selves gradually to new needs. Specific traits will of course always be estimated differently by different individuals; but all the traits idealized should possess a survival value. However, racial values and survival values do not absolutely coincide since the range of traits adapted to family life is somewhat narrower than the traits available for social life. In other words, many individuals not suited to family life may perform very useful social functions. Individuals selected for marriage should ordinarily possess characteristics suited for the two purposes of marriage, normal life of the sexes in common and parentage. Furthermore the sets of traits essential to these two purposes are not absolutely identical. The first purpose of marriage requires that individuals should be pleasing to the opposite sex, and the second requires that they should possess a good heritage and also strong parental instincts. It is evident that many persons may lack one or another of these characteristics without being incapacitated for social duties.

Extent of Sexual Selection. It is impossible to estimate accurately the number of persons in any society who will never marry; but the numbers reported as single or married at various ages are of significance and will suggest a sufficiently close approximation. Of course there is no age period which accurately divides the permanently single from those who will marry; but the age of thirty or perhaps thirty-five, would approach nearest to such a division. In 1910, of the population 35 years of age and over, 12.2% of males and 9.9% of females were reported as single. And of the population 30 years of age and over, 14.9% of males and 10.5% of females were single. The latter period gives an appreciably larger percentage of males but only a slightly larger percentage of females than age 35. After the age of 30 for women and 32 for men, the proportion of single does not vary greatly, though it declines rapidly for adults up to those ages.

Of course many persons over 30 years of age, especially males, will marry; but probably that number is more than offset by the number of persons under 30 years of age who will never marry. It seems probable, therefore, that the proportion of single over 30 years of age would be an underestimate rather than an overestimate of the number who will never marry. In the older section of the country the proportion is still higher. For example, in Massachusetts, of the native born of native parentage 35 years of age and over, 12.9% of males and 17.5% of females are single; and in the city of Boston 19.6% of males and 23.9% of females of the same class in the population are single. The extent of sexual selection therefore varies appreciably with social conditions. For the country as a whole it is no exaggeration to say that at least 10% of the women never marry. And in the Eastern states where competition is keener, and where native women are more numerous than men, the number remaining permanently single may rise to 25%. Mr. Robert J. Sprague estimates that about 15% of native women do not marry during the child-bearing period. In European countries the number is still higher. It is said that in England 25% of the adult women never marry. And Pearson makes a rough calculation that from 20 to 30% of women in other European countries never marry.

Thus while it is evident that sexual selection is a strong force in the population as a whole, its results fall even more heavily on the superior than on the inferior social classes. Bertillon gives the marriage rate in three European cities according to economic status, showing that it varies quite uniformly with wealth.1

TABLE I

MARRIAGE RATES PER 1,000 MEN OVER 20 AND WOMEN OVER 15 YEARS

OF AGE

Paris Berlin Vienna Character of District

1886-95 1886–95 1891-07

Men Women Very poor..

29.1 44.0

90.I

67.0 Poor...

27.9 44.4

80.6 52.7 Well off..

24.7 36.3

84.0

48.9 Very well off.

24.5 26.5

71.6 40.7 Rich..

21.0 26.0 56.6 Very rich....

21.1 20.5 43.4 19.1 In the United States a number of studies have been made of the marriage and birth rates of college graduates, and all indicate that a much larger proportion of college graduates remain

28.7

single than the average of the population. The results of some of these studies are given below.

The indication is that about one-quarter of the men and from one-third to one-half of the women never marry, though some of the figures include recent classes so that the proportion of married will still increase to some extent.

According to a study made in 1895) by Miss M. Shinn,2 more women graduating from coeducational institutions marry than those graduating from women's colleges. And a fact of equal significance appearing from this study is that women from coeducational colleges marry at an earlier age.

From such estimates as these one is forced to conclude not only that sexual selection is an active force in modern society, but that it is excessive for the mere purpose of eliminating the unfit; and, therefore, that from the purely racial point of view the marriage rate might advantageously be raised. However, the mere intensity of the selective process would not present a serious problem, if it always acted with precision and uniformly eliminated those least fitted for married life. Although no exact estimate of the accuracy of sexual selection can be made, casual observation is sufficient to convince one that the selective process is extremely faulty from the point of view of social utility. While it may be true that the majority of the worst do not marry and a majority of the best do marry, so that the average of the married is superior in quality to the average of the unmarried, it must be admitted nevertheless that many persons entirely unfitted for family life marry and become parents, and many persons who possess to a reasonable degree at least all the traits requisite for marriage remain single. Although the results of sexual selection cannot be entirely condemned, there is much room for improvement. And the perfection of the operation of this important form of selection would surely be a great stride in progress.

Causes of the Defects of Sexual Selection. No restraint upon the unfit. A variety of social conditions interfere with the proper action of sexual selection. For instance, social stratification resulting in the partial segregation of the degenerate and their consequent intimate association is perhaps the chief cause of the marriage of the unfit. Although the defective and the degenerate are usually rejected by the normal, inasmuch as they are not eliminated, nothing prevents them from selecting one another. In the present form of social organization there seems to be no way of avoiding this evil except by compulsion. Society must either restrain the individuals or forbid their marriages.

A narrow range of choice. Perhaps the chief obstacle to the marriage of the normal in society is restricted range of choice. In this discussion the unmarried have been referred to as those who are eliminated or rejected through the selective process; but these terms are to be interpreted in the technical sense only. It is quite undeniable that many of the unmarried are rejectors rather than the rejected; and the reason often is that their circle of acquaintance is too limited to bring them into contact with persons of the opposite sex who fulfill their ideals. In order that sexual selection should have a sufficient range of action the circle of acquaintance should be broad enough to include at least a representative number of the opposite sex. A more restricted range of choice might result in the elimination of the best.

Restrictions to sexual selection may be physical, as is the case in rural districts where the population is so sparse that few eligible persons are within the range of ordinary acquaintanceship. Or the restrictions may be social, as often happens in cities where the population though physically dense is socially isolated, because facilities for social cohesion are scarce. Young people from rural districts migrate to cities and lead lives quite as devoid of normal companionship as is life on the farm. Cities are often conceived of as aggregations, composed of all kinds and conditions of people associating more or less intimately and amalgamated into a single social unit. Therefore the city is supposed to offer limitless opportunity for association and companionship of individuals of like kind.

On the contrary cities are composed of an indefinite number of detached social groups, the members of which have little knowledge of, or sympathy with, the life of the other groups. And these groups are in turn composed of smaller circles or of individuals whose social opportunities vary greatly, some having fairly wide ranges of acquaintance and others lacking normal companionship, especially with those of the opposite sex. In comparatively few of these groups do individuals lead complete and normal social lives. Under such circumstances it is not surprising that there should be little opportunity for marriage for migrants to cities, even though many represent excellent stock.

It is of course true that marriage among the migrant class is often complicated by the question of income. But this factor need not be discussed here, for it involves the purely economic problem of wages. It is evident enough that low wages act as a deterrent to marriage, but with the majority of the unmarried in cities the economic is not the only obstacle encountered.

Although the problem of desirable marriage is of the utmost