DATA AVAILABLE FOR EUGENICS RESEARCH 155 



death rates. From these life-table values we may derive three convenient 

 yardsticks of survival, corresponding to types of natality data already men- 

 tioned: 



1. The birth rate which would tend to yield a stationary population at 

 given mortality conditions. This result is secured simply by dividing the 

 integer (2) by the expectation of life at birth. Stationary birth rates per 

 1,000 women aged 15-44 can also be determined. 



2. The ratio of children under 5 per women of specified ages, e.g., 15-^4, 

 which would be just sufficient to supply permanent replacement can also 

 be determined from the age distribution of the theoretical stationary popu- 

 lation. The ratio of the observed ratio of children to women to the estimated 

 replacement value gives a convenient index of net reproduction per genera- 

 tion. 



3. Numbers oj children per family necessary to supply family replacement 

 can also be worked out from life-table values. Some figures of this sort, 

 estimated in relation to different sets of variables, are presented in the text 

 of a forthcoming study by Frederick Osborn and the speaker, Dynamics of 

 Population (to be published by the Macmillan Company, 1934). 



It is possible to secure quantitative, albeit imperfect data on distribution 

 of various characteristics, including intelligence, in different groups. It is 

 also possible to secure quantitative indices of reproduction trends. Simply 

 by putting these two sets of data together we may get some indication of 

 what is happening to the distribution of such traits, so far as reproduction 

 is concerned. 



We have tried such an experiment, using data on distribution of intelli- 

 gence quotients in school children in New York State villages, classified 

 according to occupations of fathers (Haggerty and Nash data) combined 

 with reproduction indices of similar occupational classes derived from U. S. 

 Birth Statistics, 1928. The results of this experiment are shown on one of 

 the charts in the exhibit. 



Other data are available which show variations in natality in relation to 

 variations in intelligence within occupational classes, and in individual 

 families. There is no large, constant differential between ethnic groups in 

 this country. Immigrant mothers, as we all know, have very large families, 

 but their daughters do not seem to follow their example, to any very great 

 extent. And it is an open question whether whites or Negroes are at present 

 reproducing more rapidly. (The present speaker is inclined, on the basis 

 of new census data, to think that at present the Negroes have a slight ad- 

 vantage.) Moreover, it is difficult, in view of all the environmental com- 

 plications, to know how much significance to attach to apparent differences 

 in intellectual development between different ethnic groups. 



