¹ For examples of recent inquiries, see Lubell, Samuel, “Who Votes Isolationist and Why”, Harper's, Vol. 202, pp. 29–36 (04, 1951)Google Scholar, and The Future of American Politics (New York, 1952), pp. 129–57Google Scholar; Survey Research Center, America's Role in World Affairs (Ann Arbor, 1952)Google Scholar; Grassmuck, George L., Sectional Biases in Congress on Foreign Policy, (Baltimore, 1951), pp. 133–74Google Scholar; Almond, Gabriel, The American People and Foreign Policy (New York, 1951)Google Scholar; Bailey, Thomas, The Man in the Street (New York, 1948), pp. 238–55Google Scholar; Langer, William and Gleason, S. Everett, The Challenge to Isolation (New York, 1952)Google Scholar; Williams, Frederick, “Regional Attitudes on International Cooperation”, Public Opinion Quarterly, Vol. 9, pp. 38–50 (Spring, 1945)CrossRefGoogle Scholar.

² See Regionalism in America, ed. Jensen, Merrill (Madison, 1951)Google Scholar. Though this entire collection of essays is relevant to the question of regional definitions, see particularly Carstensen, Vernon, “The Development and Application of Regional-Sectional Concepts, 1900–1950”, pp. 99–118Google Scholar.

³ Odum, H., Southern Regions of the United States (Chapel Hill, 1936)Google Scholar, Ch. 3. Also, Odum, H. and Moore, H., American Regionalism (New York, 1938)Google Scholar.

⁴ The Census Bureau's geographic divisions are most frequently used as the basis for regional designations because census data are so commonly used by regionalists.

⁵ Ewing, Cortez, Presidential Elections (Norman, Okla., 1940), p. 9Google Scholar. Ewing defines a seven state Midwest: Minnesota, Iowa, Wisconsin, Illinois, Indiana, Michigan and Ohio.

⁶ Fenneman, Nevin M., “Physiographic Divisions of the United States”, Annals of the Association of American Geographers, Vol. 6, pp. 19–98 (1916)CrossRefGoogle Scholar.

⁷ See his The Political Parties of Today (New York, 1924), pp. 105–9Google Scholar; The Middle Classes in American Politics (Cambridge, Mass., 1940), pp. 78–85Google Scholar; “Present Day Characteristics of American Political Parties”, in The American Political Scene, ed. Logan, E. B. (New York, 1938)Google Scholar, Ch. 1; Our More Perfect Union (Cambridge, Mass., 1950), p. 71Google Scholar. In 1924, Holcombe defined the Midwest as a combination of eleven states: Ohio, Indiana, Illinois, Michigan, Wisconsin, Minnesota, Iowa, Kansas, Nebraska, and the Dakotas. A different regional pattern was described in 1938 and 1940. In 1950, basing his delineation of regions on a new, non-political basis, Holcombe described a third regional pattern.

⁸ National Resources Committee, Regional Factors in National Planning (Washington, 1935), p. 145Google Scholar. Thirty-four regional classifications, based on such varied criteria as retail shopping areas, soils, politics, and farm income, were presented in the Committee's report.

⁹ The Works of John Adams, ed. Adams, C. P. (Boston, 1853), Vol. 8, p. 37Google Scholar.

¹⁰ The Writings of Thomas Jefferson, ed. Washington, H. A. (New York, 1861), Vol. 8, p. 4Google Scholar.

¹¹ For example, in the Chicago Tribune, Feb. 1, 1951, a map of American defense zones and an accompanying editorial indicate that the United States should defend Spain, the British Isles, Africa, Iceland, and areas in the Far East and Southeast Asia as well as other points within the “Western Hemisphere Zone of Defense.”

¹² For examples of definitions used by others writing on the subject, see Weinberg, Albert K., “The Historical Meaning of the American Doctrine of Isolation”, this Review, Vol. 34, pp. 539–47 (06, 1940)Google Scholar; Rogers, William C., “Isolationist Propaganda (September, 1939—December 7, 1941)”, unpubl. diss. (U. of Chicago, 1943), pp. 5–6Google Scholar; Beard, Charles A., American Foreign Policy in the Making (New Haven, 1946), p. 17Google Scholar; Buell, Raymond L., Isolated America (New York, 1940), p. 66Google Scholar; Leopold, Richard W., “The Mississippi Valley and American Foreign Policy, 1890–1941: An Assessment and an Appeal”, The Mississippi Valley Historical Review, Vol. 37, pp. 625–42 (03, 1951)CrossRefGoogle Scholar; Peffer, Nathaniel, America's Place in the World (New York, 1945), pp. 29–31Google Scholar; Rippy, J. F. and Debo, A., The Historical Background of the American Policy of Isolation, in Smith College Studies in History, Vol. 9, p. 72 (1924)Google Scholar. See also the Survey Research Center, Ameria's Role (cited above, n. 1), for a definition similar to the one used here.

¹³ Statistics on socio-economic characteristics of the population, unless otherwise stipulated, were obtained from the Census of 1940.

¹⁴ Twenty-three districts with high “percentage isolationist” rankings were selected for this and other similar comparisons. These districts were located in Illinois, Indiana, Kansas, Kentucky, Michigan, Minnesota, Missouri, New York, Pennsylvania, Tennessee, and Wisconsin. The same number of relatively anti-isolationist districts were selected from the same states. The anti-isolationist districts contained fewer counties than did the isolationist districts.

¹⁵ The urban areas and states used were as follows: Los Angeles County and San Francisco, California; Cook County (Chicago), Illinois; Wayne County (Detroit), Michigan; New York City, New York; Philadelphia and Alleghany County (Pittsburgh), Pennsylvania; Boston, Chelsea, and Cambridge, Massachusetts; Milwaukee County, Wisconsin; Denver, Colorado; Baltimore, Maryland; Minneapolis and St. Paul, Minnesota; Cleveland and Cincinnati, Ohio; Birmingham, Alabama; St. Louis, Missouri; Louisville, Kentucky.

¹⁶ The classifications are used by Holcombe, Arthur, The Political Parties of Today, pp. 62–63Google Scholar. Metropolitan districts are those formed in or containing an urban area consisting of a city of at least 200,000 inhabitants. An urban district contains a city of at least 100,000 population; semi-urban districts contain a city of at least 50,000 persons but none of 100,000. Rural districts have no city as large as 50,000. In this comparison the isolationist districts were all those having a percentage of isolationism greater than 80 per cent. The anti-isolationist districts had less than 5 per cent isolationism and were located in the same states as the isolationist districts.

¹⁷ Most opinion polls show that higher education is correlated with less isolationist sentiments. Of thirty-nine postwar polls presenting educational breakdowns on foreign affairs questions, thirty-six found college educated less isolationist than high school educated. In thirty-eight of the polls high school educated persons were found to be less isolationist than grade school educated.

¹⁸ These maps appear in Statistical Atlas of the United States (for the Eleventh Census, 1890, Washington, D. C., 1898)Google Scholar, Maps 56–61. See also similar statistical atlases based on the 1910 census, plates 146–184, and the 1920 census, plates 147–180.

¹⁹ Findings in this research do not support the relatively strong statement of relationship between isolationism and national origins found in Samuel Lubell's works, n. 1. Though the relationship exists, it is but one of several such correlations, and does not seem to be the strongest.