¹ Data for Negroes, as distinct from those for all non-whites, are not available for all dates and geographic units used and therefore, to simplify the presentation, all figures are for nonwhites. At the national level, Negroes have constituted some 95 per cent of the non-white labor force throughout the period under study, and the two terms can accordingly be used almost interchangeably for present purposes.

² Throughout the following, male and female non-white workers are discussed together. The changes in distribution of non-whites are sufficiently similar for males and females to enable us to do this and to attempt to discuss them separately would cause unnecessary repetition.

³ The nativity status of white persons in the labor force is available only for the country as a whole in 1950, and is a result of the tabulation of a sample of the 1950 Population Census, so that the figures are not precisely comparable with other 1950 statistics used in the present paper. This sample tabulation indicates that there was a decline between 1920 and 1950 of well over two and a half million in the number of white males in the labor force who were foreign born, from 6,628,000 in 1920 to something under 4,000,000 in 1950.

⁴ In the context here the phrase “the question of who took the place of the immigrant” is not meant literally; that is, the concern is not widi substitution occupation by occupation but only with increases in the total; labor force'. In defense of this position, the great changes in the occupational and industrial structure of the labor force between 1890 and 1950 may be cited as evidence that, in fact, it would have been disastrous if the 1920-1950 components of labor force growth had been able to substitute for the 1890-1920 components only on a direct occupation-by-occupatipn basis: the labor market would have been flooded witfi unwanted manual laborers, for example, and been in desperate straits for clerical help. (Cf. Bureau, U. S. Census, Occupational Trends in the United States: 1900-1950, Bureau of the Census Working Paper No. 5, Washington, D. C., 1958Google Scholar, Table 2,.in which the proportion of the economically active in clerical and kindred occupations shows a steady incerase from 3.0 per cent in 1900 to 12.3 per cent in 1950, while the proportion of laborers, except farm and mine, shows a decline from 12.5 to 6.6 per cent over the same period.)

⁵ Non-white gains for the District of Columbia were substantially greater than those for white females when 1920-1950 is compared with 1890-1920, but the peculiar situation in the District is not relevant to the point being discussed here.