ing the year 1914 giving greater ing from 29.3 per cent. in 1911 to benefits or introducing into accident 27 per cent. in 1912. It should be policies features providing benefits not noted that the increase in premium contained in policies that are now proposed to be issued even for an additional premium." All of the recommendations adopted by the standing committee were submitted in the form of an agreement to the various companies with a request that they affirm the action of the committee and sign the agreement. Among the remaining important events relating to this form of insurance should be mentioned the enact ment of standard health and accident policy laws by New York, North Carolina, Minnesota and Vermont, and the formation of the American Association of Accident Underwriters. The formal organization of this Association occurred in May, and at the meeting there were present about 50 representatives of health and accident companies from all parts of the country. income took place entirely in the surety business, the premiums from this business increasing from $7,169,793 to $10,122,266 in 1912. Fidelity premiums, on the other hand, declined from $10.435,719 in 1911 to $9,120,918 in 1912. While three companies ceased writing fidelity and surety risks during 1912 by reinsurance, at least ten new companies entered the field. Despite the increase in competition, however, no serious attempts were made to cut rates or commissions. This favor. able condition of affairs, judging from the insurance press, has not continued during 1913. Toward the close of August it was reported that one of the largest New York companies declined to withdraw certain agents who were not working in accord with the rules adopted by the Surety Association, and that another important company resigned from the Association on the ground that: "it did not care to continue in an organization which could not control its own members." Rate wars at this particular time are considered highly undesirable in the business and for this reason the uncertain future of the Association is lamented by many under. Fidelity and Surety Insurance.The data of the issue of the Insurance Year Book for 1913 show that the combined premiums received by fidelity and surety companies aggregated $19,243,184 during 1912, or an increase of $2,285,133 over 1911. The losses increased from $4,980,430 in 1911 to $5,192,456, the ratio of losses to premiums therefore declin- writers. The chief problem which confronted the business during 1912 was the increasing liability of surety companies under contractors' construction bonds in states where workmen's compensation laws have been adopted. According to Mr. DeLeon's account in the 1913 Insurance Year Book: 3,200,645 every known hazard or contingency, including claims by workmen under these compensation laws, and for the same reason the general contractor shifts this burden whenever possible upon the shoulthese laws provide in specific terms that ders of the sub-contractor. Many of the owner shall be directly liable under certain conditions for injuries sustained by employees of contractors on work let or subject, with the result that the surety bond is conditioned for the performance of all the terms of the contract, and if liability attaches to the owner for injuries to workmen this obligation must be assumed by the contractor and the surety company. These laws materially increase the liability of the contractor and therefore of the surety company, for the owner draws his contract for the construction of a building so as to provide that the contractor shall indemnify the owner against XV. POPULATION AND IMMIGRATION POPULATION WALTER F. WILLCOX Classification by Marital Condition. -During 1913 statistics showing the population of the United States in 1910, distributed by marital condition, were given to the public. A similar Federal inquiry was carried to completion for the first time in 1890, and thus we now have figures covering a 20-year period. In the total population of the United States the proportion of single has been steadily falling and that of the married steadily rising. In 20 years the single have declined from 59.3 to 55.4 per cent. of the population, or nearly four per cent. This decrease in the second decade was more rapid than in the first. Within the same period the proportion of married rose from 35.7 to 38.9 per cent. of the population. There has been a slight increase also in the proportion of widowed. Clearly married life is becoming more and more general. In casting about for an explanation of these surprising figures the first possibility which occurs is that they may be connected with the decreasing proportion of children, or, what is another way of expressing the same fact, the increasing proportion of people of marriageable age. The correctness of this guess can be tested by excluding from the comparison all who are too young to marry. For this purpose the dividing line between children and adults is usually assumed to be 15 years. The figures show, however, that among adults, as thus defined, married life has become more common and single life less common. The proportion of married among adults has risen in 20 years from 55.3 to 57.3 per cent. and that of single has fallen from 36.9 to 34.4 per cent., while the proportion of the widowed has remained about the same. Obviously the decreasing proportion of children explains a part, but only a part, of the increasing prevalence of married life. In the southern states, that is, the South-Atlantic and South-Central, the proportion of single persons in the adult population is less, and the proportion of married and of widowed is greater, than it is in the northern states, that is, the North-Atlantic and North-Central. The proportion of single in the northern states, while greater than in the southern, is noticeably less than in the Far West, and likewise the proportion of married in the northern states, while less than in the southern, is greater than in the West. In all sections, however, the proportion of single has been falling and the proportion of married rising. Changes similar to those in progress in the whole country and its main divisions were in progress also during the 20 years in nearly all of the states; for example, among the males over 15 the proportion of bachelors decreased between 1890 and 1910 in every state except Minnesota and South Dakota; and in the same period the proportion of husbands increased in every state except Rhode Island, Wisconsin, Minnesota and the Dakotas. The proportion of widowers increased in every state except North Dakota and Wyoming. The increase in the prevalence of married life and the decrease in the postponement of marriage or abstinence from it are clear and countrywide. It might be said that these figures, referring as they do to all persons over 15 years of age, do not prove that marriage is taking place 47,332,277 100.0 44,639,989 100.0 32,425,805 100.0 30,047,325 100.0 27,455,607 58.0 23,522,121 52.7 12,550,129 38.7 8,933,170 29.7 19,721,146 41.7 21,049,696 47.2 19,720,152 60.8 21,045,983 70.0 39.6 18,092,600 55.8 17,684,687 58.9 7.1 1,471,390 4.5 3,176,228 10.6 0.4 156,162 0.5 185,068 0.6 0.2 155,524 0.5 68,172 0.2 Among women the proportion who are married at ages above 25 has remained unchanged for 20 years at 71.1 per cent. The increase of marriages below that age, then, controls the movement for the total at all ages above 15. as early as formerly. In other words, | 15. These figures show in the United States as a whole no trace of a change often believed to be in progress, whereby postponement of marriage or abstinence from marriage is becoming more general. This unexpected result is the more remarkable, since during these 20 years there has been a great increase in the proportion of urban population, which in 1890 constituted 36 per cent., and in 1910, 46 per cent., of the total population. An analysis made for 1900 shows that in cities the proportion married is distinctly less than in country districts. An explanation of these figures is probably to be sought in the very rapid increase of immigration since 1890, the large number of marriages among recent immigrants and the youthful age at which they marry. If this conjecture be correct, the growth of immigration has been more than sufficient to offset any tendency that might otherwise have been detected in the general population towards a postponement of marriage or even an abstention from it. These figures show that the proportion of young men who are married has increased 5.2 per cent. in the 20 years. The proportion of girls under 20 years of age who are married has increased about two per cent. and that of young women between 20 and 24 years of age has increased about three per cent. Obviously early marriage was more common in the United States in 1910 than it was 20 years earlier. When the higher ages are studied, somewhat different results appear. Under 35 years of age the proportion of men who were married was greater in 1910 than in 1890; above As the figures showing the increase that age the proportion was less. As of married in the total population the number of men between 15 and deserve further analysis, a distinction 35 years of age is greater than the has been drawn between white and number over 35, the increasing pro-negro. Among negro adults, or perportion of married below that age con- sons over 15, the proportion of mar trols the total results at all ages over ried is the same for the two sexes. It is greater than the proportion of married among white males and less than the proportion of married among white females. The difference between white males and white females, whereby the proportion of married white women exceeds that of married white men by over three per cent., is mainly due, no doubt, to the large excess of males in the total white population of the country. In the white population over 15 years of age the men outnumber the women by two and onequarter million, or by four in every hundred. 20-29 Death Rate by Marital Condition.During the past year a study of the comparative death rates of the single, 30-39. the married and the widowed in New 40-49. York State outside of New York City 50-59 60-69. and Buffalo, has yielded interesting results. Prior to this study the only American figures upon the death rate of the single, the married and the widowed were those found in the Fed eral censuses of 1890 and 1900, and in them only three age classes, namely, 15 to 44, 45 to 64, and 65 and over, were recognized. The ratios which resulted were practically valueless because 20 or more years were included in each age class, and in any class of that size the average age of married persons would be higher than that of single persons and lower than that of widowed. To take an example from the class between 15 and 44 years of age, the average age of American bachelors is nearly 24, that of husbands nearly 34, and that of widowers nearly 36 years. Similarly the average age of spinsters in this age class is 22.4, that of wives 31.7 and that of widows 35 years. These differences in age alone, even if there were no other difference in the mortality of the single, the married and the widowed, would make the death rate of husbands 15 to 44 years of age 21 per cent., and the death rate of widowers 26 per cent., above that of bachelors in the same age period. For the same reason the death rate of wives would exceed that of spinsters by 26 per cent. and that of widows would exceed it by 34 per cent. To get an adequate measure of the relation between marriage and mortality the age classes should include no more than five years. In the New York 70-79 DEATH RATES BY MARITAL The most obvious fact indicated by these figures is that the death rate for husbands is much lower than that cept the highest, where it is about for single men at each age group ex the same. The percentage of difference is greatest at the ages 30 to 39 and 40 to 49, where the death rate of husbands is somewhat less than onehalf that of bachelors. The table also shows that the death rate of widowers and divorced men is considerably higher than that of husbands of the same age, and between the ages of 30 and 80 is not far from the death rate of bachelors of the same age. If husbands lose their wives, then, they lose much of the chance of longevity which marriage secures them; and, in general, the younger they are, the greater proportion of this chance they lose. The following table shows corresponding figures for women: AGE PERIOD 20-29 30-39 40-49 50-59. 70-79 Here again the striking fact is that the death rate of wives is lower than that of single women of the same age. The only exception is at the age period 20 to 29, perhaps due in part to the influence of child-bearing at those years upon the mortality of wives and in part to the greater average age of the wives in that group. As women marry at the average age of about 25, the average age of wives between 20 and 30 would be noticeably above the average age of spinsters in the same group. The advantage of wives over spinsters in the matter of mortality, however, is far less than that of husbands over bachelors. The mortality of widows and divorced women as a rule is higher than that of spinsters. It also exceeds that of wives, but not as much as the mortality of widowers and divorced men exceeds that of husbands. So, from the standpoint of mortality, marriage is of less benefit to women than to men. of birth of our native population for every census since the Civil War. Before that date the figures applied only to the free population, and therefore are not to be compared with the figures since emancipation. Between 1870 and 1900 the migratory tendency of the native population of the United States as thus measured was slowly diminishing. In 1870, 23.2 per cent. and in 1900 only 20.6 per cent. of the population were living outside the state of birth. But during the first decade of the twentieth century interstate migration increased with about the same rapidity as it decreased between 1870 and 1880 or 1880 and 1890, and the proportion living outside the state of birth is now nearly as great as it was in 1880. This increase in the amount of interstate migration holds true not merely of the country as a whole, but of every state west of the Mississippi River and of 11 out of the 25 states east of that river. The 14 states in which emigration of natives showed no increase between 1900 and 1910 included all the New England states, New York, New Jersey, Ohio, Virginia, North Carolina, Georgia, Alabama and Mississippi. The census figures, also, make it possible to determine what states have. had a net gain and what a net loss through interstate migration. Those with a net loss are such as have contributed more of their native population to the population of other states in the Union than they have received from the natives of other states; those with a net gain are such as have re Attention has been called to the fact that these results differ widely from those drawn from insurance statistics, showing that the death rate of insured wives is higher than that of insured spinsters. Probably both results are correct for the classes to which they apply. The conclusions for New York are corroborated by the results in every foreign country which has made a similar analysis of its vital statistics. The divergent results reached by some students of insurance companies' experience, like other similar divergencies, can best be explained as due to the influence of selection.ceived from other states more natives For example, the death rate of insured women is often higher than that of insured men of the same age, a result in striking contrast to the low-vided approximately by the Missiser death rates of women in the general population, except perhaps in some countries among women between 20 and 30 years of age. This is probably due to the fact that a large proportion of men take out insurance policies even when they believe themselves in perfect health, while among the women who apply for and receive insurance a notable proportion have reason to suspect that their health is subnormal. Interstate Migration. The census has given figures showing the state than they have contributed to the rest of the Union. Both in 1900 and in 1910 the United States was di sippi River into an eastern area, the states of which had suffered a net loss through interstate migration, and a western area, the states of which showed a net gain by interstate migration. But the division line is moving westward; in 1880 a number of states east of the Mississippi River, namely, Wisconsin, Michigan, Illinois, Indiana and Mississippi, had gained more than they had lost by interstate migration, and even in 1900 this was true of every state west of the Mississippi River; but in 1910 it was not |