22 THE AMERICAN JOURNAL OF SOCIOLOGY
II. Effects of action by the state and by public institutions.
f) Habitual criminals. Public opinion is beginning to regard with favor the project of a prolonged segregation of habitual criminals, for the purpose of restricting their opportunities for (i) continuing their depredations, and (2) producing low-class offspring. The inquiries spoken of above (see c) will measure the importance of the latter object.
g) Feeble-minded. Aid given to institutions for the feeble-minded are open to the suspicions that they may eventually promote their marriage and the production of offspring like themselves. Inquiries are needed to test the truth of this suspicion.
/) Grants toward higher education. Money spent in the higher educa- tion of those who are intellectually unable to profit by it lessens the sum available for those who can do so. It might be expected that aid systemati- cally given on a large scale to the more capable would have considerable eugenic effect, but the subject is complex and needs investigation.
i) Indiscriminate charity, including outdoor relief. There is good rea- son to believe that the effects of indiscriminate charity are notably non- eugenic. This topic affords a wide field for inquiry.
III. Other influences that further or restrain particular classes of marriage. The instances are numerous in recent times in which social influences have restrained or furthered freedom of marriage. A judicious selection of these would be useful, and might be undertaken as time admits. I have myself just communicated to the Sociological Society a memoir entitled " Restrictions in Marriage," in which remarkable instances are given of the dominant power of religion, law, and custom. This will suggest the sort of work now in view, where less powerful influences have produced statistical effects of appreciable amount.
IV. Heredity. The facts, after being collected, are to be discussed, for improving our knowledge of the laws both of actuarial and of physiological heredity, the recent methods of advanced statistics being of course used. It is possible that a study of the effect on the offspring of differences in the parental qualities may prove important.
It is to be considered whether a study of Eurasians that is, of the descendants of Hindoo and English parents might not be advocated in proper quarters, both on its own merits as a topic of national importance and as a test of the applicability of the Mendelian hypotheses to men. Eurasians have by this time
