and over 33 as 5 to 7 per cent. The age of most frequent marriage for non-college women is 23; for college women 25; and the mean age of marriage for the former is 24.3 years; of the latter, 26.3 years. The average number of years of married life of college-bred girls is 9.6, three years less than their sisters; two less than their cousins; and two and a half less than their friends. More than half the college women 1 A Comparison of Coeducation and Women's Colleges. Overland Mo., 1890, * Statistics of College and Non-College Women. Am. Statistical Ass'n, June, ADOLESCENT GIRLS AND THEIR EDUCATION 59 l studied were teachers and three-fourths had secured economic independence before marriage. Seventy-seven per cent of the graduates married college men as against 51 per cent of the non-graduates.
Miss Shinn 1 later studied the marriage-rate of the Associ- ation of Collegiate Alumnae comprising fifteen leading colleges. Of 1,805 enrolled in 1895, only 28 per cent were married, the rate for the country at large for women over 20 being nearly 80 per cent; she concluded that " under 25 college women rarely marry," and " that but a small proportion of them have married." Of 277 of the latest three classes but 10 were married; taking only those graduates past 25, 32.7 per cent; after 30, 43.7; after 35, 49.7; after 40, 54 per cent were married. " The ultimate probability of a college woman's marriage, therefore, seems to be below 55 per cent as against 90 per cent for. other women." Taking five-year periods, most of them marry between 25 and 30. Of all the 1,805 considered, 1,134, nearly 63 per cent, are from colleges for women alone, and of these only 25.7 were married against 32.6 of those from coeducational colleges, although the latter average 0.7 of a year older than those from the women's colleges. College women marry least of all in the North Atlantic States, 23.7 per cent as against 37 per cent of the graduates of the Middle States colleges. Many of these graduates become teachers, and for no station in life save that of a nun, we are told, is marriage so unlikely as for resi- dent teachers in a private girls' school, although this is a posi- tion mostly preferred by graduates from women's colleges, while women graduates from coeducational institutions prefer teaching in the public high schools. Miss Abbott 2 found that of 1,022 Vassar graduates 37.6 per cent taught, and suggested that the association of alumnae may be recruited largely from teachers. Her conclusions agree with those of Miss Shinn, that college women marry late and in far less ratio than others.
Bryn Mawr 3 reports that in January, 1900, of the class 1 Marriage-Rate of College Women. Century, October, 1895, p. 946.
* College Women and Matrimony again, by Frances M. Abbott, M. D. Cen- 3 Education of Women, by M. Carey Thomas. Monographs on Education in the United States, 1899, p. 36.
THE PSYCHOLOGY OF ADOLESCENCE of 1889, its oldest class, 40.7 per cent were married; of the first six classes ending with graduation in 1894, 30 per cent; of the first nine classes, 20.9 per cent; and of the first eleven classes ending in 1899, 15.2 per cent. This shows the impor- tance of time and also how exceptionally old these graduates marry. The same result is shown by the following table: Opened in Percentage of graduates married.
Minnesota Cornell ) Syracuse >.« l87Q i88c IO 4 Leland Stanford Junior l8qi Chicago...This suggests that the rate of marriage of college women is decreasing and that the age at which marriage occurs is becoming steadily later. Miss Abbott (Forum, vol. xx, p. 378) showed that of 8,956 graduates of 16 colleges, 23 per cent \vere married.
It may be wrong to infer that if a small percentage of col- lege women marry, it is the college that diverts them or that they are less desirable to men. Some who go to college desire marriage less or single careers more, so that one writer 1 is surprised that there is not less marriage among girl graduates and thinks college education actually promotes it by making many marriageable who would not be so otherwise. College girls certainly have a prolonged period of probation with di- verted interest. During their course, according to another writer, very many receive and reject propositions of marriage in order to complete their education, but of course no statistics «ire available upon this point. Yet another says2 that college 1 The Marriage of Women College Graduates. Anon. Nature, September 24, 2 College Women and Matrimony again, by Frances M. Abbott, M. D. Century, i8q6.
ADOLESCENT GIRLS AND THEIR EDUCATION 593 women as a class need less to look to marriage as a means of support, and most who must at once earn their living teach. A woman, who was earning $60 a month and resigned to marry a man earning $40, exhibited rare devotion. Whether mercenary motives are increased by the luxury and expensive tastes and interests of college; whether, having tasted the fruits of the tree of knowledge, women are not less inclined to renounce it for domestic life, can not be discussed on a basis of facts or statistics.
By corresponding with class secretaries and correlating their data with those kept at the college offices, I have gath- ered the latest and perhaps the fullest data yet published on the marriage and fecundity of female graduates for three lead- ing colleges.1 If there remain errors they are those of incom- plete returns and show rather too few marriages, births, and deaths than too many, but great effort has been taken to re- duce this error to a minimum. If such a census includes all graduates up to within a year or even two of its date, the total number of marriages rarely exceeds one-fourth, for those who wed are slow to do so. The last ten classes to graduate, if taken by themselves, would show a yet far smaller proportion married. From Vassar there were 323 graduates in the first ten classes, 1867-' 76, of whom, in the spring of 1903. 179, or 55.4 per cent, had married. In the next ten classes there were 378 graduates, of whom 192 were married, 50.7 per cent. In the third ten classes, 1 887^96, were 603 graduates, of whom 169, or 28 per cent, were married. In Smith College, during the first ten years ending with the class of 1888, there were 370 graduates, of whom, in the spring of 1903, 158, or 42.7 per cent, were married. Of the next ten classes, including that of 1898, there were 1,130 graduates, of whom 331, or 28.3 per cent, were married. In Wellesley,. during the first ten years ending with the class of 1888, there were 436 grad- uates, of whom, in the spring of 1903, 203, or 46.5 per cent, were married. In the next ten classes, ending with the class of 1898, there were 1,162 graduates, of whom 296, or 25.4 per 1 See for details the study of Dr. Theodate Smith and myself on Marriage and Fecundity of College Men and Women. Fed. Sem., September, 1903, vol. x, pp.
594 THE PSYCHOLOGY OF ADOLESCENCE cent were married. From these data it appears that not very far from one-half of the graduates of all these colleges who marry, or about a fourth of all, do so after they have been out of college from ten to fifteen years, at an age of at least thirty, and that a large part of the other fourth who marry do so as late as forty. The results of this late age upon fecun- dity is, as we shall see later, about what might be inferred.
The tu quoque retort which many women love to make to men has too much basis, although the facts show results only about half as bad. Mr. Deming's admirable figures for Yale show 21.6 per cent of the Yale classes from i86i-'79 as unmarried, and the Harvard record from i87o-'79 is 26.5 per cent. President Eliot's figures for the Harvard classes of 1872— '77, although we find them somewhat unfavorably erroneous, show 28 per cent, mostly now between forty and fifty years of age, unmarried. This proportion has greatly increased, for at Yale only about two per cent of the men of corresponding age in the last century were unmarried. Then, also, men married earlier, and it was customary for clergymen, who constituted a far larger proportion of graduates than now, to marry immediately after ordination, the average age of graduation being twenty-one. Indeed, this was necessary, for domestic service was rare, especially in rural districts, and a wife was necessary to do the work of housekeeping. " The average age of marriage for women was under twenty-one, many marrying in the teens, and several marriages at the age of fourteen are recorded." 1 " Remarriage was almost uni- versal for both men and women, economic conditions rendering it a necessity." " Forty per cent of the wives of Yale graduates from 1 701-^05 did not live to bring up their children, and it took a second, and frequently in large families a third woman, to complete the.work. The problem of superfluous women did not exist in those days. They were all needed to bring up another woman's children." Of the wives of Harvard grad- uates between the years 1 658-^0 37.3 per cent died under the age of forty-five.
V. Fecundity of Educated Women. — Here the matter is worse yet for educated women. Natality rates show the 1 Miss Smith: Op. cit., p. 280.
ADOLESCENT GIRLS AND THEIR EDUCATION 595 springs of national growth or decay, and only the constant influx of foreigners prevents us from realizing the ominous forebodings with which France is facing the problem of a steady decadence of birth-rates, which prompted and circu- lated such a book as Zola's Fruitfulness. In a significant paper by Dr. Allen,1 on the New England family, which was the germ of American civilization, and where for two hundred years the homes were well-nigh models, it is shown how the birth-rate has steadily declined for half a century and that at a very rapid rate until it is lower than that of any Euro- pean nation, France itself not excepted. In 1875, there were 359,000 families in Massachusetts; of these, 23,739 consisted of only one person, 115,456 of only two, and 140,974 of only three persons. Dr. Allen estimated that only one-half of the New England mothers could properly nurse their offspring, and that the number who could do so was constantly decreas- ing. While failure to do so might be often due to lack of wish, it was usually due to undeveloped mammary glands, feeble digestion, and nervousness. This state of things, he assures us, can be found to anything like the same extent no- where else and among no other nation or race in history.
Foreign families, especially if they acquire property, approach this condition a few years after they land on our shores. The Jews and our grandmothers thought barrenness a curse, but now the bearing and rearing of large families is felt to belong to low life. Love of offspring is less intense; woman's organi- zation is changing under new conditions. Housekeeping, es- pecially as a vocation which used to be one of the most hygi- enic and ennobling occupations for body and mind, is despised and evaded, and the influence of home is diminishing. Modes of life are artificial and too expensive, so that marriages are later as wrell as fewer, the death-rate of infants among old New England families is increasing, and so are abortions and di- vorces. Invalids make poor home-builders, poor husbands, wives, and worse fathers and mothers. From 1886 to 1891, both inclusive, 103,733 children were born in families where both parents were natives; 104,884 where both were foreign; 1 The New England Family, by Nathan Allen, M. D. New England Magazine, 596 THE PSYCHOLOGY OF ADOLESCENCE 39,292 in families where one parent was native and the other foreign. During the same years 156,225 native-born inhab- itants of Massachusetts and 40,716 foreign born died. The deaths of native Americans exceeded the deaths of the foreign born by 29,796, which, adding half the deaths of unknown or mixed parentage among the foreign of the State, make the births exceed the deaths by 87,824. The foreign-born inhabi- tants of the State, including less than one- fourth, give birth to more than one-half of the children.1 First among the causes of the decline, Dr. Ellis ~ places physical and mental inability to bear and care for children, at the proper period, and sec- ondly, he places unwillingness to sacrifice ease, freedom, and enjoyment involved in parenthood, the disposition to put pleasure in place of duty, the effeminacy of wealth, the new woman movement, and foeticide, and he pleads for domestic labor as one of the best correctives. Comparing the forty years ending 1890, the native marriages average 2.3 children each, while those of the foreign born average 7.4 each.
Sterility 3 is of all degrees, from total up to inability to produce a goodly number of children who mature well through adolescence and can themselves produce healthy off- spring. It is vastly more complex in woman than in man, and, according to the averages of statistics from many sources, Duncan concludes, is due to the wife about six times as often as to the husband. In Great Britain, one of the most prolific of all lands, about one-tenth of all marriages are now unfruitful. The relative sterility of the one-child system occurs in Eng- land once in about thirteen fruitful marriages, and those with- out offspring sixteen months after marriage are beginning to be sterile, as also are those who do not at their best period bear a child every twenty months. The average age of cessa- tion in Great Britain is thirty-eight years, and a woman who begins to bear children at the age of twenty to twenty-five should continue for ten or fifteen years. Fertile marriages in England average, in fact, about six living children each.
1 Vital Statistics of Massachusetts from 1856 to 1895, by S. W. Abbott.
* Deterioration of Puritan Stock and its Causes, by Dr. Ellis. New York, 1894. Published by the author.
3 Sterility in Woman. Gulstonian Lectures, by J. Matthew Duncan. London, ADOLESCENT GIRLS AND THEIR EDUCATION 597 The first child is a source of danger to the mother, about one- fifteenth of whom die, while only one in forty-eight die at the second birth.
The chief cause of all degrees of unfertility is, according to Duncan, overnutrition, and this is true down through the animal and even in the plant world. Overfeeding or fat pro- duction in the female is not only unfavorable to fertility, but illustrates how undue emphasis laid upon individuation is antithetical to generation. The relative childlessness of heir- esses is a case in point. Excessive sexual indulgence and ex- citement are potent deterrents. Excess of desire and pleasure are often compensated by defect in reproductive energy, and so are too early marriages. " Women married under twenty years of age have much more sterility than women married from twenty to twenty-four, and the sterility of marriage be- fore twenty is less1 than that after twenty- four, and from this point it gradually increases with the age of marriage." Pre- mature and postmature women, too, have smaller children. Sterility is not a specific disease, but is the intricate product of causes as complex as modern civilization. While the rapid progress of gynecology has shed floods of light upon it, its larger problems are yet very far from solution.
Duncan had previously found 1 that the mortality of chil- dren was less if the mother's age was twenty-one to twen- ty-five than at any other quinquad; that the age of least puerperal mortality was twenty-five, and increased above and below; that the number of twins reached its maximum be- tween twenty-five and twenty-nine; that children of mothers in these years were heaviest at birth and grew tallest; that the greatest viable fecundity was at twenty-six, the fathers being thirty-one and three-tenths; that although the shortest interval between births was when the mother was twenty-six, marriage when she was below twenty-five was more fertile than after, and that the quinquad of greatest fecundity was twenty to twenty-five. In his Republic, Plato said men should bear children from thirty to fifty-five and women from twenty to forty, while Aristotle thought the man should marry at thirty-seven and the woman at eighteen. Duncan said that •Fecundity, 'Fertility, Sterility, etc., 1877.
598 THE PSYCHOLOGY OF ADOLESCENCE while thirty-three women attain the age of eighteen, only twenty-eight men live to be thirty-seven; that in nine meas- urements the womb grew after twenty (although Hecker thought the number of pregnancies was a factor) and con- tinued to grow in length till near the end of the fertile period; that the mortality of first confinements was twice that of all subsequent ones, and that mothers died least often in second confinement between twenty and twenty-five. Many who marry before twenty have no children, but if only those who do so are included, the years from sixteen to twenty are most fertile.
According to the very careful statistics compiled by Or- schansky,1 the rate of productivity or rapidity of births is greatest in women married at eighteen, and decreases gen- erally afterward. The stature of both boys and girls at birth is greatest when the mother's age is twenty-seven to twenty- eight, and also with the fifth or sixth child born, as most pelvic diameters continue to grow to this age. The productive period ends earlier with mothers who begin to bear children young.
Korosi,2 who tabulated data for 71,800 married couples, found that the maximal natality of mothers began between the ages of eighteen and nineteen, when the annual probability of birth was 44.6 and 42.2 respectively. This means that nearly one wife out of two will give birth to a child within a year from marriage. From twenty this probability declines, and at twenty-five is only 31.5 per cent; at thirty it is 24; at thirty-five it is 16; and at forty, 8 per cent. Comparable sta- tistics for Berlin and Sweden, Edinburgh, Glasgow, Norway, and Denmark, resembled closely those of Budapest. With fathers, the highest natality is at twenty-five, when it is 38.9 per cent; at thirty it is 31.7; at thirty-five, 23; and at forty, 15 per cent. We may bet, says Korosi, 70 to i against a child being born of a mother aged forty-five, and 7 to i against a father having a child during his fortieth year. Bigenous natality, i. e., where the age of both parents is considered, gives results quite different from these monogenous tables.
1 Etude sur 1'here'clite. St. Petersburg, 1894.
8 An Estimate of the Degrees of Legitimate Natality, etc. Phil. Trans, of the Royal Soc., B. Part II, 1895, pp. 781-875.
ADOLESCENT GIRLS AND THEIR EDUCATION 599 By making the complex combination necessary, it was found that for every year between twenty and forty, mothers ac- quired the highest natality possible to their age with fathers under thirty; while the latter reach their maximal natality with women under twenty, so that men between thirty-five and forty-five should choose wives between twenty and twen- ty-five, and those of forty-five to fifty, wives of twenty-five to thirty. His data also shows the interesting result that women under twenty-eight should choose older, and those above twen- ty-eight, younger husbands than themselves, while men at all ages have the greatest chance of offspring with wives younger than themselves. The most prolific marriages are when the woman is eighteen to twenty and the husband twenty-four to twenty-six. In a separate research Korosi showed that if healthy children, or those who lived to mature, be considered, the age of greatest fecundity fell about a year later.1 If the father's age goes up, that of the mother declines, but not quite in the same degree, i. e., if the father is thirty-one, the mother should be thirty-one, total, 62; if he is forty, she should be twenty-four, total, 64. These lines plotted, Galton calls iso- gens.2 If these are based on sufficient data, we can calculate the plus or minus age distance of the husband from his wife at each year which is likely to result in most children.3 The following table combines two of Korosi's, based on 10,000 marriages, showing the average fecundity of both sexes for different ages, being reduced at thirty to nearly half its initial rate at nineteen for women, and falling off one-half for men from twenty-seven to forty.4 1 Ueber den Einfluss des Elterlichen Alters auf die Lebenskraft der Kinder. Jahrbuch f. Nat. Oekon. u. Statistik. Third ser., vol. iv, p. 518. * Proc. Royal Soc., vol. Iv, p. 18 et seq. 8 Maurel,4 in an interesting study of the conditions which modify masculinity, 4 Rev. Scientifique, April 4, 1903.
600 THE PSYCHOLOGY OF ADOLESCENCE Compiling from the census reports a table regarding the decreas- ing average in the size of families in the several sections of the country we have the following: AVERAGE SIZE OF FAMILY i860 IQOO New England 4.8 4.6 South Atlantic States 5.2 5.0 Illinois 5.2 4.7 Michigan 4.9 4.4 Wisconsin 5.2 4.9 Minnesota 5.2 5.1 Missouri 5.4 4.7 North Dakota 4.3 4.9 South Dakota 4.3 4.8 Nebraska 5.1 4.8 Kansas 5.0 4.6 According to this, New England does not show either the largest decrease in ten years nor the smallest average size of families. The decrease in New England is but two-tenths of one per cent. In New York it is five-tenths of one per cent; in Pennsylvania, three-tenths; in Ohio, six-tenths; in Indiana, seven-tenths; in Illinois, five-tenths; in Michigan, five-tenths; in Wisconsin, three-tenths; in Iowa, six- tenths; in Missouri, seven-tenths; in Kansas, four-tenths. The aver- age size of families in New England is larger than in New York, Ohio, Indiana, and Michigan, and equal to that in Iowa and Kansas. But this reckoning includes all races, and has no value with regard to rela- tive size of the families of long establishment in the country as com- pared with those of later immigration.
finds it to be augmented by the youth of the parents, especially the father, when conception takes place nearest the period of marriage, also by the complete matur- ity of the ovum and perhaps also of the male element, and the vigor of the parents The higher classes develop less masculinity in general than the lower. He also finds that the proportion of male to female births, which in Europe and India is greater in about the same degree, does not hold of the African races. Reduced masculinity means real social deficit, and is, therefore, one of the best criteria of the vigor of a class of the population. Emigration, which in general favors natality, diminishes masculinity, while crossing among the Neo- Latin races and precocious marriages increase it, and syphilis, alcoholism, and arthritism diminish it. He holds that femininity is a product of enfeebled fecundation, and that paternity is a more important fact in sex determination than maternity.
ADOLESCENT GIRLS AND THEIR EDUCATION 60 1 R. R. Kuczynski * shows that in the period from 1883 to 1897 the general marriage-rate of the foreign-born in this country was three- fourths higher than that of the native-born. From 1887 to 1889 the native-born in Massachusetts have the smallest general marriage-rate. In the three censuses of 1885, 1890, and 1895, the proportion of the married among the natives was less than one-third, while the proportion among the foreign-born exceeded one-half. The native population of Massachusetts has a special birth-rate of 6.3 births for one hundred adult women in child-bearing age, whereas in Berlin the annual rate of ten for every hundred women was proven to be not quite sufficient to keep up its population, so that this State is dying at a rapid pace. The number of adult native men living in wedlock in 1885 to 1895 was three-fifths, of foreign-born, two-thirds. The proportion of native adult women living in wedlock was six-thirteenths; of foreign-born, seven-thirteenths. In 1895 the number of women who were married but childless was one-fifth among the natives, and two-fifteenths among the foreign-born.;' The average num- ber of children born to every foreign-born married woman was two-thirds higher than for the natives, viz., four-sevenths higher among German women, six-sevenths among Irish, twice as high among French Canadian women." From 1883 to 1897 the special birth-rate of foreign-born adult women was more than twice as high as for the natives.
Engelmann,2 from 1,700 cases, found that a little over twenty per cent of married women in America are childless, although his definition of sterility was the condition of those married three years without offspring. (Simpson's standard rate of sterility is eleven per cent.) Thus, he concludes that " the extremes of sterility are reached in this country." Among the laboring classes in St. Louis he found 2.1 children per married couple, in Boston he found it 1.7; in Michigan, 1.8 in recent years, but for the twenty-five years ending in 1895, 2.1. Mrs. Smith finds that among college 1 The Fecundity of the Native and Foreign Born Population in Massachusetts. The Quart. Jour, ot Economics, November, 1901, and February, 1902.
1 The Increasing Sterility of American Women. Jour, of Am. Med. Ass'n, October, 1901. 77 602 THE PSYCHOLOGY OF ADOLESCENCE