This report consolidates the data available from various sources in an attempt to assess the net U.S. balance in the international interchange of high-level manpower. The United States is shown to incur short-run losses in educational costs but long-run gains in the numerical supply of trained manpower. Nevertheless, the proportionate contribution of foreign stock to the high-level manpower pool of the United States is small, and the consequences of manpower losses to other nations is largely unknown. Current needs are for more centralized and complete information on the extent of the migration of talent to and from all countries of the world, including the United States, together with an assessment of the personal, societal, and scientific consequences of these manpower flows. With the current gaps in knowledge regarding international migration, and with the wide variation in national needs and level of development from country to country, the instituting of statutory restrictions on manpower flows by the United States would at this time appear inadvisable. I mmigration has been an important source of high-level manpower throughout the history of the United States. In recent years, the number of immigrants reported as professional, technical, and kindred workers has been increasing rapidly. In the half century since 1900, an average of more than 8,000 professional, technical, and kindred workers were admitted annually; in the decade 1951-1960, an average of 18,000 were admitted annually; and in recent years the number has increased from 22,000 in 1960 to 29,000 in 1965.' In 1960, as a result of this continued influx of immigrants, there were 287,000 foreign-born male and 129,000 foreignborn female professional, technical, and kindred workers in the experienced civilian labor force. This represents 6.8 percent of the total male * This research was partially supported by grants from the Carnegie Corporation and the Russell Sage Foundation to the Commission on Human Resources and Advanced Education. Appreciation is expressed to Lindsey R. Harmon and Herbert Soldz of the National Academy of Sciences-National Research Council for providing several of the tabulations herein reported, and to Helen S. Austin and John K. Folger for a number of helpful suggestions for revision of an earlier draft of this paper. 1 Computed from U.S. Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1957 (Washington, D. C.: Government Printing Office, 1960), p. 60; U.S. Bureau of the Census, Statistical Abstract of the United States (Washington, D. C.: Government Printing Office, 1966), p. 95. This content downloaded from 207.46.13.134 on Sun, 25 Dec 2016 06:46:35 UTC All use subject to http://about.jstor.org/terms