 1SS0. SEX, RACE, NATIVITY, OR NATIVITY Of PARENTS. 1900 1890 1880 Aggregate 4.6 3.3 6.0 6.4 4.4 8.5 8.0 5 5 10.6 Native white: 0.8 0.5 1.6 1.0 0.6 1.3 0.3 0.3 0.4 3.9 3.3 4.6 29.0 21.0 38.0 1.4 1.0 1.9 1.7 1.2 2.3 0.6 0.5 0.6 1. 1 3.6 6.3 38.6 J5.6 51.6 1.6 1.3 1.9 Native white— native parents: Native white— foreign parents: Foreign white: 5.6 4 4 6.8 Colored: 45.4 29.9 Females 60.8 The most noticeable fact shown by Table 14 is that the proportion of persons who were reported to be over 100 years old in 1900 is thirty-six times as great for the colored as for the native white population. This differ- ence seems even more remarkable when it is remem- bered that the death rate of the colored population is considerably higher than that of any other class, 5 and that consequently the real proportion of centenarians is probably less for the colored than for any other class. The relatively low proportion of centenarians among the native white population of foreign parents is to be explained by the fact that there were very few per- sons of foreign birth in the United States one hundred years ago. It will be noted that for every class of the population the proportion of reported centenarians is greater for females than for males. This should not be taken as in- dicating a greater tendency to the overstatement of age among women of advanced age than among men. In the greater number of age periods the mortality of males is higher than that of females, and consequently more females attain advanced age. For both sexes and for all classes of the population the three censuses show a progressive improvement in the accuracy of the re- turns of centenarians. Definite proof of the inaccuracy of the returns of centenarians is given in Table 15, in which the correla- tion between illiteracy and the proportion of centena- rians is shown. The various states are ranked in order of the proportion of reported centenarians in their population, as well as in the order of the per cent of their population 10 years of age and over who can not read or write. If the correlation were perfect, the ranks of the states with respect to these two ratios would, of course, be identical — that is, the differences shown in the last column of the table would be zero. So close a correlation can not be expected, for the number of re- 5 For comparative death rates see Twelfth Census, Vol. Ill, page lxix. 21 ported centenarians depends upon (1) the number of actual centenarians in the population; (2) the number of persons less than 100 years old, but of advanced age; (3) the proportion of the total number of persons of advanced age who erroneously return themselves as over 100. Only the third factor can be supposed to have any connection with the illiteracy of the population. Table 15. — Correlation between illiteracy and the number of reported a ntenarians in 100,000 of known ages, for stales and territories: 1900. STATE OR TERRITORY. Number Percent of cente- illiterate narians I in popu 100,000 of known age. Alabama Arizona Arkansas California Colorado Connecticut Delaware District of Columbia . Florida Georgia Idaho Illinois Indian Territory Indiana Iowa Kansas Kentucky Louisiana — Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming lation 10 years and over. (') (') 31.0 29.0 20.4 4.8 4.2 5.9 12.0 8.6 21.9 30.5 4.6 4.2 19.0 4.6 2.3 2.9 16.5 38.5 5.1 11.1 6.9 4.2 4.1 32.0 6.4 6.1 2.3 13.3 6.2 5.9 33.2 5.5 28.7 5.6 4.0 5.5 3.3 6.1 8.4 35.9 5 20 7 14. 5 3.1 5.8 22.9 3.1 11.4 4.7 4.0 In pro- portion In illiter- of cente- acy. narians. 5 3 1 7 14 12 18 35 39 11 43 27 44 17 16 20 7 10 6 6 32 38 35 10 12 13 29 37 11 49 21 48 17 11 4 1 31 33 19 19 37 26 28 39 48 42 2 5 23 22 27 25 46 50 3 16 22 23 42 28 9 4 38 32 10 8 40 30 36 43 20 31 34 45 45 24 47 21 8 2 49 34 15 11 11 15 26 46 25 29 13 9 24 47 30 18 S3 36 50 44 Differ- ence. 3 2 11 11 6 3 1 2 4 13 1 14 5 6 2 10 7 11 11 21 26 6 15 4 4 20 4 4 23 16 3 6 1 Less than 0. 5 in 100,000. The degree of correlation between the proportion of centenarians and the illiteracy of the population is indicated by the fact that for 24 of the 50 states the differences shown in the last column are less than 6. If there were no correlation between the two ratios the "median difference" would be as liable to be over 14 us to be under 15. The general age constitution of the state of Iowa is fairly typical of that of continental United States as a whole, and in Iowa the percent illiterate in the popula- tion 10 years old and over is less than in any other state except Nebraska. There were in Iowa, in 1900, 2,226,- 632 persons of known age, of whom 26 were reported to be more than 100 years old. The corresponding pro- portion of the population of continental United States would give 884 centenarians, or about one-fourth the reported number. "This number, however, can scarcely be regarded as even a maximum estimate; the true number was undoubtedly much less. AGE CONSTITUTION OF THE POPULATION. Median and average ages. — The simplest and prob- ably the most significant single expression of the age constitution of the population is the median age. This is the age with reference to which the population can be divided into halves — that is, half of the population are younger and half are older than the median age. In the computation of the median age, it is assumed that the population in the year of life in which the median falls is evenly distributed through that year. The death rate at the ages in which the median falls is so small that the exaggeration of the median age produced by this assumption is very slight. Table 16 shows the median age for different classes of the population at the census of 1900. Table 16. — Median age of the population classified by sex, general nativity, and race, for persons of known nor in continental United Stales: 1900. CLASS OF POPULATION. Aggregate Native born Fi treigD 1 >orn Total white Native white Native white— native parents . Native white— foreign parents Foreign white Total colored Negro Both sexes. 22. 85 211. Id 38. 12 23. 36 20.22 21.10 1- 05 38. 13 19.70 19.45 20.20 38.71 23.82 20.33 21.27 17.99 38.71 19.97 19.45 Females. 22. 43 20. 02 38.03 22. 91 20. 12 20. 93 18.11 38.04 19.46 19.44 For the aggregate population of both sexes the median age is 22.85; for males it is 23.29, and for females it is 22.43 years. The difference between the median ages of the two sexes is about 10 months. Part of this differ- ence is due to the larger number of foreign born persons among the males than among the females. Among the different classes of the population the median age is highest for the foreign born population and lowest for the native whites of native parents. The foreign born population increases through immigration, mainly of adults. This explains the high median age of the foreign born, and, coupled with a relatively high birth rate, it explains the low median age of their native children. There is a difference of 1,65 years between the median age of the native white population of native parents and that of the negro population, which indi- cates the higher birth rate and death rate of the latter class. For each class of the population except the na- tive white of foreign parents the median age is higher for males than for females. It will be noted, however, 22 that with the exception of the foreign horn population and cif the native white population of foreign parents the median age falls where it would be affected by the tendency of males to overstate theft ages for the pur pose of being counted as of voting - age. If there is any bias in the errors in the age returns of females at corresponding ages, it takes the form of understate- ment. In view of these considerations it seems unsafe to attach any special significance to the difference be- tween the median ages of males and females. Table 1" shows the median age of the population classified as living in urban and in rural districts in L90O. Table 17. — Median "•/, of the population, classified by sea and as living in cities having at least 25, 000 inhabitants, in cities having at leant 100,(11/0 (iilialiiin)it.<, ami in smaller cities and rural districts, for continental United States. 1900. Both sexes. Mules. Females. 22. 85 23.29 ■-'2. 13 25.40 21.85 25.36 22 22 25 s:t 22 ::i 25. 79 22.68 ".".HI 21.33 24.97 Smaller cities and rural districts 21 . 75 The median age of the population living in cities of over 25,0(in inhabitants is about three and one-half years greater than that of the population living in smaller cities and rural districts. This difference may he attributed to two main causes — the higher birth rate of the rural districts, and the migration from the country to the cities, which, like migration from abroad, consists largely of adults. Moreover, the cities contain a larger proportion of foreign born persons than do the rural districts. The median age of the population in cities having between 25,000 and 100.000 inhabitants is some- what higher than it is in larger cities. The difference between the median ages of the population in urban and in rural districts is slightly larger for women than for men. Table 18 shows the median age of the population, classified by sex, for every census the United States has taken. Before. 1880 the ages of the population were not given by single years. In 1790 the ages of only the free white male population were obtained, and in only two groups — under 16 and over 16. In successive cen- suses new details were added to the age classification, and since 1830 the ages of that part of the popula- tion in which the median age falls have been tabulated in quinquennial groups. For the censuses before 1880 it has been assumed for the purpose of computing the median age that the proportions of the population at the single years within the groups were as in 1900. For the colored population the returns of the negro population in L900 have been used as a standard, for no other classes were counted as ••colored"' before L870. Table 18. — Median age of tfa population classified by sex and race, for continental United stairs: 1790 to 1900. HGGREGATE. WHITE. COLORED. CENSUS, Both sexes. Males. Pe males. Both sexes. Males. Fe- males. Both Fe inn lis 1900 22.85 21.42 20 86 20.14 I'.i :w 18.83 17 Til 17. 16 16. 05 23. 29 21.. SO 21. 13 20. 22 19.71 19.12 17.78 17.06 16.57 22. 13 21.00 20. 62 20. H7 19.03 L8.66 17.73 17.27i 16.73 23.36 21.91 21.32 20.38 19.70 19.12 17.85 17.21 16. 53 15.99 15.97 23.82 21.34 21.60 20. 52 20.11 19.49 17.92 17. 15 16.49 15.87 16. 73 15.88 22.91 21. 55 21.01 ii 2 . 19. 32 18.78 17.77 17.27 16.58 10.13 16.27 19.70 17. 83 18.01 18.49 17. 65 17.33 17. 27 16.90 17. 15 19.97 17. 92 17.98 18. 19 17. 78 17.27 17.03 16 55 16 94 19.16 1890 17.7.". 18S0 1- ii 1870 1 s. 78 1860 17.62 1 851 1 17 III 1840 . 17.51 1830 17. 1 1 1820 17.36 1810 .. 1800 . . . 1790 .. The most significant fact shown by the results in Table 18 is the general increase in the median age of the population. This is probably due to a combina- tion of factors, among which are a decreasing birth rate, a decreasing death rate, and an increase in the number of the foreign born. The median age of the aggregate population of both sexes has increased during every decade since 1820. The total amount of this in- crease in the eight, decades is 6.2 years, or an average of about two-thirds of a year per decade. The greatest apparent increase has been between 1890 and 1900, but this is probably due in part to the fact that in 1890 many persons gave their ages as at their last birthday, not- withstanding the fact that the question in that census called for ••age at nearest birthday." This error un- doubtedly makes the median age for that census some- what too low. The median age of the white popula- tion increased during each decade from 1810 to 1900, the increase amounting in the ninety years to 7.4 years, or an average increase of about five-sixths of a year in a decade. For both the white and the colored population the increase in the median age has been greater for males than for females. So far as the white population is concerned this may be partly clue to the large excess of males in the increasing foreign born population. Table 19 shows the median age and quartiles for states and territories at the census of 1900. The quar- tiles might be called ''secondary medians." One-fourth of the population are below and three-fourths are above the first quartile; three-fourths are below and one- fourth are above the second quartile. 23 Table 19.— MEDIAN AGE AND QUARTILES OF THE TOTAL POPULATION: 1900. STATE OR TERRITORY. Continental United States North Atlantic division New England Maine New Hampshire Vermont Massachusetts Rhode Island Connecticut Southern North Atlantic New York New Jersey Pennsylvania South Atlantic division Northern South Atlantic Delaware Maryland District of Columbia Virginia West Virginia Southern South Atlantic North Carolina South Carolina Georgia Florida North Central division Eastern North Central . . Ohio Indiana Illinois Michigan Wisconsin Median age. 27.11 27.92 28.50 27. 96 26.91 26.24 26.85 25.29 26.17 '.'.5. 15 24.20 19.98 21.72 24.63 23.39 27.10 20. 28 20.28 18.70 18.68 18.04 18.80 20. 39 23.35 24.09 24.89 23. 96 24. 57 22. 26 First quartile. 10. 52 13.49 13.66 14.40 13.47 13.47 13. 15 13.15 11.92 12. 64 11.85 11. 20 10.12 11.81 11.10 15. 0U 9.39 9.17 8.61 8.49 8.29 8.53 9.10 10.86 11.29 11.99 11.45 11.02 11.66 10.18 Second quartile. 38. 75 10. 81 42. 85 46.33 16.08 16. 48 11. 57 41.18 12. 18 40.30 40.85 40. 23 39.49 38.05 40.86 39.62 41.12 36. 82 35. 32 33.81 34.81 32.27 33.58 34.78 40.10 41.10 40. 48 38.79 41.29 39.20 STATE OR TERRITORY. Continental United States — Continued North Central division — Continued. Western North Central Minnesota Iowa Missouri North Dakota South Dakota Nebraska Kansas South Central division Eastern South Central Kentucky Tennessee Alabama Mississippi Western South Central Louisiana Arkansas Indian Territory Oklahoma Texas Western division Rocky Mountain Montana Idaho Wyoming I !olorado - - New Mexico Basin and Plateau Arizona Utah Nevada Pacific Washington Oregon California Median age. First quartile, 22.23 22. 02 22. 96 22. 36 20. 77 20. 74 21.60 22.22 19.21 19.65 20. 73 19.91 18.77 18. 46 18.84 19. 36 18.66 18.11 19.94 18.68 25. Ml 24. 93 26.60 22.45 24.89 25.92 21.12 21. 33 24.21 19.07 28. SO 27.12 25.80 25. 33 28.04 10. 29 9.78 10.71 10. 53 8.70 9.29 9.98 10. 62 8.76 8.97 9.56 9.26 8.58 8.43 8.54 8.69 8.55 8.16 8.79 8.49 12. 02 10.93 12.11 9.54 11.53 11.89 8.86 10.60 8.49 14.56 11.81 12. 19 14.14 Second quartile. 38.24 37. 52 39.30 38.11 35.26 37.40 37.53 39.21 34.18 34.83 36.52 35.34 33.57 32. 51 33.46 33.97 33.21 31.96 35.56 33.24 40.24 38.49 3S.14 37. 69 36. 91 39.19 37.27 37. 12 38. 32 35.19 44.91 41.59 39.55 40.70 42. 81 The median age is highest in the New England and the Pacific states, and lowest in the Southern South Atlantic and the Western South Central divisions. The maximum, 28.50 years, is in New Hampshire; the minimum, 18.04 years, is in South Carolina. The region in which the median age is higher than that of continental United States as a whole includes all of the New England and North Atlantic states, the South Atlantic states north of the Potomac river, all the states of the Eastern North Central division except Wisconsin and Iowa, and all the states of the Western division except New Mexico, Utah, and Idaho. Map 1 shows the geographic distribution of median ages, classified in three arbitrarily selected groups. The distribution of the quar tiles necessarily has a very close correlation with that of the median. The position of the lower quartile, however, is influenced especially by the proportion of children in the popula- tion, while the position of the upper quartile is more 4091— Bull. No. 13—04 1 responsive to variations in the proportions of persons of advanced years in the population. It is significant, therefore, that the maxima for the lower quartile are in California and Nevada, while the maxima for the upper quartile are in Maine, Vermont, and New Hamp- shire. For both quartiles the minima are in South Carolina and Indian Territory. Table 20 shows the average age of certain classes of the population in 1900. 1890, ami 1880. The average age is obtained by ascertaining the sum of the ages of the population and dividing this sum by the number of the population. For this purpose it is assumed thai the population is distributed evenly through each year of life; thus, the population reported as 47 years old is assumed to average 47i years. As the age tables do not show the ages of centenarians by single years, it has been assumed that their average age is that which the Massachusetts census of 1895 reports for that state — 102£ years. 24 MAP 1.— MEDIAN AGE OF THE TOTAL POPULATION, FOR STATES AND TERRITORIES: L900. Less than 20.52 years. 20.52 to 25 years. More than H 25 years. 25 Table 20.— AVERAGE AGE OF THE POPULATION, CLASSIFIED BY RACE AND NATIVITY, FOR CONTINENTAL UNITED STATES: 1900, 1890, AND 1880. All ages Unknown Known Sum of known ages. Average age AGGREGATE. 1900 75, 994, 573 200, 584 75, 793, 991 1,989,260, 140 26.2 Isoo 62,622,250 162, 168 62,460,085 1,569,020,1'J4 25.1 I SMI 50,155,783 1,234,564,930 24.6 NATIVE WHITE. 1900 56, 595, 379 120, 172 56, 475. 207 1,364,925,515 24. 2 1890 1880 45,862,023 36,843,291 96, 524 45,765,499 36,843,291 1,056,226,318832,145,935 23.1 22.6 FOREIGN WHITE. 1900 1890 1880 10, 213, 817 24,880 in. 1ss,m:i7 410,196,526 40.3 9, 121, 867 24,617 9,097,250 344,032,595 37.8 6, 559, 679 6,559,679 251, 272, 979 3S. 1900 9, 185, 379 55, 532 9,129,847 214,138,444 23.5 1890 1880 7, 638, 360 41,024 7, 597, 336 168,761,281 22,2 6, 752, 813 6,752,813 148, 146, 065 21.9 The average age, like the median, has risen during the period from 1880 to 1900, the total increase for the aggregate population amounting to 1.6 years, while the increase in the median age in the same period was about 2 years. The increase in the average age of both the, native white and the colored population, also, was 1.6 years, while the increase in the average age of the foreign white population was but one-tenth of a year less. During the period 1880 to 1890, in which an unprecedent- edly large number of immigrants came to the United States, the average age of the foreign born population declined. During the decade 1890 to 1900 the immigra- tion was less, so that at the end of the decade there was a relatively large number of foreign born persons who had been in the United States more than 1<> years. This caused an increase of a year and a half in the average age of the foreign white population during that decade. The average age, like the median, is highest for the for- eign white and lowest for the colored. In computing both the average and the median age it has been assumed that in 1890 the age question was answered as asked, that is, that the age reports represent the ages of the population at the nearest birthdays. If it be thought that for most of the returns the age was really age at last birthday, corresponding changes should be made in the average and median ages. On this supposition the median age for that census would be raised by 6 months. The average age would be, for the aggre- gate population, 25.6 j 7 ears; for the native white. 23.6 years; for the foreign white, 38.3 years; for the colored, 22.7 years. Productive and nonproductive ages. — An age classifi- cation which is frequently used is that into productive and nonproductive ages. It has been said that from the economic point of view this is the most important of the statistical groupings of the population. 1 The idea is that the population can be grouped by ages in such a way r as to differentiate those who have to bear the brunt of the economic struggle and those who are economically dependent. The division usually adopted is into three groups, persons under 15 and per- sons over 60 years of age being classed as econom- ically unproductive. Of course the limits of these groups are arbitrarily- chosen and it is quite certain that they are not very accurate. The census of 1900 showed that the per cent of males who were engaged in gainful occupations was as follows: 10 to 14 years of age. 21.4; 15to64 years of age, 90.8; 65 years and over. 68.4. For females the per cents were as follows: 10 to 14, 8.1; 15 to 64, 21.4; 65 and over, 9.1. The upper limit of the group of "productive ages" ought to be higher than either 60 or 65 years. Care should be taken, there- fore, not to place too much emphasis on the economic significance of the classification into "productive" and "nonproductive" ages. Yet this classification is not without utility. It is desirable for many purposes to have a classification of the ages of the population less detailed than the usual grouping in 5-year or 10-year periods, and for such purposes this one serves as well as any other. The year 15 marks off with a rough accuracy 7 the years of childhood from those of adult life. The year 60 marks quite as accurately the completion of the period of "middle age" and the beginning of "advanced years." Tables 21 and 22 show the population in these three age groups classi- fied by sex, race, and nativity for the censuses of 1900, 1890, and 1880. 1 Wagner, Grundlegung der politischen Oekonomie, 3d ed., Vol. I, part 2, page 606. Cf. Engel, Der Werth des Menschen. Table 21. 26 POPULATION IN SPECIFIED M,K GROUPS, CLASSIFIED BY SEX, RACE NENTAL UNITED STATES: 1900, 1890, AND 1880. AND NATIVITY, Foil CONTI- 1900 1890 1880 BEX, RACK, "It NATIVITY. Under 15 years. 1 i to 59 years. CO years and over. Under 15 years. 15 to ft9 yi nr . 60 years and over. Under 15 years. 1ft to ft',' years. 60 years and over. Aggregate: Both sexes 26,124,985 13,196,049 12 928 936 44,797,145 23,020,391 21,776,754 4,871,861 2, 472, 585 2,399,276 22,242,200 11,290,0118 10,952, 192 36, 342, 563 18,681,914 17,660,649 3, 875, 322 1,992,429 1,882,893 19, 109, 3C2 '.i,i-,'.i(i,:i-'i 9, 119,043 28,218,743 14, 376, 077 13,842,666 2,827,678 1, 152, 122 1,375,256 Total white: i 1 39 '. L07 11, 124,409 22, i,S90 11, 13 i, 181 10,871,209 511.126 267, 926 253 200 3, 607. 469 1,802 942 1,804,527 39,719,251 20, 169,665 19 149 86 31,536,818 16,007,985 15 528 833 ■-. i- !, 133 4, 161,680 3,720,753 5,077,894 2,550,726 2, 527, 168 4,427,377 2,241, 187 2, 186,240 2,931,999 1,461,594 1,470, 105 1,495,378 779, ft 11! Tl.i.sSft 444, 4S4 231,44.8 213,036 19,044,788 9,672,145 9,372,643 18,313,429 9,300,876 9,012,553 731,359 371,269 360, 090 3,197,412 1,617, sua 1,579,549 32,292,739 16,642,776 15,649,964 26,086,799 12,693,734 12,393,065 7,20ft.:' 10 3,949,041 156 399 1,049,824 2,039,139 2,010,685 3,525,222 1,810,766 1,714,456 2,365,271 1,195,848 1,169,423 1,159,951 614,918 645, 033 350, 100 181.663 168, 437 16,123,101 8, INS, 222 7,934,879 15, 699, 594 7,974,423 7,725,171 423, 507 213, 799 209, 708 2, 986, 261 1,502,099 1,484,162 24,759,003 12,648,785 12,110,218 19,334,892 9,721,948 9,012,944 5,424,111 2, 926, 837 2,497,274 3,469,740 1,727,292 1,732,448 2,520,866 1 293 893 1,808,805 Native while: 912,894 995 'HI Foreign white: TlJ.Hhl 380,999 331,062 Colored: 306 812 158,529 148,283 Table 22.— PER CENT THAT POPULATION IN SPECIFIED AGE GROUPS FORMS OF TOTAL POPULATION OF KNOWN AGES, CLASSIFIED BY SEX, RACE, AND NATIVITY, FOR CONTINENTAL UNITED STATES: 1900, 1890, AND 1880. 1900 1890 1880 SEX, RACE, OR NATIVITY. Under 15 years. 15 to 69 years. 00 years and over. Under 15 years. 15 to 59 years. 60 years and over. Under 15 years. 1ft to 59 years. 60 years and over Aggregate: 34.5 34.1 34.8 59.1 59.5 58.7 6.4 6.4 6.5 35.6 35.3 35.9 58.2 68.5 57.9 6.2 6.2 6.2 38.1 38.0 38.2 56.3 66.3 56.2 ft. 6 5 7 5 6 Total white: 33.8 33.4 34.2 39.0 38.9 39.0 5.0 1.7 5.4 39.5 39.3 39.7 59.6 60.0 59.1 55. 8 66.0 55.7 80.3 81.1 79.3 55.6 55.6 55.6 6.6 6.6 6.7 5.2 5.1 6.3 14.7 14.2 15.3 4.9 5.1 4.7 34.7 34.4 35.1 40.0 40.1 39.9 8.0 7.5 8.6 42.1 42.2 42.0 58.9 59.2 58.5 54.8 54.7 54.9 79.2 80.0 78.3 53.3 53.1 53.5 6.4 6.4 6.4 5.2 5.2 5.2 12. n 12. ft 13.1 4.6 4.7 4.5 37.2 37.0 37.3 42.6 42.9 42.4 6.5 6.1 6.9 44.2 44.3 44.1 67.0 57.2 56.9 52. ft 52. 2 52. 7 82.7 83.1 82.2 51.2 51.0 51.5 ft, s ft, 8 ft. 8 Native white: 4.9 4.9 4.9 Foreign white: 10. s 10.8 10.9 Colored: 4.6 4.7 4.4 27 The age group containing the years 15 to 59, which may be called "productive ages,'' contained in 1900 a larger proportion than in either 1880 or 1890. Foreverv class of the population this group was relatively larger in 1900 than in 1890, and for the native white and colored populations it was relatively larger in 1900 than in 1880. The decline in the proportion of the foreign white population at productive ages in the decade 1880 to 1891 I was undoubtedly clue to the large foreign immigration of that decade. The decline of immigration in the decade 1890 to 1900 brought with it a slight increase in the proportion of foreigners of productive age and a marked increase in the same class over 60 years of age. For the aggregate population a slightly larger propor- tion of men than of women were of productive age at each of the three censuses. For the native white and colored population, however, the proportion of the. female population who were of productive age was larger than the corresponding proportions for males in 1890 and 1880, but the differences between these propor- tions are too small to be of sio-nificance. The approximate equality between the per cents of the native white and the colored population who are of productive age is noticeable. It should be remembered, however, that the native white include the native chil- dren of foreign parents, so that this close approxima- tion is more interesting than significant. The very marked excess of the percentage in productive ages among the foreign born over the corresponding per cents for the other classes of the population is self- explanatory. The general increase in the proportion of the popu- lation at productive ages has been accompanied by an increase in the proportion 60 years of age and over. For the native white and negro populations these facts seem to indicate a decreasing birth rate and possibly an increase in the average length of life. For even- class of the population the proportion under 15 was less in 1900 than in either 1890 or 1880. Table 23 shows a similar grouping of the population by age for the most important European countries. Table 2.3.— POPULATION OF EUROPEAN COUNTRIES IX SPECIFIED AGE OKOUPS AND PER CENT THAT POPU- LATION IX EACH GROUP FORMS OF THE TOTAL POPULATION OF KNOWN AGES. 1 Austria . Belgium 1890 Denmark 1890 England and Wales 1891 France Date of census. Germany 1890 Holland. 1890 1890 1891 ISM I 1890 ls'jll 1891 Spain 1890 Sweden 1890 Switzerland I 1890 Hungary. Ireland . . Italy Norway . . Portugal . Scotland . Ali known 23,895,413 6,069,321 2,170,752 29,002,525 38,112,731 49,428,470 4,511,169 17,339,226 4, 702, 964 28,455,948 1,986,955 5, 030, 564 4, 025, 647 17, 252, 472 4,784,981 2,933,334 Under 16 years. 8, 160, 450 1,989,448 755, 702 10,172,235 10,000,152 17, 372, 100 1,647,231 6,401,888 1.529,067 9,158 457 712,310 1,669,139 1,432,225 5,741,024 1,593,316 941,269 15 to 69 years. 13,850,693 3,491,864 1,193,288 16,684,809 23,343,426 28,111,728 2,448 266 9, 746, 809 2,681,784 16,750,238 1,051.201 2, S57, 460 2,276,776 10,219,423 2,643,034 1,716,082 CO years and over. 1,884,370 ;,SS.(KIS 221,762 J. 1 15, 4S1 1,769, 153 3,944,642 415,672 1 , 190, 529 492, 113 2, 547, 253 223, 444 503, 965 316,646 1,292,025 r,is.,;:;i 275,983 Under 15 years. 34.1 32.8 34.8 35.1 26.2 35.1 36.5 36.9 32.5 32.2 35.9 33.2 35.6 33.3 33.3 32.1 15 to 59 years. 60 years and 58.0 57.5 55.0 -.7. 5 61.3 56.9 :.l . ;; 56.2 57.0 58.9 62.9 56.8 56.5 59.2 55.2 68.5 7.9 9.7 10.2 7.4 12.5 8.0 9.2 6.9 10.5 8.9 11.2 10.0 7.9 7.5 11.5 9.4 1 Computed from the tables given by Bertillon, in Statistique Internationale resultant des Recensement< de la population executes dans les divers Pays de PEurope pendant le XIX« siecle et les epoques preeedentes. [Paris, 1S99.] 28 When these results are compared with those for the aggregate population in 1900, it appears that the United States has an unusually large proportion of its population in the group of productive ages. Only France and Spain— countries in which the population is more nearly stationary than in any other countries of Europe— have a larger per cent of their popula- tion in productive ages than has the United States. It will he noted that the censuses of the European countries included in Table 23, except that of Italy, were taken in. or near. 1890. Comparing with these results the returns for the United States in L890, it appears that not only in France and Spain, but also in Switzerland and Italy, the population of productive age was a larger proportion of the total population of known ages than in the United States. The main cause of the relatively high proportion of productive ages in the United States is the importance of the foreign born element in our population. For most European coun- tries emigration exceeds immigration. The native white population and the colored population (96 per cent of whom are native) of the United States in 1900 included relatively fewer persons who were of pro- ductive age than did any of the countries shown in Table 23, except Holland, Denmark, Norway, and Swe- den. Making the comparison with the United States census of 1890, only Holland and Norway had a smaller proportion at productive ages than our native white, and only Norway a smaller proportion than our colored population. It is significant that all of the European countries shown in the table had in 1890 a larger per cent of population more than 60 years old than had the United States in either 1900 or 1890. Table 2-1 shows the population at productive and nonproductive ages in cities and rural districts, for 1900. The age returns for cities have not been tabulated in such a way as to make it possible to use the same upper limit for the period of productive ages as in the other tables. Accordingly, those over 15 and less than 65 years of age have been considered as of productive age. The table shows that the per cent of the population in the productive age group is larger in the cities than in the rural districts, this difference being more marked for females than for males. It should be noted, however, that the proportion who are of productive ages does not increase with the size of the city, on the contrary the per cent for cities of between 25,000 and 100,000 population is larger than for cities of over 100,000 population. Persons over 65 years of age, also, are relatively more numerous in the smaller cities. Chil- dren under 15, however, are found in relatively greater numbers in the large cities than in cities of less than 100,000 population. The difference between the age constitutions of tin 1 urban and rural districts, so far as the relative numbers in productive and nonproductive ages is concerned, is to be explained largely by the migration from the rural districts to the cities, although differences in birth and death rates and in the number of foreign born residents, also, must be taken into account. Table -4. — Number <»»,000 and 100,000 inhabitants, and in smaller cities and rural districts, for continental United States: 1900. SEX OR AGE GROUP. IN CITIES HAV- ING AT LEAST 100,000 INHAB- ITANTS. IN CITIES HAV- ING BETWEEN 25,000 AND 100,000 INHAB- ITANTS. IN SMALLER CITIES AND RURAL D I S- TRICTS. Number. Per cent. Number. Per cent. Number. Per cent. Both sexes: Total known ages... 14, 170, 385 100.0 5,486,787 100.0 56,136,819 100.0 Under 16 veins .. 15 to 61 years 65 years and over Males: Total known ages . . . 1,233, 163 9,484,824 452, 598 7,042,879 29.8 67.0 3.2 100.0 1,583,551 3,706,976 196, 260 2,726,413 28.8 67.6 3.6 100.0 20,307,971 33, 397, 208 2,431,640 28,919,733 36.2 59.5 4.3 100.0 Under 15 years . . 15 to 64 years 65 years and over Females: Total known ages . . 2,116,837 4,722,751 203, 291 7,127,506 30.1 67.0 2.9 100.0 789, 196 1,848,488 88,729 2,760,374 28.9 67.8 3.3 100 10,290,016 17,366,319 1,263,398 27,217,086 35.5 60.1 4.4 100.0 Under 15 years . . 15 to 64 years 65 years and over 2,116,626 4,761,673 219,307 29.7 66. s 3.5 794, 355 1,858,488 107,531 28.8 67.3 3.9 10,017,955 16,030,SS9 1, 168, 242 36.8 58.9 4.3 Table 25 shows the population in productive and non- productive age groups, for states and territories and for geographic divisions, at the census of 1900. 29 Table 25.— AGGREGATE POPULATION IN SPECIFIED AGE GROUPS, AND PER CENT THAT POPULATION IN SPECIFIED AGE GROUPS FORMS OF THE TOTAL POPULATION OF KNOWN AGES: 1900. All known ages. Under 15 years. 15 to 59 years. 60 years and over. PER CENT. STATE OR TERRITORY. • Under 15 years. 15 to 69 years. 60 years and over. 75, 793, 991 26,124,985 44,817,325 4,851,681 34.5 59.1 6. 1 21, 004, 724 6,262,717 13, 147, 796 1,594,211 29.8 62.6 7.6 5, 676, 777 1,533,222 3,540,609 502, 946 27.5 63.5 9.0 692, 824 410,460 342, 778 2,795,818 427, 642 907,255 15,427,947 189, 333 106, 592 94,877 767, 628 120,471 254, 321 4,729,495 421,630 256, 600 207, 303 1,804,845 275, 401 574, 930 9, 607, 187 81,861 47, 368 40,698 223,345 31,770 78,004 1,091,265 27.3 26.0 27.7 27.5 28.2 28.0 30.6 60.9 62.5 60.5 64.5 64.4 63.4 62.3 11.8 11.6 11.8 8.0 7.4 8.6 7.1 7,257,889 1.879,890 6,290,168 10,415,167 2,111,000 577,518 2,040,977 4,076,112 4,600,426 1,173,529 3,833,233 6, 768, 419 546,464 128,843 415, 958 581, 636 29.1 30.7 32.6 39.1 63.4 62.4 60.9 55. 3 7.5 6.9 6.6 5.6 4,451,908 1,599,545 2, 572, 406 279, 957 35.9 57.8 6.3 184, 226 1,183,950 278,423 1,850,296 955,013 6, 963, 259 57, 981 393, 546 69, 616 710,620 in;:, ssii 2,475,567 112,720 709, 921 Iss.sti; 1,024,334 536,558 11,186,013 13,525 mi. is:: 19, 935 115, 442 50, 672 301,679 31.5 33.2 25.0 38.4 38.5 41.5 61.2 60.0 67. 8 56. 4 56. 2 53. 4 7.3 6.8 7.2 6.2 5.3 5.1 1,888,944 1,338,512 2, 209. 974 525, 829 26,279,235 ts'j mi 572, 277 '.HH.SIIL' 204, 027 8,857,120 1,003,337 Tim, n- 7 1,183,873 298,716 15, 640, 593 103, 206 66, 148 109, 239 23,086 1,781,522 41.4 42.8 41.5 38.8 33.7 53.1 52. 3 , r iH. 6 56.8 59.5 5.5 4.9 4.9 4.4 6.8 15, 955, 736 5,196.557 9,611,150 1, 148, 029 32.6 60.2 7.2 4,160,574 2,511,164 4,810 ! • 2.417, 166 2, i.'.N. 10,323, 199 1,282,471 813,069 1.588,685 772,834 739, 998 3,660,563 2,543.854 1,614,282 2,921,410 1,457,077 1.174,527 ■ 6, 029, 443 324, 249 183,813 300, 161 188, 045 151,761 633, 493 30.9 32.4 33.0 31.9 35.8 35.5 61.3 60.3 60.7 60.3 56.8 58.4 7.8 7.3 6.3 7.8 7.4 6.1 1.747,292 2. 226, 632 3,098,259 Ills, in:, 11*1. Mi;'. 1,064,638 1,467,440 14,030,794 637,801 1,082,065 126,492 164,566 388,288 612, 7811 5.1194,219 1,007 583 1,306,949 1,832.974 181,137 226, 509 619,468 856,823 7, 699, 370 101,908 161,070 183,220 11.776 20, 759 56,882 97,878 637,205 36.5 34.1 31.9 39.4 38.6 36.5 34.9 40.6 67.7 58.7 69.2 56.9 56.2 58. 2 58. t 64.7 5.8 7.2 6.9 3.7 South Dakota 5.2 5.3 6.7 4.7 7,523,255 2, 995, 706 4,134,316 393,233 H',1.8 55.0 5.2 2, 140, 400 2, 012, 844 1,821,980 1,648,031 6,507,539 809,142 783, III 752, 679 650,444 2,698,513 1,204,630 1,123,947 982,363 823,376 3,565,054 126, 628 105, 456 86,938 74,211 243,972 37.8 38.9 41.3 42.0 41.5 56 5 55.8 53.9 53.2 54.4 5.7 5.3 4.8 4.8 4.1 1,378,419 1,306,390 389,352 396, 794 3,036,684 4.064.071 660,646 543,713 167,642 168,538 1,267,974 1,235,817 752,496 735,887 210,582 221,774 1,644,315 2,571,147 65, 277 26. 790 11.128 16, 182 124, 295 257, loT 40.7 41.6 43.0 40.0 41.8 30.4 64.6 64.3 54.1 55. '.I 54.1 63.3 4.7 4.1 2.9 4.1 4.1 6.3 1,225,300 397,383 773.405 54, 512 32.4 63.2 4.4 242,084 161,182 92.304 535, 150 194,580 439,444 71.130 58,917 28, 312 163, 147 75, 877 164,669 162, 019 94,639 61,320 346, 786 108,641 249, 756 8,935 7, 626 2,672 25,217 10, 062 25,119 29. 4 36.6 30.7 30.5 39.0 :'.:. 5 66.9 58.7 66.4 66.0 55.8 66.8 3.7 4.7 2.9 4.5 5.2 6.7 121,642 275,917 41,885 2, 399, 327 511,844 412,604 1,474,879 40, 455 113,350 10,764 673, 865 75, 279 147,219 27, 258 1,547,986 5,908 15,348 3,863 117,476 33.3 ■11.1 25.7 28.1 61.9 53. 3 65.1 64.5 4.8 Utah 5.6 9 'J 7, 1 157.899 126, 136 389, 831 328,857 259,815 969, 314 25,088 26, 654 125,734 30.8 :n 6 26.4 64.3 63.0 65.1 1 9 6. 1 8.6 30 Considering first the results for the main geographic divisions, it appears that the maximum per cent of productive ages is found in the Western division, with the North Atlantic. North Central, South Atlantic, and South Central divisions following, in the order named. The per cents for the minor geographic divisions do not show any marked deviation from this general order. The maximum per cent of 64.5 is found in the Pacific division and the minimum of 53. -i in the Southern South Atlantic division. The study of the results for single states will be facilitated by Table 26, which shows the states in order of the proportion of their population who are of pro- ductive age. Table -'<;. STATES AND TERRITORIES IX ORDER OF THE PER CENT OF THEIR TOTAL POPULATION OF KNOWN AGES WHO ARE FROM 15 TO 59 YEARS OF AGE: 1900. RANK. State or territory. Per .■.■lit. RANK. State or territory. Per ''''"I 67.84 (16.93 66.43 65. 08 65.04 64.99 64.55 64.40 64.25 13. 38 63. 37 62. '.17 62. 49 6'2. 43 61.88 61.31 61.19 60.94 60.86 60.73 60.48 60.30 60.27 59.90 59. 16 '20 58.72 ■' 27 58. 65 28 58. 39 •29 58.19 30 57. 07 31 56. 89 32 56.84 33 56.81 34 56. 52 III 56. 26 36 56.18 37 55. 89 38 55.83 39 55.79 40 55.30 i ihio 41 54.59 4-2 54.33 43..., 54. l.i 44 54.08 53. 92 46 53.57 47... Utah 53.36 48 53. 19 •21 49 53.12 25 50 52.30 i The states in which the proportion of the population of productive ages is greater than that for continental United States as a whole (59.10) include all of the states in the North Atlantic division, together with Delaware, Maryland, and the District of Columbia in the South Atlantic division: Ohio. Michigan. Indiana, Illinois, and Missouri in the North Central division: and all of the Western division except Idaho. Utah, and New Mex- ico. There is a very close agreement between this area and the area in which the median age is higher than it is for continental United States as a whole. The distribution of tlie population in 10-year age groups. — For most purposes served by the study of the age constitution of the population a classification in 10-year periods is sufficiently detailed. Tables 27 to 32 show such a grouping of the population, classified by sex, race, and nativity, for the censuses of 1900, 1890, and 1880. The number at advanced ages is so small that all those over 80 years of age have been included in one group. 1 J It will be noted that each of these groups begins with a year which is a multiple of 10. For some purposes it will be found con- venient to use a system of grouping in which the lowest age in each group is an odd multiple of 5. Such a distribution of the popula- tion for 1900 can be found in the Abstract of the Twelfth Census, pages 11 to 18. 31 Table 27.— POPULATION IN SPECIFIED AGE GROUPS, CLASSIFIED BY SEX, RACE, AND NATIVITY, FOR CONTINENTAL UNITED STATES: 1900. 8EX, RACE, NATIVITY, OR NATIVITY OF PARENTS. Under 10 years. 10 to 19 years. 20 to 29 years. 30 to 39 years. 40 to 49 years. 50 to 59 years. 60 to 69 years. 70 to 79 years. 80 years and over. Aggregate: 18,041,751 9, 113, 008 8,931,7-13 15,636,323 7,833,492 7,802,831 13,864, 157 6,948, 123 6,916,334 10, 520. 820 5, 518, ISC, 5, 002, 634 7,701,778 4,093,752 3,608,026 5,154,001 2, 709, 879 2,444.122 3, 094, 289 1,584,836 1,509,453 1,403,698 711,188 692,510 373, 874 176, 561 197, 313 Total white: I i,558,278 7, 373,804 7,684, 174 15,358,717 10, 639, 101 L3.502, 1-7 6,777,393' 6 725,034 12,629,110 6,280 T, 6 8,895,343 4,487,432 4, 407, 'Jl ; 3. 733, 773 1,860,948 B73, ;i! 429, 013 144,298 1, ' ] I ITT, 7 .>7 2,074,012 1,022,392 1,051,620 12,098,024 6,088,363 6,009,661 10,081,813 5,042,656 5,038,6 >7 7.IH 1,251 3,538,73] 3, 17., ...'ii 3, 067, 062 1 503, 925 1,563, 137 2,016,711 1,045,707 971,004 1,766.433 8 i9,7l 906, 673 1,706.651 819,518 887, 133 9, 465, 019 1,979,794 1. 185,225 7, 113,770 3,646,28s 3, 467, 482 4,958,931 2,664,600 2,394,331 2,154,839 1,081,688 1,073,151 2,351,249 1,017,743 1,055,801 • 517,409 999, 294 495, 501 503, 793 6, 957, 821 3,707,148 5,151,489 2,681,382 2.47(1.107 3,892, 158 2,034,079 1,858,079 1,259,331 647,303 612,028 1,025,766 7 13, 957 :-;m;.uii 347,585 346,015 1,655,198 2, 436, 270 2,218,928 1,557,853 2,756,421 1,417,349 1.339,072 452, 382 233, 601 218,783 1,446,395 785,320 661,075 498,803 273, 609 225, 194 470, 163 252,611 217, 652 2, 815, 953 1,382,025 1.860,182 934,704 925, 47S 1 686,010 845,434 840, 576 174, 172 B9.270 955, 771 199,224 156,547 278,336 150,908 127,428 ■,.i 141,979 122,379 633,875 844,576 120,941 128,635 784,850 390,682 394, 168 59,726 30,259 29, 167 441,241 231,01)1 210,240 117, ssl ,9,246 , ,635 112,802 16,710 325, 607 Males 155. 267 170,340 Native white: 105,949 121.292 Native white — native parens 210. 266 97, 718 5, 242, 109 4, 71H.616 .. IS 2,343.098 199,561 100,294 99,267 2,486,473 1,239,204 1,247,269 2,418, 11:1 1,204,897 1,213,516 112,548 Native white — foreign parents: 16, 975 8,231 8,744 Foreign white: .- .... 49,318 49,048 Colored: 18,267 21,294 26, 973 Negro: 15,890 20,297 25,693 Table 28.— PER CENT THAT POPULATION IN SPECIFIED AGE GROUPS FORMS OF THE TOTAL POPULATION OF KNOWN AGES, CLASSIFIED BY SEX, RACE, AND NATIVITY, FOR CONTINENTAL UNITED STATES: 1900. SEX, RACE, NATIVITY, OR NATIVITY OF PAR] Under 10 10 to 19 years. 20 to 29 years. 30 to 39 years. 40 to 49 50 to 59 years. on to 69 years. 70 to 79 years. SO years and over. Aggregate: 23. 8 2S.5 21.1 20. 6 20. 2 21.0 is.:; 18.0 18.6 13.9 14.3 13.5 10.2 10.6 9.7 6.8 7.0 6.6 4.1 4.1 4.1 1.8 1.8 1.9 0.5 0.5 0.5 Total white: 23.3 23. 1 27. 2 27.2 26. 1 26. o 6 30.3 30.0 1.9 1.8 2.1 27. 2 27. 27.4 27.5 27.6 27 1 20.3 20. 7 22. 4 22. 2 22. 5 21.8 21.6 23.9 24.0 8.6 7 - 9 23.4 23. 23.7 2::. 6 23. 4 23 - 18.2 17. s 17.8 17.(1 IS. 1 17.2 17.0 17,:: 19.6 19.2 20.0 19.8 19.0 20.7 is 8 19.9 19.4 18.8 20.0 14.2 14. i; 1.: • 12.6 12.7 12. ■"> 12. 1 12.3 11.9 1.; - 23. 1 24.2 21.7 11.0 11.7 11.4 11.4 11. 1 11.4 10.4 10.9 10.0 9.1 9.4 S.9 9.5 9.8 9. 2 8.0 s.3 7.8 17.7 18.7 16.6 8.1 8.4 7.9 7.9 8.0 7. s 7.0 7.1 5.7 5.8 5.6 6.8 i'.. 8 6.7 2.9 3.0 2.8 14.2 14.3 14.1 5.5 6.0 :.. 5. 1 5.8 4.9 4.2 4.2 4.2 3.3 3.2 3.3 4.1 4.1 4.2 1.1 1.1 1.1 9.4 9.1 9.7 3.1 3.3 2. s 3.0 3.2 2.8 1.9 1.9 1.9 1.5 1.5 1.5 1.9 1.9 2.0 0.4 0.4 0.4 4.3 4.2 4.5 1.3 1.3 1.3 1.3 1.3 1.3 0.5 0.5 0.5 Native white: 0.4 0.4 0.4 Native white— native parents: 0.5 0.6 0.6 Native white— foreign parents: 0.1 0.1 0.1 Foreign white: 1.0 0.9 1.1 Colored: 0.5 :. Ii 1 Negro: 11.7. 0. 7. 0.6 32 Table 29.— POPULATION IX SPECIFIED AGE CROUPS, CLASSIFIED BY SEX, RACE, AND NATIVITY, FOR ( '( iXTIXENTAL UNITED STATES: 1890. SEX, RACE, OR NATIVITY. Under ID years. 10 to 19 years. 20 to 29 years. 30 to 39 years. 40 to 49 years. 60 to 69 years. 60 to 69 years. 70 to 79 years. 80 years and over. Aggregate: 1 1,208,693 7,715,221 7,493,470 13,591,072 6,823, 198 6, 707. 57 1 11,424,453 5,803,204 5,621,249 8,444,791 4,476,708 3, 968, 083 5,917,158 3, 072, 706 2,844,452 5,995,595 2,080,585 1,918,013 2,468,144 1,284,337 1,183,807 1.(191, S13 562,735 632, 078 312,365 145, 357 167, 008 Native white: 12, 717, 836 6,457,977 6,259,859 334, 980 170,110 164,870 2, 155, S7."> 1,087,134 1,068,741 10, 749, 645 5,404,155 5,345, 190 917,674 458,817 158,857 1,923,753 960,526 963, 227 8, 105, 541 4,069,248 4, 036, 293 1,989,613 1,078,769 910, 844 1,329,299 655, 187 671.112 5,771,850 2,989,846 2,782,004 1,812,912 1,042,570 770, 342 860, 029 444, 292 415, 737 8.667.(199 1.8.58, 404 1. si 15.055 1,647,769 911.8,572 739, 197 602,290 305, 670 296, 620 2, 388, 257 1,214,920 1,173,337 1,234,351 661,472 572, 879 375, 990 204, 193 171,797 1,459,413 745, 095 714,318 789, 221 420,846 368,375 219,510 118,396 101, 114 702,555 356,662 345, 993 300,221 169, 531 140, 690 92,037 46,642 45, 395 203,303 94, 191 109,112 Foreign white: 70,509 34,541 35, 968 Colored: 38,563 16,625 21,928 Table 30.— PER CENT THAT POPULATION IN SPECIFIED AGE GROUPS FORMS OF THE TOTAL POPULATION OF KNOWN AGES, CLASSIFIED BY SEX, RACE, AND NATIVITY, FOR CONTINENTAL UNITED STATES: 1890. SEX, RACE, OR NATIVITY. Under 10 years. 10 to 19 years. 20 to 29 years. 30 to 39 years. 40 to 49 years. 50 to 59 years. 60 to 69 years. 70 to 79 years. 80 years and over. Aggregate: 24.3 24. 1 24. 6 21.8 21.3 22.2 18.3 18.2 18.4 13.5 14.0 13.0 9.5 9.6 9.3 6.4 6.5 6.3 3.9 4.0 3.9 1.8 1.8 1.7 0.5 0.6 0.6 Native white: 27.8 27.9 27.7 3.7 3.5 4.0 28. 4 28.3 28.4 23.5 23.3 23.7 10.1 9.3 11.0 25.3 25.0 26.6 17.7 17.6 17.9 21.9 21.9 21.9 17.5 17.1 17.9 12. 6 12. 9 12.3 19.9 21.1 18.5 11.3 11.6 11.1 vii 8.0 8.0 18.1 18.4 17.7 7.9 8.0 7.9 5.2 5.2 6.2 13.5 13.4 13.8 5.0 5.3 4.6 3.2 3.2 3.2 8.7 8.5 8.8 2.9 3.1 2.7 1.5 1.5 1.5 3.3 3.2 3.4 1.2 1.2 1.2 0.5 0.4 0.5 Foreign white: 0.8 0.7 0.9 Colored : 0.5 0.4 0.6 Table 31.— POPULATION IN SPECIFIED AGE GROUPS, CLASSIFIED BY SEX. CONTINENTAL UNITED STATES: 1880. RACE, AND NATIVITY, FOR SEX, RACE, OR NATIVITY. Under 10 years. 10 to 19 years. 20 to 29 years. 30 to 39 years. 40 to 49 years. 60 to 59 years. 60 to 69 years. 70 to 79 years. SO years and over. Aggregate: 13,394,176 6,782,840 6,611,336 10,726,601 .. ;:s:;.,-,('.9 5, 343, 032 9, 168, 393 4, 664, 425 4,503,968 6, 369, 362 3,271,467 3,097,895 4, 568, 256 2, 322, 468 2, 236, 788 3,111,317 1,641,629 !, 109,688 1,830,095 964, 356 SO,, 75.9 776, 507 388,602 387, 905 221,076 99,464 121,612 Native white: 11, 057, 502 5, 612, 691 5, 444, 911 185,068 93, 059 92,009 2,151.606 1,077,190 1,074,416 s, 614, 930 4, 327, 580 4,287,350 617,2.51 305. 060 312,191 1,494,420 750, 929 743, 491 6,744,802 3, 418, 239 3, 326. 563 1,199,371 639, 132 56(1, 2X9 1,224,220 607,054 617, 166 4,075,850 2,048,672 2,027,278 1,551,896 852, 726 699, 170 741,616 5711,169 371,447 2,708,796 1,347,996 1,360,800 1,343,831 725,794 618, 037 505, 629 248,678 256, 951 1, 832, 606 941,393 891,213 950, 201 524,865 425, 336 328, 510 175, 371 153.139 1,112,895 589, 378 553, 615 491,824 268, 869 222, 955 195, 378 106,109 89, 269 523, 710 259, 660 264, 050 176, 305 90, 696 84, 609 77, 492 38, 246 39, 246 142, 202 63.S56 78, 346 Foreign white: 44, 932 21,434 23,498 Colored: 33,942 14,174 19,768 Table 32.— PER CENT THAT POPULATION IN SPECIFIED AGE GROUPS FORMS OF THE TOTAL POPULATION OF KNOWN AGES, CLASSIFIED BY SEX, RACE, AND NATIVITY, FOR CONTINENTAL UNITED STATES: 1880. SEX, RACE, OR NATIVITY'. Under 10 years. 10 to 19 years. 20 to 29 years. 30 to 39 years. 40 to 49 years. 50 to 59 years. 60 to 69 years. 70 to 79 years. 80 years and over. Aggregate: 26.7 26.6 26.8 21.4 21.1 21.7 18.3 18.3 18.3 12.7 12.8 12.6 9.1 9.1 9.1 6.2 6.4 6.9 3.7 3.8 3.5 1.6 1.5 1.6 0.4 0.4 0.5 Native white: 30.0 30.2 29.9 2.8 2.6 3.0 31.9 31.8 31.9 23.4 23.2 23.5 9.4 8.7 10.3 •22.1 22.2 22. 1 18.3 18.4 18.2 18.3 18.2 18.4 18.1 17.9 18.3 11.1 11.0 11.1 23.6 24. 2 23.0 11.0 10.9 11.0 7.3 7.2 7.5 20.5 20.6 20.4 7.5 7.4 7.6 5.0 5.1 4.9 14.5 14.9 14.0 4.9 5.2 4.6 3.1 3.2 3.0 7.5 7.6 7.3 2.9 3.1 2.7 1.4 1.4 1.5 2.7 2.6 2.8 1.1 1.1 1.2 0.4 0.3 0.4 Foreign white: 0.7 0.6 0.8 Colored: 0.5 0.4 0.6 33 Considering first the results for 1900, it will be seen that for each class of the population, except the foreign white, the number of persons less than 10 years old is greater than in any other decennial period. For the foreign white population the maximum number is in the group 30 to 39 years. The diminution in the numbers of the aggregate population in successive age groups has a certain regularity, as is shown in Table 33. Table 33. — Excess of population in each specified age group over pop- ulation in next higher group, for continental United States: 1900. AGEGRfiVl'. to 9 years... 1 ii i o 19 years. . 20 to 29 years. 30 to 39 years. 40 to 49 years. . 60 to 59 years. 00 to 69 years. 70 to 79 years. Excess over Number in group. number in next high- er group. Per cent. 13,044,751 2,408,428 13.4 15,636,323 1,771,866 11.:; 3,343,637 24. 2 2,819,042 26.8 7,701,778 2,547,777 33.1 5,154,001 2,059,712 39.8 3,094,289 1,690,591 54.8 1,403,698 The per cent which the difference between the numbers in the first and second groups form of the number in the first group is somewhat larger than the corresponding per cent for the second group, but for all the other groups the rate of diminution increases as the age increases. Some interesting results are yielded by a comparison of the distributions of the total population of various classes to the several age groups. The figures will be found in Table 28. A larger per cent of the total native white population of foreign parents than of any other class of the population is in the groups to 9 years and 10 to 19 years. This is explained, of course, by the fact that the adult foreign white population has been increasing faster than the adult native population. A larger per cent of the foreign white population than of any other class is found in each of the groups above 19 years. Larger proportions of the negro population than of the native white population of na- tive parents are less than 10 years old and between 10 and 20 years old. In the groups composed of the years of age from 20 to 79 there are relatively more native whites of native parents than negroes. The same proportion of each of these two classes of the population claims to have reached fourscore. Nearly one-fourth of the aggregate population are less than 10 years old and over three-sevenths are less than 20. Less than one-seventh have accomplished half the possible hundred years of human life, and only 23 out of every 1,000 have passed the allotted "three- score years and ten." It is scarcely necessary to point out that these results have no bearing on the question of the average length of human life. In an increasing population, like that of the United States, the lower age groups are normally better filled than in a station- ary or decreasing population. Over one-fourth of the negro population, as well as of the native white population of native parents, are less than 10 years old. Nearly one-half of the latter and more than one-half of the former are less than 20 years old. Among the native white population of native parents 133 in 1,000 have passed their fiftieth year, and 24 in 1,000 have passed their seventieth year. About one-tenth of the negro population are more than 50 years old, and only 18 in 1,000 are more than 70 years old. Three-tenths of the native white population of foreign parents are less than 10 years old, but only 1 in 53 of the foreign white population is in the same age period. Considerably over one-half of the native white population of foreign parents and one-tenth of the foreign white population are less than 20 years old. Among the native white population of foreign parents only 1 in 22 has passed the fiftieth year, and only 5 in 1,000 have passed the seventieth year. About three-tenths of the foreign white popu- lation are at least 50 years old, and 53 in 1,000 are at leasi 70 years old. For the aggregate population there are relatively more females than males in each of the groups below 30 years, and in the group 70 to 79 years. Males are relatively more numerous in the three groups compris- ing the years of life from 30 to 59. In the groups 60 l" 69 years and 80 years and over, the relative num- bers of the two sexes are approximately equal. In the native white population of native parents, as well as in the foreign white population, females are relatively more numerous in each of the groups below 30 and above 59 years. In the period of middle life there are for each of these two classes relatively more males than females. In the native white population of foreign parents the proportions of males less than 10 years old and between !•» and 60 years old are larger than the proportions of females. There are relatively more females than males in the groups of 10 to 19 and 20 to 29 years, while the proportions of the two sexes in the groups 30 to 39, 60 to 69, 70 to 79 years, and 80 years and over are approximately equal. The excess of the relative numbers of one sex or the other is distributed for the negro population in the same way as for the native white population of foreign parents, except that among the negro population there is a larger per cent of males than of females in the group 60 to 69 years, and a larger per cent of females than of males in the group 80 years and over. A comparison of the returns for 1900 with those for 1880 shows that for the aggregate population of both sexes there has been an increase in the proportion in each age group above 29 years, and a decrease in the proportion in each of the two groups below 20 years, while the proportion in the group 20 to 29 years has remained about constant. The same results appear when the per cents for the two sexes are considered separately, except that for males there has been a decrease in the proportion in the group 20 to 29 years, which is exactly balanced by a corresponding gain for females. This gain in the proportion of the population who are in 34 die more advanced age groups lias been fairly constant during the two decades, for the same general results arc gleaned from a comparison of the per cents for L890 and L900, although, of course, the differences are not so marked. It is to be noted that while those tabulated as •"under 10" in 1890 included only those who reported their ages as being less than 9.1 years, this group contained a proportion of the population larger than the proportion of those who reported themselves as less than 10 in 1000. Comparison of tin 1 returns of the foreign white pop- ulation in 1900, 1890, and 1880 shows that throughout the two decades there has been an increase in the per cent of that class of the population who are in the groups above 49 years. In the other age groups there has been a relative decrease during the two decades. except that the proportion of the population between 20 and 29 years of age was greater in 1900 than in 1880. and the proportion between 30 and 39 years of age was markedly greater in L 900 than in 1890. Between 1880 and 1900 the age constitution of the native white popu- lation so changed as to leave a larger or equal propor- tion in each of the groups above 29 years and a smaller proportion in the other groups, this decrease being most marked for children under 10 } T ears of age. The per cents for the colored population show a marked decrease between L880 and 1900 in the proportion in the first ten years of life, which is balanced by a slight increase for each of the other groups. Tables 34 and 35 show the population of cities and of the rest of the country, classified by 10-year age periods. The age returns of cities were not tabulated in such a way as to make it possible to use the form of grouping that has been used in the preceding tables. Table 34.— POPULATION IN SPECIFIED AGE GROUPS, CLASSIFIED BY SEX AND AS LIVING IN CITIES HAVING AT LEAST 100,000 INHABITANTS, IN CITIES HAVING BETWEEN 25,000 AND 100,000 INHABITANTS, AND IN SMALLER CITIES AND RURAL DISTRICTS, F( U; CONTINENTAL UNITED STATES: 1900. Under 5 years. 5 to 14 years. 15 to 24 years. 25 to 34 years. 35 to 44 years. [5 to ' years. 55 to 64 years. 65 years and over. In cities having at least 100,000 inhabitants: 1,505, 171 757, 108 748,063 549,319 276, 428 272,891 7,115,838 3,599,776 3, ,16,062 2. 7117, 992 1,359,429 1 , 368, 563 1,034,232 512,768 521, I'll 13,192,133 6,690,240 6, 501, 893 2,730,811 1.279,389 1,451.422 1,095,020 515, K% 579, 124 n 065,274 6, 679, 746 5,485,528 2, 766, 216 1, 390, 240 1,375,970 1, 050, 331 580,862 619, 169 8, 268, 933 4,803,756 3,965, 177 2, 058, 737 1,082,452 976, 285 781,601 108,881 372,720 6,371,609 3,381,448 2. KID. 161 1,230,610 629, 101 601,515 493,677 263. 349 240. 228 4,673,248 2,520,008 2,153,240 697, 944 341, 563 :;;,o,:;M 286, 447 139,500 146, 947 3,018,144 1,581,361 1.436,783 1 >2,698 203, 291 249, 307 190,200 88, 729 107. 531 In cities having between 25,000 and 100,000 inhabitants: In smaller cities and rural districts: " 131 040 1,263,398 1,168 242 Table 35. — Percent that population in s/» ci/inl age i/rou/ix forms of the total population of known ages, classified bg sex ami as tiring in cities having at least 100,000 inhabitants, in cities having between 25,000 and 100,000 inhabitants, and in smaller cities and rural districts, for continental United States: 1900. Under 5 years. 5 to 14 years. 15 to24 years. 25 to 34 years. 35 to 44 years. 45 to 54 years. 55 to 64 years. or, years and over. In cities having at least 100,000 inhabit- ants: Both sexes 10.6 10.8 10.5 10.0 10.1 9.9 12.7 12.4 12.9 19.2 19.3 19.2 18.8 18.8 18.9 23. 5 23.1 23.9 19.3 18.1 20.4 20.0 18.9 21.0 19.7 19.3 20.1 19.6 19.7 19.3 19.1 19.5 18.8 14.7 14.9 14.6 14.6 15.4 13.7 14.3 15.0 13.5 11.4 11.7 11.0 8.7 8.9 8.4 9.0 9.3 s.7 8.3 8.7 7.9 4.9 4.9 5.0 5, 2 5.1 5.3 5.4 5.5 5.3 3.2 2 9 Females In cities having be- tween 25,000 and 100.000 inhabitants: Both sexes 3.5 3.6 3.3 Females In smaller cities and rural districts: 3.9 4.3 4.4 Females 4.3 The groups under 15 years, and each of the two groups in which persons more than 55 years old are distrib- uted, contain a smaller proportion of the urban popu- lation than of the population living in " smaller cities and rural districts." Moreover, the group 15 to 24 years contains a larger proportion of the latter class than it does of the population living in cities having at least 100,000 inhabitants. All of the other groups are relatively larger in the cities of over 25,000 inhabitants than in the rest of the country, this excess being most marked for the group 25 to 34 years. It appears that the greater average maturity of the urban population, shown by its higher median age, is due to the relative deficiency of children and the relatively larger popula- tion in the period of early middle life. A slightly larger proportion of the population is in the groups 15 to 24, 45 to 54, 55 to 64, and 65 years and over in cities of 25,000 to 100,000 inhabitants than in larger cities. The difference between the per cents of the popula- tion in the group 5 to 14 years in the cities and in the rural districts is somewhat more marked for females than for males, but in the group 15 to 24 years the pro- portion of the male population is smaller and of the female population larger in cities than in rural dis- tricts. This may indicate that the migration of females from the country to the city takes place, on the aver age, at an earlier age than that of males. Tables 36 and 37 show the distribution of the popu- lation of the states and territories and the geographic divisions in 10-year age periods. 35 Table 36.— TOTAL POPULATION IN 10-YEAR AGE PERIODS: 1900. STATE OR TERRITORY. Continental United States. North Atlantic division . New England Maine New Hampshire. Vermont Massachusetts . . - Rhode Island. - - . Connecticut Southern North Atlantic New York .... New Jersey... Pennsylvania South Atlantic divi^ien Northern South Atlantic Delaware Maryland DiBtrict of Columbia . Virginia West Virginia Southern South Atlantic North Carolina . South Caroline . Georgia Florida North Central division . Eastern North < lentral Ohio Indiana ... ] Llinois Michigan Wisconsin . Western North Cei Minnesota Iowa Missouri North Dakota . South Dakota . Nebraska Kansas South Central division . Eastern South Central Kentucky . Tennessee Alabama .. Mississippi. Westein South Central. Louisiana Arkansas Indian Territory . < iklahurna. All known ages. 75,793.991 18,014,751 Under 10 years. 21.004.721 ! 692, 824 110, 400 313.778 3. 795. sis 127,642 907,265 I. - ., 1J7. 'J 47 7,257,889 1,879,890 6,290,168 1. 151,908 L84, 226 1,183,950 278.42: 1,850,296 5,963,259 1,888,944 l 338 il2 2,209,974 525, B29 15.955.730 4. ISO. 574 2.511,104 I. 310,256 2,417, 156 ,i. ■ . 10,323,499 1,747,292 3,098,259 318, 105 400, 833 1 1,030,794 7,. 'V23, 25.') 2, 1 10, 400 2,012,844 1,821,980 1,548,031 6,507,539 1,7.7s. II!' 1,306,390 389,352 396, 794 Texas 3,036 584 Western division 4,064,071 129.026 73, 095 64, 698 538,298 177,966 3 287, 119 Rocky Mountain 1,225,300 Montana Idaho Wyoming . . . Colorado ... New Mexico. Basin and Plateau Arizona . Utah ... Nevada.. Pacific. Washington Oregon California ... 242,084 161,182 92, 304 535, 150 194, 5*0 439, 444 121,642 275.917 1 1 885 511 . 844 112,604 1. 174,879 1,467,212 103, 171 1,416,736 2.827,331 1,100,282 39, 235 267,329 16 B81 189,683 27.7. 154 1 . 727. 039 547, 076 397,914 638, 997 143.052 6,051,138 3,542,279 867, 624 548,247 1,093,805 7.24.717 507,886 2,508,859 417,, 737 520, 064 734,800 89,985 107, 266 266, 545 344, 162 3, 955, 832 2,073,530 557, 489 539,668 523, 994 452. 379 1 882,302 391.261 377, 232 117.-36 111,276 884 797 282,441 51,733 11.937. 20. 469 114,276 54,028 116,140 28, 784 79, 980 7,376 457, S45 109, 666 84,737 262. 942 10 to 19 years. 15, 630, 323 3,796,851 9 17, 964 122, 782 67, 617 60, 645 467, 197 76, 031 153, 692 2,848,887 1,284,470 341,093 1,223,324 2,384,986 969, 410 36,619 246, 181 47, 548 424, 339 214, 723 1,415,576 20 to 29 years. 13,864,457 4, 001, 660 1,070,164 117,054 73, 992 57, 169 561,792 84, 531.1 175,627 1,397,801 354,636 1,179,059 1,871,660 809. S9| 449, 222 331,783 519, 343 115, 228 5, 433. 246 3,221.77.1 822,1.1.6 522, 232 97.7., 1 1 18 480, 77,8 441,050 362. 211 167,336 071,644 64,735 ST. 660 2 ; .j,".. 325,856 ::.28::. 538 1.77,7,186 480, 420 470, 353 433,507 372,906 1,526,352 315,119 313, 7194 93,049 s; 566 717,221 737, 702 217/228 36.77,7 31,647 15,616 39, 323 91,960 22,083 63,010 6,867 428, 514 33, 157 216, 537 61 . 267, 322, 359 176,576 1,061,766 30 to 39 years. 1(1. 7.20. .82(1 3,300,692 878,033 96,223 60,331 47,874 461,146 145, 126 1.178,017 302, 982 941,060 1,197,487 564,210 26, 121 163,790 46,649 210,290 117,060 633.277 321, 46J 337, 361 402,018 102,920 4,695,358 2,8)8, Us 441.111 891,463 420, 598 342, 086 1,816,91(1 309, 281 397,291 7.66,612 7.8. ,,,,3 66, 744 190, 407 257, 970 2,549,972 193,238 134,970 239, 493 65, 576 3,745, 185 2, 330, 882 603 156 354,995 719,271 313,471 1,369,435 3s-,, 187 366, 303 285,639 1,180,537 252, 248 305,084 428, 655 46, 375 52,294 113,441 186, 506 833, 195 27.6,367 236. 477 70. 542 67,81,3 549, 349 745. 807 267, 391 231,618 189,709 163,477 735, 924 159,758 140,239 It. 179 52, 505 338,943 689, 037 333 193 217.330 50,9 l 27,761 20, 898 100. 935 76, 451 23, 480 45, 078 7, 893 92,337 81.201 264,973 92. 196 72,849 27(1.81.8 7.11.789 25. 362 17.335 96,974 26, 876 60,511 19, 747 34,235 ,, 529 93, 18.: 65,479 152,2 40 to 49 years. 7,701,778 2,359,836 64.8,63: 81,631 49, 598 39, 707 323, 649 49,875 104, 112 1,711,204 830, 494 210, 002 670, 708 913,348 430, 827 20, 865 124,760 32,7.60 1711, lit 82,528 183, 521 157, 793 102, 191 175,821 46, 716 2,741,497 1,715,084 461,840 269, 618 504, 506 269, 977 209, 1 13 1,026. 113 172, 688 221, 1 HI 305,690 31,459 11 068 105, 486 1 47.. .882 1,200,377 637,430 194, 199 168,904 17.1,' .HI 119.383 562, 947 116. (KIT 113, 5 33, 451 36, 725 263.(199 486, 720 143,372 29 222 iei 68i 111,287, 67.414 19,770 42,905 13, 763 33.8117 5, 335 63, 890 50, 400 186.153 50 to 59 years. 5,154,001 1,596,940 461,023 64,247 37, 856 32,027 220, 391 33, 774 72,728 553, 431 139, 163 443, 323 638, 729 297.338 14, 404 84,870 23,7,8., 118.069 56, 400 341,401 116,948 70, 145 125, 063 29, 245 1,831,989 1,150.260 318 s;.,, 191,118 316, 002 1.89,890 131.371 681,729 1113, 189 151,647 207, 638 15, 470 25,042 69 - .7 108,8,8,; 793, 571 60 to 69 years. 3, 094, 289 994, 212 304, 442 48, 151 27,280 22,918 138.391 20, 167 47,332 68'.,, 77(1 341,298 83, 181 265, 291 371,291 177,: 8,679 51,734 13,236 72,410 31,750 193. 452 64,1 43,315 70,807, 15,266 716.328 3110,37,1 117,052 191, 991 118,018 88,813 III-, 37(1 444,236 134,086 129, 542 100,582 80. 026 349, 335 74, 630 72. 123 18.967 24.438 158,877 292, 772 78, 218 13,691 10, 170 5,029 37,449 11,876 26,358 7,877 14, 459 4,022 63,388 99,316 121,045 7,695 13, 730 37,962 63. 231 430, 503 249,601 77, 267 69, 150 55, 156 48. 028 180, 902 41.589 35, 943 8,014 12.022 83, 334 173,685 38,272 6, 572 5,421 2,005 17,411 6,863 16,036 1,019 9,346 2,671 34,881 31,281 122,034 17.602 17,681 84,094 70 to 79 years. so years and over. 470, 760 151,811 25, 189 15,132 13,190 65,355 9,017 23, 028 318, 949 161,906 36, 486 120, 557 162, 767 80, 292 3,823 22, 971 7,, 287, 33, 326 14,887 82,475 29,954 17, 392 29, 173 5, 956 524, 145 342, 450 97,088 63, 570 87,133 , ,,202 19, 163 181,689 31,020 13,333 50,337 3,346 5,702 15,393 26, 662 177, 993 108, 178 35, 343 29, 657 23,936 19, 242 69, 815 17,618 13, 630 2,506 3, 691 32, 434 68,033 12,393 1 , 929 1 , 827, 7,13 6,720 2,376 7,169 1,427 4,779 963 48,471 0,208 7,410 34, 793 3ti Tu.i e :57. -PER CENT THAT POPULATION IN EACH SPECIFIED AGE GROUP FORMS OF THE TOTAL POPULATION OF KNOWN AGES: 1900. STATE OR TERRITORY. Under 10 years. 10 to 19 years. 20 to 29 years. 30 to 39 years. 40 to 49 years. 50 to 59 years. 60 to 69 years. 70 to 79 year-. 80 years and over. 23.8 20.6 18.3 13.9 10.2 6.8 4.1 1.8 0.5 20.7 18.1 19.1 15.7 11.2 7.6 4.7 2.3 0.6 19.1 17.0 19.2 16.8 11.6 8.3 5.5 2.7 0.8 18.6 18.0 is. a 19.2 19.6 19.6 21.3 17.7 16.5 17.7 1(',.7 17.8 16.9 18.5 16.9 18. 16.7 20.1 19. s 19.4 19.0 13.9 14.7 14.0 16.5 15.9 16.0 15. 7 11.8 12.1 11.6 11.6 11.6 11.5 11.1 9.3 9.2 9.3 7.9 7.9 8.0 7.3 6.9 6.6 6.7 5.0 4.7 5.2 4.5 3.7 3.7 3.8 2.3 2.1 2.6 2.1 1.2 1.2 1.3 0.7 0.6 0.8 0.5 20.2 21.5 22.5 27.1 17.7 18.1 19.5 22.9 19.3 18.9 18.7 18.0 16.2 16.1 15.0 11.5 11.5 11.2 10.7 8.8 7.6 7.4 7.0 6.1 4.7 4.4 4.2 3.6 2.2 1.9 1.9 1.6 0.6 0.5 0.5 0.4 24.7 21. S 18.2 12.6 9.7 6.7 4.0 1.8 0.5 21.3 22. 6 16.8 26.5 26. 9 29. 19.9 20:8 17.1 22.9 22.5 23.7 is. II 18.3 22.0 17.4 18.5 17.8 14.3 13.8 16.8 11.4 12.3 10.6 11.3 10.5 11.7 9.2 8.6 8.1 7.8 7.2 8.5 6.4 5.9 5.7 4.7 4.4 4.7 3.9 3.3 3.3 2.1 1.9 1.9 1.8 1.6 1.4 0.6 H.5 II . 5 0.5 0.4 0.4 29.0 29.7 28.9 27. 2 23.0 23.8 24.8 23.5 21.9 20.7 17.0 17.6 is. 2 19. 6 17.9 10.2 10.1 10.8 12.5 14.2 8.3 7.6 8.0 8.9 10.4 6.2 5.3 5.7 5.6 7.0 3.4 3.2 3.2 2.9 4.3 1.6 1.3 1.3 1.1 2.0 0.5 0.4 0.4 0.3 0.5 22.2 20.2 17.9 14.6 10.7 7.2 4.5 2.1 0.6 Ohio 20.9 21.8 22.7 21.7 24.6 24.3 19.8 20.8 19.9 19.9 21.3 21.1 18.2 17.6 18.5 17.4 16. 6 17.9 14.5 14.1 15.6 14.2 13. 5 13.7 11.1 10.8 10.5 11.2 10.1 9.9 7.7 7.6 6.6 7.8 6.6 6.6 4.8 4.7 4.0 4.9 4.3 4.0 2.4 2.1 1.8 2.3 2.4 1.8 0.6 ii 5 0.4 0.6 0.7 0.4 25.5 23.3 23.7 28.3 26.8 25.0 23.5 28.2 20. 7 21.0 21.7 20.3 21.9 21.8 22.2 23.4 17.7 17.8 18.3 18.4 16.7 17.9 17.6 18.2 14.5 13.7 13.8 14.6 13.0 13.5 12.7 11.3 9.9 10.1 9.9 9.9 10.2 9.9 9.9 8.5 5.9 6.8 6.7 4.9 6.3 6.6 7.4 5.6 3.6 4.5 3.9 2.4 3.4 3.6 4.0 3.1 1.8 2.2 1.6 1.0 1.4 1.4 1.8 1.3 0.4 0.6 0.4 0.2 0.3 0.3 0.4 0.4 27. 6 23.4 18.2 11.3 8.5 5.9 3.3 1.4 0.4 26.0 26.8 28. 8 29.2 28.9 22.4 23.4 23.8 24.1 23. 5 18.0 18.2 18.3 18.5 18.1 12.5 11.5 10.4 10.6 11.3 9.1 8.4 8.5 7.7 8.6 6.3 6.4 5.5 5.2 5.4 3.6 3.4 3.0 3.1 2.8 1.7 1.5 1.3 1.2 1.1 0.4 0.4 0.4 0.4 0.3 28.4 28.9 30.2 28.0 29. 1 21.1 22.9 24.0 23.9 .... | 23.' 6 1S.1 18.6 18.1 18.1 17.1 18.1 18.3 11.6 10.7 11.4 13.2 11.2 17.0 8.4 8.7 8.6 9.3 8.7 12.0 5.4 5.5 4.9 6.2 5.2 7.2 3.0 2.8 2.1 3.0 2.7 4.3 1.3 1.0 0.6 0.9 1.1 1.7 0.4 0.3 0.2 0.2 0.3 0.3 23.1 17.7 19.0 17.8 11.7 6.4 3.1 1.0 0.2 21.4 26.0 22.2 21.4 27. s 26.4 15.2 19.6 16.9 17.5 20.2 20.9 21.0 17.2 22.6 18.9 16.8 17.4 21.0 15.7 18.8 18.1 13.8 13.8 12.1 10.4 11.1 12.6 10.2 9.8 5.6 6.3 5.5 7.0 6.1 6.0 2.7 3.4 2. 2 3.2 3.5 3.7 0.8 1.1 0.6 1.1 1.2 1.6 0.2 0.3 0.1 0.2 0.4 0.4 23.7 29.0 17.6 19.1 18.1 22.8 16.4 17.9 19.3 16.3 18.8 18.2 16.2 12.4 15.6 17.1 11.3 8.6 12.7 12.5 6.5 5.3 9.6 7.8 3.3 3.4 6.4 6.0 1.2 1.7 2.3 2.0 0.4 Utah 0.5 0.6 Pacific 0.4 21.4 20.5 17.8 18.1 19.7 17.3 18.1 17.6 IS. 4 18.3 15.9 17.1 12.5 12.2 12.6 6.8 7.6 8.3 3.4 4.3 5.7 1.2 1.8 2.3 0.2 Oregon 0.4 0.5 37 Among the main geographic divisions the South Central division has relatively the largest population under 10 years of age, and between 10 and 20 years of age. The North Atlantic division has relatively the largest population between 20 and 30 years of age, and in each of the groups 50 years and over. The Western division has relatively the largest population between 30 and 40 and between 40 and 50 years of age. The rela- tive minima for the various age groups are as follows: Under 10, and 10 to 19, tire North Atlantic division; 20 to 29, the North Central division; and each of the higher age groups, the Western division. The age constitution of the North Central division is most nearly representative of that of continental United States as a whole. The age constitutions of the North Atlantic and the Western divisions are alike, with respect to the small proportion of the population in the first 20 years of life, but for the other age groups there are marked differ- ences between the two divisions. These differences are clearly apparent in Table 38. It shows that over two- fifths of the population of both the North Atlantic and the Western divisions is more than 30 years old — a larger proportion than in any of the other divisions. But the population 30 years of age and over is distributed very differently for the two divisions. In the North Atlantic division a smaller per cent of this part of the population is less than 60 years of age than in any other division. In the Western division a larger per cent of the population is between 30 and 60 years of age than in any other division; that is, the North Atlantic division contains an unusually large number of persons of advanced age, while the Western division contains an unusually large number of persons in middle life. Table 38. — Per cent that population 30 years of age and over form* of the total population if known ages, and per cent that the population betwi i n SOand 60 gears of age and 60 years of age and over forms of the population 30 years of age and over, for main geographic divisions: 1900. Per cent that population 30 years of age and over forms of total population of known ages Per cent that population between 30 and 60 years of age forms of the popula- tion over 30 years Per cent that population 60 years of age and over forms of the population over 30 years North Atlantic division. 82.0 South North Atlantic Central division, division. 82.6 82.4 South Central division. Western division. 30.2 12.4 84.5 85. S 15.5 It now remains to consider the distribution of the population by 10-year groups for the states and terri- tories. Anything like an adequate discussion of these results would require so much space that it seems best to substitute for such an analysis Table 39, which shows the states and territories in the order of the proportions of their population in the various age groups. 38 Table 39.— RANK OF THE STATES AND TERRITORIES IN THE PER CENT OF THEIR POPULATION OF KNOWN AGES IN SPECIFIED AGE GROUPS: 1900. UNDER 10 YEARS. 10 TO 19 YEARS. 20 TO 29 YEARS. 30 TO 39 YEARS. 40 TO 49 YEARS. RANK. terril Per cent. State or territory. Per (Tilt. State Or territory. Per State or territory. Per cent. or territory. Per cent. RANK. ] Territory ... South Carolina 30. 239 29. 728 29. 138 28. 987 28.962 28.914 28.876 28 18 i 28 261 2s. II 11 27. 767 27. 21 if, 26. 927 26.811 26. 701 26 165 26.046 !6.017 ilO 25. 036 24.580 23.717 23.663 23. 17 1 23. 357 22.739 22 580 22. 523 22. 176 21.832 21.705 21.416 -I | 6 21.370 21.354 21.297 20.901 20. 537 20. 21ft 19.616 19.580 19.254 18.875 18. 623 17.954 17.828 17.610 16. 838 South Carolina 'J 1.7-7 24.089 23.989 28.898 23.793 23. 782 23.500 22.934 22.861 22.837 22 M 22 145 .j.j 206 22.068 21.914 21.869 21.793 21 . 678 21 345 20.988 20. 797 20. 793 20.731 20 831 20. 209 19.887 19.877 19.855 19.820 19. 681 19.634 19.448 18.154 18.144 18. 010 17.779 17.722 17. 698 17. CM 17. 544 17.288 17.078 16. 940 16.918 16.710 16.473 16. 395 15.183 22.641 22.004 21.048 ■_>ll 091 19.766 19.574 19.3 >s 19. 302 19. 259 18.865 18. 861 18.844 is. 711 18.599 18.533 18. 489 is. 152 18.406 18.362 18.289 18. -8S 18.239 18.198 18. 191 18.146 18.11,8 18.102 18. 091 18. 0"1 18.027 17. 998 17.996 17.885 17.843 17. 701 17.656 17.584 17.580 17.567 17. 422 17. 398 17. 224 17.088 17.018 16. 895 16. 777 16. 678 16. 651 16. 555 16. 338 20.980 18 7-n 18.264 is. 121 17, Ml 16.755 16.494 10. 23 1 16.231 10.117 1 , 996 15.885 15.870 15.735 15.588 1ft. S77 14.901 14. 699 1 1.665 14. 532 11. 186 14.342 1 1.208 14.137 13.967 13. 889 13.835 13.831 13.812 13.702 13. 550 13.473 13. 232 13.046 12.710 12.493 12. 471 12. 408 12. 257 11.0911 11. 607 1 1 12 1 11.365 11.162 10. 837 10. 73ft 10. 560 10.412 10. 230 10. 084 12.737 12.622 12. ".97 12. 482 12.21ft 12.084 12,071 11.782 11.694 11.663 11.601 11.070. 11.476 n ii 11.320 11.314 11.171 11.10.8 11.113 11.127 10. 737 10. 003 10.638 111 188 10. 349 10 246 10.160 10. 122 10.066 9.941 9.908 9.883 9.880 9.S66 9.255 9.194 9.073 8.884 8. 701 8.664 8.641 8.628 8.592 8.504 8.416 8.391 8.353 7.956 7.712 7.635 1 2 Dial of Columbia .. Wyoming Washington i 'aiitornia 2 Indian Territory ... ■1 Texas Rhode Island Washington Utah ft 6 Carolina North Carolina Texas Dist. of Columbia . Massachusetts New Hampshire. . i tonnecticut 7 8 9 Dist. of Columbia. Rhode Island in New Jersey Connecticut Rhode Island 10 11 .Ninth Dakota 11 12 Utah Massachusetts. - . . < lonnecticut New York 12 13 New Mexico Florida W.st Virginia Pennsylvania 13 1 1 Idaho Nevada Illinois Pennsylvania New Hampshire... North Dakota Ohio 14 15 West Virginia 15 16 West Virginia Mississippi North Dakota California .* Maryland 17 South Dakota Virginia New Jersey IS South I ia kohl 18 19 20 20 21 Minnesota 21 22 Delaware Pennsylvania Maryland 22 23 23 24 24 25 Minnesota North Dakota New Mexico 25 26 Indian Territory ... Arkansas Maine South Dakota New Mexico Wisconsin 26 27 27 28 Maryland New Mexico Iowa 28 29 Maryland Wyoming Delaware Washingti m New Hampshire. . . . 29 30 30 31 Ohio 31 82 Minnesota North Dakota 32 33 33 34 Pennsylvania Arizona South Dakota ... 34 :)."> Washington Oklahoma 35 36 36 37 Colorado Washington Rhode Island South Carolina Kentucky 37 38 Utah 38 39 Ohio West Virginia Louisiana 40 Virginia 40 41 West Virginia Utah 41 42 Massachusetts Indian Territory .. 42 43 Oklahoma North Carolina Indian Territory . 43 44 Dist. of Columbia .. Connecticut 44 4ft Louisiana Tennessee North Carolina . . . 4ft 46 New Mexico 46 47 New Hampshire California Dist. of Columbia .. Massachusetts New Hampshire Mississippi 47 48 South Dakota 48 49 North Carolina South Carolina South Carolina . . . 49 50 Utah 50 Table 39.— RANK OF THE STATES AND TERRITORIES IX THE PER CENT OF THEIR POPULATION OF KNOWN AGES IN SPECIFIED AGE GROUPS: 1900— Continued. 50 TO 59 YEARS. State <>r territory. Nevada Vermont Maine New Hampshire.. Dist. of Columbia. California Connecticut Rhode Island Massachusetts Indian Territory . Michigan Delaware Ohio New York Indiana i tregon Kansas New Jersey Maryland Pennsylvania Colorado Washington Iowa Missouri Illinois Nebraska Wisconsin Arizona Tennessee Virginia Idaho Kentucky South Da kota North Carolina . . . Oklahoma New Mexico Minnesota West Virginia .... Georgia Montana Florida Arkansas Alabama Wyoming Louisiana i la rolina . . . Otah Texas Mississippi North Dakota Per cent. an:: 343 273 223 171 .'71 016 . SOS .883 .871 .819 . 683 625 .611 .581 . 120 .403 .168 -Ills 998 815 811 7112 562 503 176 136 ::,S1 ;m ',. 248 191 159 1114 '.««, 906 659 657 562 544 520 lis 111 2|ii 211 60 To 69 YEARS. -Ian' or territory. Maine Vermont New Hampshire. . Nevada California Connecticut Massachusetts Michigan Ohio Dist. of Columbia. Rhode Island Delaware New York Indiana Iowa Kansas . < New Jersey Maryland " U iSi onSin i tregon Pennsylvania Illinois Virginia M issouri . Pei cent. (1 6 5, 5, i 4 I 1 4 4 4 4 I 4 4 I I 4 4, 7, :: Minnesota Kentucky Nebraska New Mi xico Washin Tenm ssi e South I' North < Utah Idaho West Virginia . . Arizona lo South Cat Georgia Mississippi . ... i iklahoma Alabama Louisiana as Texas . Montana 2 North Dakota 2. [g 2 Indian Territory 2 950 ii.si; 646 377 702 217 'J 77 882 827 771 716 711 7U2 661 160 145 127. 7,711 298 285 217 'I'll "17, 907 628 610 7,,-; 189 136 127 392 :;s7 325 304 236 21 1 1 II 12 1 1711 027 "17 "H7, 771 711 7 17 417 172 711 T" 79 YEARS. State 1 1! territory. Vermont New Hampshire. , Maine Connecticut \\ isci msin California ohin Massachusetts - - . Nevada Michigan New York Iowa Indiana Rhode Island Delaware New Jersi ; Maryland ' Pennsvh Dist. oi ' iiluiiiliiu Kansas Illinois Virginia Oregon Minnesota Utah Kentucky North Can linn - . . West Virginia South Dai ■ i South Carolina ... Louisiana ppi Washington New Mexico Idaho Colorado Texas North Dal i Arkansas -"a Indian Territory . Wyoming Prr cent. SO YEARS AND OVER. -tiilr nr territory. Vermont New HllMi|,-liir, Maine Connecticut Massachusetts Wisconsin Ohio Michigan Rhode Island New York Iowa Delaware Nevada Indiana Virginia Dist. of Columbia Maryland New Jersey North Carolina . . Pennsylvania California Mississippi Utah Louisiana Illinois Alabama Minnesota New Mea Kentucky Georgia .' West Virginia Kansas South Carolina... Arizona essee . * iregon Florida Nebraska South Dakota Texas Arkansas Washingl hill In i North Dakota Colorado Oklahoma Montana Indian Territory. Wyoming Per cent. 1 2 3 4 5 6 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 26 27 28 29 30 31 32 33 34 35 36 37 38 39 4ft 11 42 13 44 45 4t; 47 48 49 50 40 The number and proportion of children vn tht 'popu- lation. Perhaps the most important and interesting results of the age inquiry arc the statistics of the num- ber and proportion of children in the population. These statistic-- make it possible to come to some knowl- edge of the rale of the " natural increase" of the popu- lation of the differenl classes and in the same geographic areas. The proportion of children in the population is, of course, primarily a function of the birth rate. Table 40.— NUMBER AND PEB CENT OF THE TOTAL POPULATION UNDER 1 YEAR AND TJNDEE 5 YEARS OF AGE, CLASSIFIED BY SEX, RACE, .VXD NATIVITY, FOB CONTINENTAL UNITED STATES: 1900, 1890, AND 1880. but it is also affected by the rate of infant mortality, which varies widely among different classes of the pop- ulation. The migration of children in the first lew years of life is so small as to he. for most purposes, negligible. Table 40 shows the number and proportion of chil- dren under 1 and under ."> years of age, for the censuses of 1900, L890, and L880. lilOO 1890 1SSII BEX, RACE, NATIVITY, <>K NATIVITY OP PARENTS. Number. Percent. Number. Per cent. Number. Percent. i nder i Under 5. Under 1 . Under 5. Under 1. Under 5. Under 1. Under 5 Under 1. Under 5. Under 1. Under 5. Aggregate: 1,916,892 969, 2">7 947,635 9, 170, 628 4, 633, 612 4,537.016 2.5 2.5 2.6 12.1 11.9 12. 2 1,566,734 7911, 373 767,361 7,634,693 3,884,869 :;, 7 19, 82 1 2.6 2.5 2.5 12.2 12. 1 12.3 1, 117,9X3 734,024 713.9f>9 6,914,516 3,507,709 :>,, 400, so7 2.9 2.9 2.9 13.8 13.8 Females 13.8 Total white: Both sexes 1,665,007 844,238 B20.769 1,661,005 842,22] S1N.7M 1,157,534 587,815 569,719 503, 171 254, 106 249,065 4,002 2, (117 1,985 251,88 125, 019 126, B66 244,510 121,329 123, 18] 7,919,952 1,011, 155 1,908, 197 7,867,583 3,882,695 :,, 164,881 2,773,201 2,691,680 2, 102,702 1,211,687 1,191,015 52, 369 26,567 2 i,802 1,250,676 622,157 628,519 1,215,655 604, 187 611,168 2.5 2. 5 2.5 2.9 2.9 2. 9 2.8 2.8 2.8 3.2 3.2 3.2 (') (') (') 2.8 2.7 2.8 2.8 2. S 2.8 11.9 11. s 12.0 13.9 i '. 9 13.9 13.4 13.4 13. 4 15. 1 15. 1 15.2 0.5 0.5 0.5 13.7 13.6 1.: 8 18 - i.: 8 13.8 1 3 ,9 120 694,766 664,354 692, 626 662,288 941,657 182,794 158,863 413,257 209 s32 203, 425 !, 106 2. HO 2,066 207,614 104,607 103, 007 6.579,648 3,351,104 1 28 44 6,493,019 3,307,064 3, 185,955 i. i50,682 2,323,933 2, 226, 749 1,942,337 983, 131 I, 86, 629 44,040 1,055 045 ,33,71 , 521,280 2.5 ' i 3.0 8.0 2. 9 2.7 2.8 2.7 :; 6 3.6 3.6 el 2.7 2.7 2.7 12.0 11.9 12. 1 14.2 14.3 111 13.3 ; : 13.2 111. 9 17.0 16.8 1.0 0.9 1.0 13. 9 13 9 1,218,787 620, 296 598 191 1, 212.7:17 617, 307 595, 430 5,800,151 2,949,449 2. -,,n. 702 5,737,780 2, 918, 193 2,819 ..-7 2.8 2.8 2.8 3.3 3.3 3.3 1. 4 1.: .; 13.4 Native white: Mules 15.6 15. 7 15. 5 Native white — natr\ e pfl r- ents: Buthsexcs (Native white— foreign par- ents: Foreign white: 6,050 2,989 3,061 229, 196 113,72s 115 168 62, 371 31,256 31,115 I, in,".'.;, ■Y.-.2UI Vt, hi, 1 0. 1 :;. 1 3.4 8. 1 0.9 0.9 1.0 16.5 10.5 16.5 ■ ' Less than one-tenth of 1 per cent. Considering first the results for 1900, it appears that the smallest proportion of children in each of the two groups shown in the table is found in the foreign while population, and the largest proportion in the native white population of foreign parents. These figures indicate simply that the proportion of children among our immigrants is small, and that the foreign white population is contributing relatively more to the nat- ural increase of the population than is any other class. This latter fact does not mean necessarily a relatively higher fecundity for the foreign born population, for birth rates are not of much significance unless differ- ences in the sex and age constitution of the various classes of the population are taken into account. The age constitution of the foreign born population is espe- cially favorable to a high birth rate, for. as has been seen, a relatively large proportion of that class of the population is of adult age. It will be noticed that the per cent of the negro population under 1 year of age is the same as the cor- responding per cent for the native white population of native parents. This does not indicate necessarily 41 that the birth rates of the two classes are approxi- mately equal, for the errors in the reports of children's ages must be taken into account, and these have been shown to be more important for the negro than for anv other class of the population. Moreover, infantile death rates are usually large in the same classes of the population in which birth rates are large, a fact which tends to equalize the proportion of children in different classes of the population. In 1900, for the registration area, the ratio of the deaths of children under 1 year of age to 1,000 births was 142.8 for native white children, and ^!'7.n for colored children. The proportion of the total population in each of the two age periods under consideration decreased between 1880 and 1900, the decrease being more noticeable for the colored than for the native white population. Comparisons of the figures for 1890 and 1900 would be misleading, because of the abnormal amount of over- statement of children's ages in 1890, and because those tabulated as less than 5 years old in that year included only those children who were reported as less than 4| years old. Table -±1 shows the number and per cent of the pop- ulation in these age groups for the most importanl countries of Europe. Table 41. — Number and per cent of th< populatio I and under 6 years of age, for I peat countries. 1 NUMBER. PER CENT. CODS i RY. All known Under 1. [ i ■ . Under 1. Austria 1890 23,895, n 1 ^.OTKT'J. 7 . 12.8 Belgium 1890 6,069 321 2. 5 Denmark 1890 1 England and Wales - 1891 2. e 12. 2 1890 1.7 8.7 i termanj 1890 19, 128,470 1,428,687 i,, 128,836 J • Holland 1890 1,511,169 592,903 2.9 Hungary 1890 i 139 !2i 2, 150, 167 3.2 11. 1 1 pi land 1891 i ;n ,964 90,789 1.9 Hi.ii [taly 1880 155,94! 7m] 599 •1.1 12 1 Norway 1890 - 261, 126 2.9 1890 5, 049, 129 115,957 g 2.3 11.7 Scotland 1891 1,025,647 107,652 2.6 12.5 1890 17,252,472 154,880 ■ 12 1 Sweden 1890 4,784,981 121,835 2 5 12. ■> Switzerland . . 1890 2, 933, 334 65,969 . m 2. 2 11.0 1 Compiled from Bertillon, Statistique Internationale. Only in France, Ireland. Portugal, and Switzerland. of the 16 countries shown in the table, is the proportion of children under 1 year of age less than in the United States. This seems to indicate that "natural increase" is not so important a factor in the growth of the popu- lation of the United States as in the majority of Euro- pean countries. It is probable, however, that in most of these countries the ages of children are reported more accurately than in the United States. It should be remembered, also, that migration of adults has operated to increase the proportion of children in most European countries and to decrease the corresponding proportion in the United States. Table 42 shows the number and per cent of chil- dren under 1 and under 5 years of age in rural and urban districts. Table 42. — Number and per rent <>/ (hi total population under J ii. .-<■ inn! under ~> years of age, classified by s< a and as living in cities having at least 100,000 inlmbuants, in cities having between 25,000 and lOO.Ooo inhabitants, and m smaller cities and rural districts, for continental United states: SEX OB AGE PERIOD. IN CITIES HAY- 1 1 LEAST ■'P INHAB- ITANTS. IN CITIES I'll ING BETWEEN 1 "''II AND 100,000 INHAB- ITANTS. I N S M A L L E R CITI E.S A N 1> 1 1 : I I. D I S- TRICTS. \ Mill In l Per eent. Number. Per cent. Number. Per Under 1 year: 748,063 ■J. 3 ■2.2 10. 6 in - l-i. D 2.1 2. 1 2.1 10.0 1" i 732,843 7,115,838 8,599,776 3, 516,062 ■• 7 •' 6 ! :ilu^ ■ :. Bo1 h sexes 2.1 12.7 Females 12. 9 The proportion of children under J year of ace is smaller in cities of over LOO, 000 inhabitants and in cities having between 25,000 and 100,000 inhabitants than the corresponding proportion in smaller cities and rural districts. The same is true for children under •> years of age. In each case the difference is more marked for females than for males, but. that false inferences may not be drawn from this fact, it should be noted that in cities having over 25,000 inhabitants 4'.>.s per cent of the population are males and 50.2 per cent, females. The per cent which the number of children of each sex forms of the total number of children is substantially the same for cities and for rural district-. Table 43 shows the number and per cent of the pop- ulations of the states and territories that are under 1 and under 5 years of age. 42 Table 43.— Xl'MBER AND I'll; CI'.XT OF THE TOTAL POPULATION UNDER 1 AND UNDER 5 YEARS nF AGE: 1900. NUMBER. PER 1 1 I STATE OR TERRITORY. NDMBEB. PER CENT. Under l. Under 5 Under 1. Under ft. Under 1. Under 5. Under 1. Under 5. 1,916 922 9,170,628 2.5 12.0 Continental United States— Cont'd, North Central division— Cont'd Western North Central 263, 121 1,264,617 2.5 176 BIO 2,244,321 2.3 10.7 12.3 47,056 54,768 75,587 10,198 11,893 27. 793 35, B29 416. 533 !( 364 086 47. 783 55,217 133,747 172. 122 2, 016, 078 2.7 2. i 2. 1 8.2 3.0 2.6 2.4 3.0 13.1 11.8 117,940 554, 254 2 1 9.9 11.8 13,503 B.048 6, 755 60, 492 9,368 19,774 358, 870 65, 690 38,231 32,852 282,237 43, 452 91, 792 1,690,067 1.9 2.0 2.0 2.2 2.2 2.2 2.3 9.5 9.3 9.6 1" 1 10.2 10.1 11.0 North Dakota 15.0 Eastern South Centra] Southern North Atlantic 219, 169 1,055,904 2.9 14.0 61,790 57, 671 54, (in 45, 307 197, 364 284,230 274,923 267,300 229, IM 960, 174 2.9 2. 9 3.0 :; ii 13.3 13.7 43,57] 155,778 304,465 753, 490 206, 446 730,13] 1.447,579 2.2 2.3 2.5 2.9 10.4 11.0 11.6 13.9 14.7 14 8 Western South Central 14.8 119,817 ;«K.IM 2.7 12.6 39, 084 39,281 12,783 12. 112 93 B04 87,563 199, UK. 189,81] 59,985 58, ,30 152, : 12 423,997 2.8 3.0 3.3 3.1 3.1 2.2 14.5 Northern South Atlantic 14 ft 15.4 14 8 4. 167 28, 398 4,758 ft2.ftSN 30,106 184,648 19, 796 134,584 23,160 249,055 135. 465 885,529 2.3 2.4 1.7 2.8 3.2 3.1 10.7 11.4 8.3 13.5 14.2 14.9 Texas Western division 14 9 Districl of < loluml 10.4 30, 496 143, 285 2.5 11.7 Southern South Atlantic. . , . Montana 5, 629 1,653 2.10ft 11,950 6, 159 13. 100 26, 979 21,560 10, ft20 56 ' 27.227 60, 391 2.3 2 ',' 2.3 2.2 :;. 2 3.0 11.1 60, 255 43, 002 66,327 i ,,064 631,551 _•-:;, 712 203, 651 32ft. 473 72. 693 3,038,653 3.2 3.2 3.0 2.9 2.4 15.0 15.2 14.7 13. 8 11.6 13 1 13.7 3. 153 9,176 771 13 967 14,785 11,852 3,754 220 321 Eastern North Central 368, 4-J7 1,774,036 2.3 11.1 Utah 8.3 15.2 Nevada 1.8 9. Ohio S9.::ft'.i 57,993 114,392 54,161 52. 522 131, sin 274 799 550,036 260,65! 256,734 2.2 2.3 2.4 2 2 •Is 10.4 10. 9 11.-1 10.8 12. 4 Pacific 1.8 9.2 10,761 8,069 2:,, 137 41.111 125. 937 2.1 10.4 2.0 I" 1.7 While for continental United States as a whole the proportion of children under 1 is 2.5 per cent (or 1 in every 40 of the. population), this proportion varies in the individual states from 3.3 per cent (or 1 in every 30) in Utah and Indian Territory to 1.7 per cent (or 1 in every 60) in California and the District of Colum- bia. The proportion in Pennsylvania. Wisconsin, and Iowa is the same as that for continental United States. The states in which the proportion of children under 1 is greater than it is for continental United States, as a whole, form two well-defined groups. One of these areas includes all of the states east of the Mississippi river and south of the Potomac and Ohio rivers, together with Arkansas, Louisiana, Indian Territory, Oklahoma, Texas, New Mexico, Arizona. Utah, and Idaho. The other group is composed of Wisconsin. Minnesota. North and South Dakota, and Nebraska. The rela- tively high proportion of children in the states of the South Atlantic and South Central divisions is to be traced to two main causes — the large negro population and the relatively small urban population of this section of the country. Those states of the North Central division in which the proportion of children is relatively large are states that combine a large foreign born population with a relatively small urban popula- tion. New Mexico, Arizona. Utah, and Idaho, with their high proportion of children, present a contrast to the other states of the Western division. All of the states in which less than 2.2 per cent of the population are under 1 year of age. with the exception of Ohio. Nevada, and the District of Columbia, are in New England or on the Pacific coast. The per cent of children under 5 varies from 15.4 in Indian Territory to 8.3 in the District of Columbia. In Wisconsin, Arizona, and Iowa the per cent of children under 5 is approximately the same as in con- tinental United States as a whole (12.0 per cent). In general the rank of the states with reference to the proportion of children under 5 is about the same as their rank with reference to the proportion of children under 1. The greatest variations are found in New Mexico, which ranks sixth among the states in the percentage of children under 1 and only fourteenth in the percentage under 5; in Mississippi, ranking four- teenth and seventh, respectively: and in West Virginia, seventh and thirteenth, respectively. The significance of these proportions as indicative of the fecundity of the population of different classes and areas should not l>e overestimated. Other things being equal, the number of children in the population will vary, not 43 only with the infantile death rate, but also with the pro- portion of females in the population, the proportion of thos*e who are married, and the proportion of married women who are of child-bearing age. Accordingly, a more significant proportion is that of the numherof chil- dren in the population to the number of married women of child-bearing age. The presentation of such ratios. how ever, lies outside the -rope of the present study. Oth r nil. groups. — Adequate discussions of the popu- lation of military age, voting age. and school age will lie found in Twelfth Census. Vol. II. '11,, proportions of th. sexes in different ag< qroups. — The tables hitherto considered have show n the per cent of the total population of each sex who are in specified age groups. Table 44 shows the percentage thai the number of each sex foil!!, of the total population in specified groups. Table -i-i. — /'/ r cent of eacl groups, fo All known age? 5 years. Under 1 year. 15 to I" s ' :,,< ■JH (o 24 years ■ ... 35 to 39 yi a • - .... in 69 yean rs.. 75 i.i 79 years mi in 84 years... -'.' year- - re inn years and over . AGGREGATE. Per Per NATIVE WHITE. ' 19. I 51.2 111. 41.1 47.8 17 3 1- B M . I 54.0 61.4 50.7 19.9 111. 7 .... I 7.1 I i 18.0 lis Per cent IJ. 4 17. - 19 WHITE. Per ■111. 7 I 49.6 1'J. J 50.1 49.9 49. 1 17.4 57 1 51.6 54.3 17.1, 52 l' 17. s 7>il 7 51 :'. !8.8 i- 7 51.3 46.7 n n , , 39, 1 17. 1 34 7 male. 50.2 50.1 51.1 50.0 IS I 45. 7 60.6 60.6 65.3 The interpretation of the results shown in Table 44 is closely dependent upon the fact- with reference to the proportion of the sexes at birth. The census obtains the total number of births by adding to the number of children reported as I old at the dal the census the number of deaths of children who, if they had survived, would have been less than a year old at the census date. For the registration area the deaths of children under 1 year an- abstracted from the state or city registration returns; tor bhe rest of the United States they are obtained by the census enumerators. The reports show that of all births in the United States during the twelve months preceding tin- taking of the census of 1900, 51.2 per cent were of males.' For the registration area the per cent is 51.1. There is reason to believe that this percent is somewhat too low. In the states of Massachusetts, Connecticut, and Rhode Island the registration of births is fairly complete, anil what omissions there are may be supposed to be divided proportionately be- tween the sexes. The proportions that registered male births made of the total number of births in these states for the twelve months preceding the census were as follows: Massachusetts. 51.2; Connecticut, 51.7; Rhode Island, 51.7. The corresponding per cents as returned by the census were: Massachusetts. .Mi. 7: Connecticut. 51.0; Rhode Island. 51.0. It is probable that the true per cent for the United States is nearer 52 than 51. Table 44 shows that in the native white and foreign white populations this excess of males over females continues for the first three quinquen- nial age groups, although the excess of male- in each of these groups, as well as among children less than a year old. is smaller than the excess of males in the total number of births. In the negro population, on other hand, there is tor children under 1 and in each of the first two quinquennial i ipsan excess of females over males. The explanation of these facts is found in the different infantile death rales of the - and of the different classes of the population. Of the total number of deaths of children less thi year old among the white population of the registration 56.3 per cenl were of miles and i:\. 7 | were of females. Th ponding per cents for t colored population are 53.8 and 46.2. The death rate per 1. nun for white children under 1 in the fi tion area was 143.4, while the corresponding rate for colored children was 297. It is probable that on count of the overstatement of the ages of children under 1 these death rale- are too high, the exaggeration In ater for colored children than for white. How- ;. the fact that there are more females than males less than a year old in the colored population must be attributed to its excessive infantile death rate. Table 4."i shows the death rates of the sexe- in differ- ent age groups for the registration area. Tl rates can not be supposed to be accurate, but for the present purpose it is sufficient if they show 'lie relative death rates of the two sexes with approximate accuracy. Table 45. — Deaths at s " of «'» the fact that the registration area is not fairly repre- sentative of continental United States as a whole, since it contains a disproportionately large urban population. In the registration area 59 per cent of the population are iii cities of over 8,000 population, and 41) per cent are in cities of over 100,000 population. The corre- sponding per cents for continental United States are 31 and 19. Table 46 shows the death rates of the two sexes, classified by aye periods, for the population of the rural parts of registration states (including all towns and cities of less than 8,000 population). Table 40. — Deaths at specified ages per 1,000 population qfsamt ages for tli,' ruralparts of registration states: 1900. AGGREGATE. NATIVE WHITE. FOREIGN WHITE. COLOREH. Males. Fe- males. 103. C 31.2 3.2 5.3 7.3 8.2 15.1 73.6 Males. Fe- males. Males. Fe- males. Males Fe- Under 5 yeara 131.0 37.6 3.2 5.2 6.4 7.8 16.0 80 129. 5 37.3 3.1 5.1 6.6 7.7 15.1 80.1 102.3 30.8 :;. i 5.3 7.4 8.0 14. 8 72.5 106. S 26. 9 4.1 5.2 5.4 8.3 is.;t 79. 7 121.9 27.3 :; 3 5.0 6. 9 S.7 17.0 77. 1 246.9 711 6 5. 6 6.11 8.4 9.6 211. 6 79. 8 191.4 63. 6 6.4 15 to 24 i ears 25 to 34 \ ears 9.5 10 ii 13. 5 : t years 65 j ears and over - - 22. 5 us, 9 Table 4<1 shows that in the rural districts, through- out the period of middle life, the mortality of males is less than that of females. This throws light on the results shown in Table 44. In the native while population there are more males than females in each group below 15 years, but owing to the higher death rate of males in the period of child- hood and youth the groups of 15 to L9 and 20 to 24 years show an excess of females. During the period of mid- dle life the higher death rate of females in the rural districts seems to overbalance the higher urban death rate of males, for each of the age groups between 25 and 75 years contains more males than females. In the period of old age the mortality of males is greater than that of females in both cities and rural districts, so that the age groups composed of the years of life above 75 show a marked excess of females, this excess increasing as age advances. In the colored population the higher death rate of females begins at an earlier age than in the native white population, so that the group in to 14 years in the negro population (which constitutes 96.2 per cent of the colored population of continental United States) contains more males than females. With the exception of the group 30 to 34 years, in which the numbers of the two sexes an 1 approximately equal, each group between 15 and 45 years contains more females than males. In view of the fact that the negro popu- lation is preeminently a rural population (only L2.9 per cent of the negro population of continental United States living in cities of over 25, Out) inhabitants, as against 27. 7 per cent for the white population), it seems difficult to explain this excess of females in early middle life. All the registration states, with the exception of the District of Columbia, are in the North Atlantic or North Cen- tral divisions, and it is quite possible that the small colored population in these states is not fairly repre- sentative of the large negro population of the South. However, examination of the returns for the nonregis- tration area does not clear up the difficulty, for in each quinquennial age group between 10 and 50 years more deaths of females than of males are reported for the colored population. 1 These returns are admittedly in- complete, but there is no a priori reason to suspect that the error is a biased one. It may be thought that the death rates of 1900 are not fairly representative of those of past years, which have been the factors in shaping the present age and sex constitution of the negro population. An examination of the returns for the censuses of 1890 and 1880, however, shows that there has been no marked change in the age and sex distribution of the mortality of the colored population. 2 The population and mortality returns seem in this par- ticular to be irreconcilable. It is possible that during the period of middle life the deaths of females are re- ported more accurately than those of males. The proportion of the sexes in the foreign white population in different age groups depends on the rela- tive numbers of the two sexes among the immigrants of various ages, as well as upon their relative death rates. In view of this fact it is remarkable that the re- sults presented for the foreign white population shown in Table 44 show so small a deviation from those for the native white population. The excess of males in the total foreign white population is greater than in the total native white population, but this greater excess is confined to that part of the foreign white population which is in the age periods comprising the years of life above '25. The majority of the foreign white popu- lation, as of the native white population, in the groups 15 to i'.' and 20 to 24 years are females. The sex distribution of the population in different age groups for cities and rural districts is shown in Table 47. 62 and 63. Part J. litres 15to20; Tenth page xxxii. 'Twelfth Census, Vol. IV, pages 65 ! Eleventh Census, Vital Statistics, Census, Vital Statistics, Tart 1, page J 45 Table 47. — Per cent distribution, by sex, of population of specified age in cities having at least 100,000 inhabitants, in cities having between ?S,000 and 100,000 inhabitants, and in smaller cities and rural districts, for continental ' hited States: 1900. AGE PERIOD. All known ages Under 5 years Under 1 year ... 1 to 4 years 5 to 9 years In t<> 14 years 15 to 19 years 20 to 24 years 25 ti> 29 years 30 to 34 years 35 to 44 years 45 to 54 years 55 to 64 years 65 vears'and over IN CITIES MAV- IS'; AT I I \- J 100, ( N- HABITANTS. Per Ci n' male 49.7 50. 3 60.5 50. 3 5(1. 1 I'M' 17.:; I', 5 19. 1 51.6 52. 6 51.1 48. '.' 44.9 Per cent female, 50.3 49.7 49.5 49. 7 49. 9 50.4 52. 7 53. 5 50.9 IV 1 47.4 IS 9 51 . 1 55. 1 IN CITIES HAV- ING BE- TWEEN 25,- 000 AND 100,- 000 INHABIT- ANTS. Pel cent male. 49.7 50.3 50.3 50.3 50.0 49.2 47.3 47.0 49.6 51.6 52.3 51 :: 48.7 45.2 Per cent female 50.3 49.7 49.7 49.7 50.0 50.8 52. 7 53.0 50.4 48.4 17.7 48.7 51.3 54. 8 N SMALLER CITIES A N D RURAL DIS- TRICTS. Per cent male. Per cent female. 51.5 50.6 50. (', 50.6 50.6 50.8 50.4 50.5 51.7 52. 5 53.1 53.9 52. 4 52. 8.5 49.4 49.4 49.4 49. 4 49. 2 49. 6 49.5 48.3 47.5 16, 9 46.1 47. 6 48.0 In every group of the age classification used in Table 47 there are, in "smaller cities and rural district-." more males than females. If the period of 65 years and over, however, could be subdivided into smaller groups, it is quite certain that more females than males would be found in the groups comprising the more advanced ages. In cities there are more females than males in each of the groups of age between 10 and 30 years and in each of the groups for 55 years and over. This difference between the constitution of the population of cities and rural districts is due partly to differences in the numbers and ages of male and female migrants from the rural dis- tricts to the cities, and partly to differences in rural and urban death rates. Table 48 shows the death rates of the population in specified age groups for the registra- tion cities not in registration states. These cities are selected because they are more characteristically urban than the cities in registration states. All places of more than 8,000 inhabitants in which sufficiently accu- rate records of deaths are kept are " registration cities," but 70 per cent of the population of such of these cities as are not in registration states are in cities of more than 100,000 inhabitants. Table 4S. — Deaths at spt cified ages per 1,000 of the aggregate popula- tion ofsami ages, classified by sex, for registration cities mil in regis- tration states : 1900. SEX. Under 1 year. Under 5 years. 5 to 14 years'. 15 to 24 years. 25 to 34 years. 35t,,44 years. 45 to 64 years. 65 years and over. 194.0 156. 60.4 50.8 5.2 4.9 s. 1 6.9 10.8 14.4 28. 9 21. 6 103.8 Females 9.1 11.4 88.7 Upon a comparison of the rates shown in Table 48 with those shown in Table 46 for the aggregate popula- tion of the rural parts of registration states, it appears that for each age group the death rate is higher in cities than in rural districts. Moreover, in cities the mortality of males is greater than that of females for every age group, while in the rural districts the death rate of females, as already mentioned, is higher than that of males for the years of life between 15 and 45. The higher death rate of males accounts, in part at least, for the fact that in the urban population between the ages of In and 20 there are more females than males. However, the differences between the death rates of the two sexes for the age groups comprising the years of life between 5 and 35 are so slight that this excess of females suggests that women between 15 and 25 years old migrate to the cities in greater num- bers than do men of corresponding ages. Our statis- tics do not afford any absolute proof of this theory. In Europe, however, it is indisputably a fact. 1 Notwithstanding the fact that the differences between the mortality of males and females in cities are greater for the period 35 to 64 years than for the period 5 to 34 years, males are more numerous than females in the groups 30 to 34, 35 to 44, and 45 to 54 years. This fact can be explained only on the supposition that more males than females among persons of adult age migrate from the country to the city. 1 Biicher, "Ueber die Yerteilunjj der beiden Geschlechter auf der Erde," Allgemeines Statistiches Archiv, Vol. II, page 369 ff. APPENDIX A. THE ADJUSTMENT OE THE RETURNS. For most of the purposes to which census age returns are put, groups of five, ten, or more years are sufficiently detailed, and the discussion of the subject of errors has shown that such groups, when formed with due attention to the nature of the errors in the returns, may be supposed to represent the actual age constitution of the population with reasonable accuracy. It may happen, however, that for some purposes — such as the construction of a life table — it becomes necessary to know the number of persons in a given year of life. In such cases the errors in the returns stand in the way of satisfactory results. For this reason attempts have been made in various countries to modify the census returns in such a way as to eliminate as many of these errors as possible. In some ways the problem is similar to that with which actu- aries have to deal in the construction of life tables based on the experience of life insurance companies. On the other hand, how- ever, there is adistinct difference between the two problems. The errors in the unadjusted life tables are accidental errors arising from paucity of observations; that is, they would be eliminated if the number of observations were indefinitely large. The errors in the census age tables are systematic errors and take certain definite forms. The formulas employed in the adjustment of life tables are usually based on the theory of errors, and consequently can not be applied to the census returns, for until known errors are elimi- nated the law of error is not applicable. This point is mentioned, because in several instances formula- based on the theory of errors have been used in the adjustment of census age returns. 1 The thing to be desired in an adjustment of the age returns is a smooth series that will adhere as closely as possible to the facts. The aim should be to eliminate irregularities caused by misstate- ment of age, while retaining those corresponding to actual irregu- larities in the age constitution of the population. An accurate selection of this kind is, of course, impossible, but it should be noted that (in a graphic representation of the age returns) real irregularities would usually take the form of flexures covering a 1 For a list of such adjustments, as w ell as a general treatment of the subject under discussion, see an article by the present writer on "The adjustment of census age returns." in the Western Reserve University Bulletin, November, 1902. period of several terms of the series, while the irregularities caused by errors are more likely to appear as angular deflections, corre- sponding to abnormal values of single terms. It has already been shi iwn that groups of terms are more liable to be accurate than are single terms, and especially is this the case when the groups are so constituted that the probability of the equality of positive and negative errors is a maximum. To obtain the closest agreement with the facts such groups should contain as few terms as consider- ations of accuracy will permit. The discussion of age groups has shown that the quinquennial groups in which the lowest year is a multiple of 5 tit these conditions as closely as any others. Hence, in the present adjustment, it has been thought best to retain the different quinquennial groups intact, and simply to redistribute the numbers at the various ages within each group. The fact that the groups containing multiples of 10 are some- what larger, relatively, than groups containing odd multiples of 5 leads to some irregularities in the adjusted series, but it has been thought that these are more than counterbalanced by the advantages of the agreement of corresponding groups of terms of the adjusted and unadjusted series. The method of adjustment used has been adapted from that used in the English census. Intermediate values were interpolated by the method of differ- ences in the -cries log Q^ log Q l0 , log Q 16 , log Q ao , etc., in which ij represents the number reported at and above the age x. After the adjusted numbers at and above each age had been found in tin- manner, the number at each age was obtained by subtrac- tion. A separate interpolation was made for each sex of the native white population of native parents, the native white popu- lation of foreign parents, the foreign white population, and the colored population. The results for both sexes, for the native white population, and for the aggregate population, were then obtained by the proper additions. For the groups 5 to 9, 90 to 94, and 95 to 99 Newton's coefficients were used, while Bessel's coeffi- cients were used for all the other groups. In both formulas five orders of differences were used. The ages under 5 were not included in this adjustment, because the mortality in the first few years of life is so high that an adjustment of these years based on the pro- gression of the Q, series taken at quinquennial intervals would : the facts. 147) 48 Adjustt d " 1,916,892 7,25 3,664,355 947, 635 3,589,381 1,661,005 6,201 8 12,221 5,142,667 R18.784 3,063.911 3 4 1,768,078 1,830 332 1,824,312 1,831,014 8,874,123 893, 2113 925 160 920, 335 4,479.396 874,815 905, 072 903, 977 9(15,517 4,394,727 1.521 536 1.555 , 15 1,559 1,561.399 7.491,134 770, 755 791,936 789 ' ■ 790,5-1 .,788,622 750. 781 771,509 77 ; 770,818 3.702,512 s ' 7 R 9 1,801,004 1,804,1 15 1,787,1119 1,768,605 1,723,350 8,080 '. ,536 907, 644 904,428 891,745 .87:,, 043 4,0,83,1111 900, 168 B96 .01 382 .9] 866,860 848,307 3,997, 195 1,5(12,759 1,533,296 1,500 -oil 1,464,882 1,429.301 6, 647, 673 785,654 774,07,8 760, 355 725 525 3,361,671 777,105 759,218 710.541 721.-70 703, 778 3,286,002 III 1 1 1 • IS 14 10 to 14 years IS 1,675,884 1,643,425 1,616,163 1,588,650 1 . 556, 1 12 7,556,089 8 19, 18 1 832,569 816,51 1 -mi. 542 3,750, 15] 825 100 -IK 8 16 799,619 7--., LOS 772, 180 3,805,538 1,394,324 1,7,5(1, 035 1,329, 169 1,298,853 1,265,001 5,981,443 706, 763 5-9. (172, 1911 555, 197 638 359 2, 981 1. 709 6,-7,551 671,020 1157,275. 643. 356 626, 792 2,994,734 Hi 17 18 M •m °1 1,549,050 1,504,562 1,513, 199 1.517,980 1,471,292 7,335,016 7(17.207 7 16,343 7 >i 706 744,2,87 731,908 3,624,580 781,8 19 748,219 762, 493 775,1,95 739, 384 3, 710, 436 1,257,352 1,203,8114 1,197,181 1,187,923 1,135,120 5, 11.5, 302 535.753 509, 758 597, 578 5-5,11,53 570,502 2,689,295 633. 564 594.106 599. 606 602,840 564,618 2.726,267 ■-. ■>4 ffl ■•■; o 24 years ?7 1,501,839 1, 191,(144 1,468,846 1, 144,638 1,428,049 6, 529, 441 738. 742 733,33:, 723,881 717.318 709, 104 3,323,543 755,097 758, 109 712,9(5. 727, 320 718, 945 3, 205, 898 1,132,818 1,111,113 1,083,884 1,056.072 1,031,675 4, 665, 751 561,816 550, 172 538, 268 525, 937 513, 102 2, 353, 361 571,002 560, 941 545,616 530,135 ,515.573 2,312,390 ?8 .,,, 311 31 ::• 25 to 29 years 33 1,387.321 1,34,8,325 1,305,117 1,262,931 1,225,747 5,556,039 698, 628 682, 798 664,643 646, 676 630,798 2, 901, 321 688, 693 665, 527 640, 474 616, 255 594, 949 2,654,718 1, 003, 748 969, 453 932, 411 595.290 863, 849 3,830,761 504,134 487,971 470, 384 453, 177 437, 695 1,958,744 499,614 181,482 462, 027 443,113 426. 154 1,872,017 34 ■;:, ■in 37 35 30 to 34 years 39 1.173.097 1,138,972 1,111,668 1,083,540 1,048,762 4, 964, 781 608, 377 592, 734 580, 734 568, 074 551.402 2, 616, 865 564,720 546, 238 530, 934 515, 466 497, 360 2, 347, 916 823,881 792, 289 755, -8,; 739. 496 709,209 3, 283, 009 419, 790 404,609 391, 685 378, 711 363, 949 1,687,544 404,091 387,680 374,201 360, 785 345,260 1,595,465 40 41 1" 43 44 45 1,041,370 1,018,521 994,692 969,033 941,165 4,247,166 547, 278 536, 066 524,390 511,637 197, 494 2, 255, 916 494,092 482, 455 470, 302 457, 396 443,671 1,991,250 689, 564 671,343 658, 101 643,021 620, 980 2, 886, 031 353,503 344.529 338, 345 331,132 320, 035 1,497,876 336, 061 326,814 319, 756 311,889 300, 945 1,388,155 46 47 4H 49 5(1 SI 915,642 883, 678 851,008 813, 558 783, 280 3, 454, 612 485, 420 469, 071 452,939 431.731 416, 755 1,837,836 430, 222 414, 607 398, 069 381,827 366, 525 1,616,776 030 813 601,661 579, 073 551,522 532, 962 2, 265, 458 321,155 311,820 301, 492 285, 978 277, 431 1,183,506 299, 658 289, 841 277, 581 265, 544 255, 531 1,081,952 59 .Ml ;.i 55 :.rl 57 739,734 712, 045 690, 538 670, 526 641,769 2, 942, 829 392,966 378,5(19 367, 159 357.471 341.731 1,564,622 346, 768 333, 536 323, 379 313,055 3(1(1,113,8 1,378,207 495, 692 472, 324 451,790 433, 855 411,797 1,830,589 258. 174 246, 414 235, 791 227, 432 215,595 955, 956 237, 518 225,910 215, 999 206, 423 196.102 874,633 58 59 fin 61 IV fi3 642, 418 618, U6 588, 278 558,816 534,901 2,211,172 343, 466 330, 080 312, 669 295, 774 2-2,353 1,145,257 298,9 288, 336 275, 609 263, 042 252, 258 1,065,915 402, 9S8 385, 101 365, 735 346, 846 329, 919 1,378.211 212, 083 202, 200 190.577 179, 964 170,832 694,994 190,905 182, 901 171,85s 166, 552 159.087 683, 220 M 65 fir, 67 68 69 486, 976 461,194 442, 372 422,941 397, 689 254, 523 239, 725 228, 936 218,002 204,071 232, 453 221,469 213, 436 204, 939 193,618 305, 646 289. 128 275, 537 261,856 246. 047 15,5,705 115,251 138,811,8 131,406 122, 791 149.938 112.8 17 136, 729 130. 450 123, 256 7(1 56 years 71 7? 73 1 Not adjusted 49 and nativity, for Continental United States: 1900. NATIVE WHITE— NATIVE PARENTS. NATIVE WHITE — FOREIGN PARENTS. OEEIGN WHITE. RED. Total. Males. Females. Total. Males. Females Total. Total. Males. ' Females. 40,949,362 20,8 19 B47 20.099,515 15,646,017 7,836,603 7.S09.414 10. 213, "17 5,51 . 9,185,379 4.614.713 1 1,157,534 4,307,347 587, 815 2, 185, 386 569, 719 2,121,961 503, 471 1,899,23] 254, 406 957,281 941,950 48,367 2.017 2 1,. 7.70 1,985 23.S17 2.71, "S3 998, 791 125,019 497, 138 2 i 1,067,983 1,087,237 1.1174,499 1.077.62s 5,174,220 541,787 551,514 544.814 547,27] 2,623,791 526, 196 :.:::.. 723 529,68 i 530,357 2,550,429 153,553 176,508 185,399 2,316,914 228. 96s 240, 122 243, 310 1,164 831 224,585 2 10. SIS 21", 16,1 1,152,083 . !.'• ■ 147,192 5, 117. ". 117 73, 727 3,700 5,224 6, 785 -, 108 73, 465 2"", 352 1,23.7,737 lis. .76,3 127.911 124, 163 126.. 701 617.047 126, 591 618,750 A t ( t 1 . 1177, 152 1,056,756 1, 035, 242 1,013, 173 '.(■11 '-'.17 l 660,390 541,404 ,34,1 91 526. 054 616, 150 505, 192 2,364,797 535, 748 522, 065 509,18" 197,223 480.205 2, 295. 593 485, 607 17,., 540 451.509 1,98; 244,250 - 996,874 241,357 237, 153 '1 .:i 9 22,313 31.7 il 44, 677 311,565 3,194 16,191 20, 267 157,632 6,374 11,, ,, 21,288 153. 933 228,677 251, "311 1,120 ill, 688 123, ""3 127,879 12.". 466 116,989 125, 653 17 1,50 126.36 1 123.241 c 11 1 i: i l 971,278 950,6 A 932, 233 913,582 392 643 4,234,953 183,276 172,967 162,637 451,926 2,122,635 177,287 '■,. • , a 150 94 llli 717 2,112,318 123,046 409, 372 397,236 385. 271 :7J 3 ■- 1,746,490 212,772 15,730 199,229 [92 ...i 186 283 864,0' : 210,271 19". 007 192,411 186 07 7 882, Hi. 46.771 62,47] 77. 70 561.746 24, 131 77,671 31,59 . 35. 46.7 38,787 271, 381 22, 640 26.231 30, "76, 3.7.423 38,763 290, 365 234,789 229, 71 1 221.223 218,909 213,561 1,012, 00 118,590 11.7,909 112.723 109,580 10,1.936 116, 199 113, 605 111,500 106, 625 520, 539 l l l i l 2 893,330 848,229 847, 188 844,6110 801,226 3,805 609 442, 401 133, 163 424,657 115,991 406, 123 1,903,864 450,929 415,066 122,831 I2S.6S9 394.803 1,901,745 364, 022 355 63 - 349, 696 343.243 333, 894 1,609,953 181,387 176.. 795 172, 921 169,092 16,1,079 785, 131 182,635 179,010 1 76, 775 1,1 1 ,1 169,815 B24, >22 -i. 98, 255 112,86 : 126,914 136,960 919, 182 12, 288 17,717 54, 187 61.226 65,863 456, 186 44, 466 50, 738 58,376 65,688 71, 097 463, 296 20 1,9.70 202, 1 13 203. 152 203. 143 199, 212 999, .'72 101,131 99, 06S 98,641 97.97" 103, 819 103, 375 104, 511 105, 165 103,669 2 2 2 2 2 2 802,231 7*3, 131 760, 983 738, S16 720, 148 3,208,642 loo.ooi 390,946 380,765 370,631 361,461 1,634,867 402,170 392, 185 380,218 36s, 1"., 37.8,687 1.573,775 330, 587 327,682 322.901 317.256, 311,527 1,457,109 16,1.7.75 159, 226 157,503 155,306 151,641 71". 194 168, 332 16", 156 161.950 159,886 73", ,,1 . 159,684 173,524 184,786 195, 197 206,291 1,09 77, 107 84,640 91,699 9", 297 104.447. 589,521 82, .777 88,884 96, :»i" 101,848 507, 708 209, 337 200, 176 19 69 190.0":; 766, 461 99. 319 98, 723 95,914 93.0"! 91,. 779 880,661 109, .71" 108,284 104,262 100,285 98, 524 385,800 2 2 •. 3 3 3 690,693 666, 909 640,886 616,747 594, 408 2, 659, 360 349. 425 388,223 326, 680 315, 455 305, 084 1,372,529 341,268 327,686 311.205 301,292 289, 324 1,286,831 313.055 303, 544 291.526 279,543 269, 441 1,171,401 154, 709 149, 748 143, 704 137, 722 132, 611 586,215 I ,.; 7'... 147, S22 141,821 136, 830 ,- . !-, 209, 240 215, 147 220, 208 22 1 , 51 1" 228, 126 1,178,683 109, 174 111,046 11", 131 122,290 125, 577 660, 702 100,066 101,101 101,771 102, 218 102,549 512, 981 174, 333 163,725 152, 498 142, 133 133, 772 551,595 85, 320 B0, 78] 7.7, "2.7 71,209 67,526 281, 875 89, 013 82,944 76, 673 70. 924 66.246, 269, 720 3 3 : 3 3 3 ,.7,,.::ss 549,505 531,373 513. 653 494, 441 2, 299, 571 293, 394 283,260 274,305 265,512 256, 058 1,192,071 276, 994 266,245 257,068 248, 141 ■.Ms. 383 1,107,500 253,493 242, 784 234, 513 225. si:: 214, 768 983,438 126,396 121,349 117.380 113,199 107,891 495, 473 127.097 121,135 117,133 112.641 106, 877 487, 965 229,0] 1 233, 18] 235. 6,03 237.183 238,702 1, 177,566 127, 490 130, 170 132. 733 134,685 135, 624 672,804 101, 624 103,011 102,870 102, 19" 103, 078 504, 762 120,202 113.502 110,179 106, 861 11,0, -.,,1 504,206 61,097 67,955 56,316 54,678 51,829 256, 517 59. 105 55, 547 63, 863 52, 183 49,022 247, 689 ? 4 1 4 1 479, 973 468, 197 460, 438 451,968 438,995 2, 104, 551 248. 401 242, 192 238,724 234.544 227,910 1,096,825 231,572 225,705 221.714 217, 424 211.085 1,007,726 209, 591 203, 146 197, 663 191,053 181,985 781, 480 105, 102 102,037 99, 621 96,588 92,125 401,051 104, 489 101,109 98, oi2 94,465 89,860 380, 429 244 788 212.397 235, 728 229,015 225, 690 966, 112 139,089 138, 127 134,756 131,284 129, 548 557,300 105, 649 104,268 100.972 97,731 96. 142 408, 812 li,7,"6," 104.783 100, 863 96,997 94, 495 395, 023 54, 686 53, 410 51,289 49, 221 47,911 200,740 52, 382 51,373 49, 574 47, 776 46, 584 194, 283 4 4 4 4 4 6 441,543 432, 701 421.441 409, 734 399, 132 1,787,607 229, 554 225, 22 1 219, 709 213, 887 208, 451 937, 254 211,989 207, 177 201.732 195.817 190, 681 850,353 179, 270 168, 960 157,632 141, 788 l 13 530 477.851 91,601 86,596 81,783 72,091 68,980 246,252 S7.669 82, 364 75,849 69, 697 64,850 231.599 209, 161 200, 344 192, 979 185,809 177, 819 840, 220 121, 112 116, 034 111,298 106.715 102, 141 468, 466 88,049 84,310 81,681 79,094 75, 678 371,754 85,668 81,673 78,956 76. 227 72, 499 348, 934 43, 153 41,217 40, 149 39,038 37, 183 186, 864 42,515 40, 456 38,807 37 I"" 35, 316 163, 070 5 6 5 5 5 5 378, 463 366. 901 357.381 349,255 335, 607 1,551,811 197. S4S 192, 124 187, 126 183, 785 176, 371 811,724 180, 615 174,777 170, 255 165. 470 159,236 740,087 117.229 105, 423 94. 409 S4.600 76.190 278,778 60, 326 54,290 48, 665 43, 617 39, 324 144,232 56. 903 51,133 15,744 40,953 ::,. 866 134,546 172, 499 169, 203 168, 592 167, 281 162. 645 803, 392 97. 348 94, 806 93, 950 92, 742 89,620 440,079 75, 151 74,397 74,642 74,539 73,025 363, 313 71,543 70, 518 70, 156 69,390 67, 327 308, 848 37, 444 37,289 37.418 37,297 36, 416 168, 5S7 34,099 33, 229 32, 738 32,093 30.911 140, 261 a 5 7 1 e 335, 182 324, 138 310, 554 296, 922 285,015 1,204,610 177, 009 170, 653 162, 323 154, 140 147,699 605,625 15S, 173 153. 485 148. 231 142.782 137.416 598, 985 67. 806 60, 963 55,1 si 49, 924 44.904 173, 604 35, 074 31,547 28,554 25,824 23. 233 89, 369 32,732 29. 116 26, 627 24. 100 21.6.71 84,235 169, 269 166,410 161,058 155, 146 151,209 643,003 93, 322 91, 451 88,199 84,833 82,274 345,241 75, 947 71,9.79 72. S.79 70. 613 68,935 297. 762 70, 161 66, 905 61, 185 56, 724 53, 773 189, 955 3", 961 36,429 33, 593 30, 977 29, 527 ln.7.,,22 32. 100 30.476 27,892 25, .547 24, 246, 84,933 ■ , : 264,339 261,446 241,045 230, 341 217, 439 134, 312 126,834 121, 065 115,242 108, 142 129, 997 124, 612 119,980 115, 099 109, 297 41,307 37, 682 34, 492 31,515 28, 60S 21,366 19,447 17,743 16,164 14,649 19,941 18,235 16, 749 15.351 13, 959 138. OS", 132,864 128, 876 124.473 I 118.101 75, 172 71, 685 69, 143 6 62,804 63,517 6.1,179 59, 733 58,036 55,297 42,6.41 39, 202 37,959 36, 612 33, ,11 23,,,13 21,7.79 20,985 20, 159 18, 476 18, 99s 17,443 16,974 16, 453 15, 065 - 50 Adjusted ages of tlu population classified by sex, race, and iGE I AGGREGATE. NATIVE WHITE. Total. Males. Females. Males. Female 1.7'.' 1 ' S7I, 196 1,075,627 539. 430 -- 394,987 339 772 1,302,926 10 ■ r, i 183,863 173,977 L92, 113 184,380 166 15 ... ■,. 23s, 772 226,970 215. 125 1 132 119. 605 113,711 Ills. (173 102,206 95,836 395,274 115 1.. 113 [01 llll. 322 g . ii-. 76 77 7s 7" M 297,371 27.x, SI 7 200, 859 242 935 ■223,4 14 683,841 152.476 142,719 133. 707 124. 421 114,343 149,609 1 14,895 127, 162 lis. .ill 109,101 434,232 179,678 168,061 156,943 145.741 134,142 52.5. 772 90 169 84,681 79,109 75, 161 67,664 263, 590 89 209 77,834 72 290 2112,182 8" 8 1 65 R7 208,972 192,464 176,855 161,032 114,. 518 106,573 98,040 89, 995 81 , 794 73, 207 261 , 579 102,399 94, 424 86,860 79, 238 71,311 258,278 123, 818 114,053 105,232 96,219 v,;. i ,ii 318.. Mil 112,297 57,296 52,75 1 48,118 43,115 61,521 56,757 52,411,8 48,101 151 153 SS 8q 'in 91 93 SI 95 96 47 130,101 116 409 103 504 91,123 78,660 251,512 65,906 5S 824 62,050 16,634 122.273 64.255 51, 154 45, 489 39, 195 129,239 79.214 71,269 63 548 56, 1 18 48,625 155, 642 35, 312 31, 338 27, 613 25, 751 74, 697 39,890 5... (157 32,210 24,861 . 80, 945 1 qq 68,649 58,650 19 131 41 , 080 33, 702 88,600 33,846 28,700 23,987 19, 741 15,999 40,742 34, 803 29,950 25, 1 1 1 21,339 17,703 47, 858 42.095 36, 118 30,742 25,702 211,1185 55, 860 20,522 17. 172 14,722 12, 168 11, 813 25,036 21,573 18,646 16,020 15. 5.1 11 172 30, 82 1 urn ins 104 105 Iflfi 27, 184 21,544 16,811 13, 022 10, 039 23, 992 12, 805 10, 007 7,671 5, 833 4,426 9,858 1 1,3711 11,537 9, 1 in 7,1 Ml 5,618 14,134 17 271 13,699 10,623 8, : 19 6,148 12, 998 7. 922 6,193 4.731 3 .. : 2,636 5 268 9,5 15 1...I . 3,51 7.75H 107 110 in 7,537 5, 722 4,408 3, 478 2,847 6,266 3,267 2,415 1,795 1,347 1,034 2,417 1,270 3.31 17 2,613 2,131 1,813 3,849 4, 551 3,299 2,352 1 . 1.51 1 115 2 ::<■'■ 1.902 1,348 543 435 796 2, 6 19 1 ■' 5 1. 112 1,008 7111 1, 199 ii" 115 2,029 1,558 1,183 878 618 3,504 763 587 451 348 268 1,271 1 266 H71 732 530 350 2,233 - 13 582 1111 281 195 446 300 204 137 92 63 152 533 57s 257 189 132 294 51 nativity, for Continental United States: 1900 — Continued. NATIVE WHITE— NATIVE »ARENTS. NATIVE WHITE— FOREIGN PARENTS. FOREIGN WHITE. ' "I'iRED. Total. Males. Females. Total. Males. Females. Total. Males, Females. Total. Males. Females. 965. 900 483, 454 182, 146 109,727 55, 976 53. 751 285, 7*3 259,248 170,705 91,954 78,751 74 212, 323 202.912 193, 653 183,861 17:;, Til 106, 130 101,460 96, 961 92, 151 86, 752 106, 193 101, i i 96 1 9 91,710 86,399 26, 1 19 24,058 21,772 19,667 17.7S1 13, 475 12, 251 11,112 10, 055 9,083 12, 974 11,807 10, 660 9,612 8 698 118,260 11 1,009 109, 176 101, 152 98,834 62, 154 59. 921 57, 360 54,576 51, 169 54,0V. 52,116 47. 865 36,602 34, 173 31.792 20,515 19. 58,8 is, 180 17, 195 16,248 V, 140 17,036 15,712 11 9, 13,93 > 75 76 77 7S 79 720.110 361,980 358, 130 64,445 33, 294 81, 1M 410,710 213, 111 197 299 107,631 58,954 48,677 80 163.763 153, 800 144,199 134,3 * 123, 990 S2.26S 77,319 72.5211 67,564 62, 309 81,495 76, 181 71,679 66,794 ■ 1 681 15,915 11,2.1 12.711 11,383 10,152 8,201 7, 362 6, 589 5,887 5.255 7.714 6,889 6. 155 5 i ii; 4,897 92, 552 87, 271 82, Oil'. 77, 255 71, lllli 18,031 15,278 12, 868 40, 222 37, 042 14. 521 41,993 37, 033 34, 154 25. 1 1 1 22, 995 21, 151) 19. 939 18,106 13,976 12,760 11,7811 10, 751 9,787 11,165 10,235 9 120 9, iss 8,369 81 82 S3 si B5 488,649 244, 57 1 241,075 37, 123 19, 016 18, 107 282,325 148,068 134.257 75, 744 , 37, 793 86 114, 841 105. 963 97, Ml 89, I'.'l 80,510 57,656 63, 130 48, 983 44. 707 10,098 57. 185 ,2,833 18,858 11,7^7 in 112 8,977 8,090 7,391 6, 725 5, 940 4,641 4,166 3,781 3,411 3,017 4,336 3,924 3,610 3, '1 1 66,941 61.713 51,295 45. 793 35, 052 29, 701 26 927 24,023 31.SSVI 29 348 21,770 I'.'l i 16,698 18.51s 12,275 9,221 6,719 6,069 8,989 8,319 7,510 6,769 6, 206 87 88 89 90 91 296, 201 146, 108 150 093 22, 603 11,243 11.360 1,,- 'i|i 42. 137 21.295 20,842 92 73, 636 66,239 ,9,036 52, 149 45,141 36, 540 32,810 29,093 25,625 22, mo 33,429 29. 948 26, 524 23, 101 5, 57 s 5,030 1,512 8, 184 2,784 2,502 2,215 1,988 1 72 1 2, 794 2, 528 2,2H7 2,1111 1.7611 III, -Mi ,:i, 072 31, 'i- 27.211 23, 169 21,394 18,854 16.473 14.192 12,020 19,492 17,218 .... 13,049 11,149 10, 061 9,068 s. 108 7.731 1,658 8, 829 3,381 4,410 1 169 8,9115 3, 185 93 94 95 96 97 144.314 69,087 75. 227 11,328 5,610 5,718 69, 075 35. 390 .;:; | 28,. 795 12. 186 14,609 98 39,086 33,513 28, 196 23,803 19, 116 19,026 16, 177 13, 609 11,231 9, 1 1 20, 060 17, 386 14,887 12.572 10,372 3,009 2, 605 2,211', 1,899 1 , 569 1. 196 1 . 295 1.113 937 769 1,513 1,310 1,133 962 800 19, 559 16,338 13, 118 10,982 3,803 10, 102 8, 118, 6,882 5, .'i.'.l, 1, 187 9. 457 7,925 6,561 5, 376 4,366 6,995 6,194 5. 2 16 1, 118 3,914 3.222 2,815 2, 8S3 2,1117 1,749 3,773 3,379 2,863 2, 129 2. 165 99 100 101 102 103 51,542 22, 992 28,550 4,318 2,044 'J, '.'71 22. 167 10,802 11,365 10, 573 4,904 5,669 104 15.950 12,643 9, 802 7. 186 5.661 7,275 5,686 1,348 3,266 2, 117 8,675 6, 957 5, 151 1,220 3,244 1,321 1,11,1. 821 633 487 647 507 383 288 219 674 549 438 345 268 7,859 5. 181 1,130 3, 534 2. 689 1.99< 1 1,474 1.115 3. 525 2,772 2,140 1,648 1,280 2,854 2,058 1,781 1, 196 1,8,19 1,125 950 mi:, 1,75 1,505 1,259 1,108 976 821 105 106 1117 IDS 109 11, 940 4,815 7. 125 1,058 453 605 5, 321 2,339 2,982 2 251 3, 422 110 4,184 3.032 2. 159 1,515 1,050 1.740 1,233 858 587 397 2, 111 1,799 1.301 928 a 367 267 193 136 95 162 115 82 56 38 205 152 111 80 57 1.788 1,327 975 7ii7 516 819 590 423 299 208 977 737 552 408 308 1.18H 1,096 n 1, 120 1,186 546 477 432 105 391 644 619 649 715 795 111 112 113 114 115 2, 077 695 1,382 218 101 117 1 112 609 803 2 . i9 1,012 1.547 116 756 527 365 251 175 26S 179 US 7S 52 188 348 217 176 123 77 55 39 27 20 32 25 19 14 11 45 30 20 13 9 10 831, 266 212 168 181 143 116 94 76 249 193 151 lis 92 766 6111 513 385 255 232 2111 199 162 129 484 400 314 223 126 117 lis 119 120 121 393 129 264 53 23 30 391 178 213 2,667 941 1,726 122 / APPENDIX B. BIBLIOGRAPHY OF DISCUSSIONS OF AGE STATISTICS. In addition to representative modern discussions of age sta- tistics, a few older studies of primarily historical importance have been included in the following list of references. Valuable dis- cussions will also be found in Census Reports, especially those of Germany, Switzerland, Hungary, Italy, India, and New South Wales. J. ( hius ill an nta'irr il, stiiii.iiii/K,, Paris, 1896, page Traiti ili Statixiii/m. 'leuxieme edition, Paris, 1886, page Bkrtili.on, 460 ff. Block. M. 432 ff. CoNRAn, J. Grundriss zum Studium der politischen Oekonomie, Vol. IV, Statistik, I, Jena, 1900, page 68 ff. Exgel, E. Der Werth des Menschen, Berlin, )ss3, Vol. I. page 57. Faer, Wm. Vital statistics, London, 1 885, page 37 ff. Fircks, A. Freiherr, von. BevolkerungMehrc uml lleeiilh rungspolitik, Leipzig, 1898, page 67 ff. Haisuofer, M. Lehr und Handbueh der Statistik, Vol. II. zweite auflage, Vienna, 1882, page 209 ff. Holmes, Geo. K. Age, etc., in "The Federal Census," Publications of the American Economic Association, New Series, No. -'. page 55 ff. Jastrow, J. Some 'peculiarities in the age statistics of the United sini, s. Science, Vol. V, page 461 ff. King, W. A. The decrease in tin' proportion of children, Political Science Quarterly, Vol. XII, page 608 ff. Levasseur, E. La population franc.aise. Vol. II. Pari", 1891, page 257 ff. Mayr, i ;. vox. Statistik und Gesellschaftslehre, Vol.11, Bevolkerungs- statisiik, Freiburg, 1897, page 73 ff. Mayo-Smith. Statistics and sociology. New York, 1895, page45ff. Mischler, E. Altersgliederung der Bevblkerung, Worterbuch der Volkswirtschaft, Vol. 1. page 59 ff. Morpurgo, E. Die Statistik und die Sozialwissenschaften, Jena, 1S77, page 498 ff. Pareto, V. Cows cTiconotnie politique, Vol. I, Lausanne, 1896, page 75 ff. Perozzo, L. Studi sulla composhione delta popolazione per tin in Italia i' m altri stati, Annali di Statistica, Rome, 1885. Rauchberg, H. AUersgliederung del Bevblkerung, Hand worterbuch der Staatswissenschaflen, zweite auflage, Vol. I, page 275 ff. Die Bevblkenngs Oesterreichs, Vienna, 1895, page 106 ff. Schmoller, G. Grundriss 'la- allgemeinen Volkswirlschaftslehre, Vol. I, Leipzig, 1900, page 159 ff. Stratss, C. Allersverhdltnisse der Bevblkerung verschiedener europ'd- isch r inn! ausser-europaischer StaaU n, Jahrbticher fur National- okonomie und Statistik, dritte folge, Vol. II, page 905 ff. Wag nek, A. Grundlegung >/> r politisch n Oekonomie, dritte auflage, Vol. 1, Part 2, Leipzig, 1893, page 606 ff. Wright, Carroll I). Outline of practical sociology, New York, 1902, page 33 ff. Young, Ai.lyn A. The comparative accuracy of different forms of quinquennial age groups, Quarterly Publications of the Ameri- can Statistical Association, Vol. VII, page 27 ff. The enumeration of children, Quarterly Publications of the American Statistical Association, Vol. VII, page 227 ff. The adjustment of ct nsus one returns, Western Reserve Uni- versity Bulletin, November, 1902. o (53)