-C HANGES in both the density and the distribution of the civilian population have been recorded in most of the United States since the outbreak of World War II.' Between April I, 1940, and March I, 1943, thirty-six states lost population. North Dakota, with a decrease of I5.6 per cent, experienced the largest percentage loss, and four othersVermont, South Dakota, Montana, and Idaho-lost more than io per cent of their prewar population. The twelve states that gained population during this period were Washington, Oregon, California, Arizona, Nevada, and Utah west of the Rocky Mountains; Michigan in the interior of the country; and Connecticut, Delaware, Maryland, Virginia, and Florida along the Atlantic coast (Fig. 7). These population changes present problems of growing concern. Will the areas that have gained population since 1940 keep their gains at the close of the war, or will people be forced to migrate from them in order to make a living? Can the areas that have lost population since the outbreak of the war be repopulated, or will people continue to move from them in the postwar period? Geographers, because of their ability to judge the capacity of regions to support people, are now in a position to contribute to the solution of these problems.2 This paper sets the stage for work along such lines by suggesting a method of population analysis, and by submitting an examination of population patterns brought into focus by that method.