² Stedman, Murray S. Jr.,, “American Political Parties as a Conservative Force,” Western Political Quarterly, Vol. 10 (1957), p. 395CrossRefGoogle Scholar.

³ DeGrazia, Alfred, “General Theory of Apportionment,” Law and Contemporary Problems, Vol. 17 (1952), p. 261Google Scholar.

⁴ For example, see: Kneier, C. M., City Government in the United States, rev. ed. (New York, 1947)Google Scholar, ch. 6; MacCorkle, S. A., American Municipal Government and Administration (Boston, 1949), pp. 52–54Google Scholar; Snider, C. F., American State and Local Government (New York, 1950), pp. 169–172Google Scholar.

⁵ For example, before the 1955 Reapportionment Act in Illinois, and at a time when the Illinois Legislature was characterized as rural-dominated, 53 per cent of the Senators and Representatives came from districts in which more than half the population resided in a city or cities of 10,000 or more. Fifty-seven per cent represented districts in which more than half the population resided in a city of 5,000 or more.

⁶ Key, V. O. Jr.,, American State Politics (New York, 1956), p. 234Google Scholar.

⁷ See Kneier, op cit.; Snider, op. cit.; Baker, Benjamin, Urban Government (Princeton, 1957), pp. 325–328Google Scholar; Baker, Gordon E., Rural versus Urban Political Power (Garden City, 1955), pp. 15–19Google Scholar.

⁸ It should be noted that while Chicago is part of Cook County, St. Louis County and St. Louis City are separate and independent areas and governments. The latter exercises both municipal and county powers. No state legislative district is in both St. Louis City and St. Louis County.

⁹ In the Missouri House each county has at least one seat, but a multiplier effect in the apportionment formula written into the Constitution gives increased representation to heavily populated counties—not, however, in direct proportion to the population of the county. Thus, St. Louis City has 22 per cent of the population and 11 per cent of the seats, Jackson County 14 per cent of the population and 8 per cent of the seats. The Missouri Senate is reapportioned decennially on the basis of population. At present St. Louis City and Jackson County contain 36 per cent of the population and hold 32 per cent of the Senate seats.

¹⁰ Keefe, W. J., A Study of the Role of Political Parties in the Legislative Process, Illinois General Assembly (unpublished Ph.D. thesis, Northwestern University, 1951), pp. 73–74Google Scholar. See also Keefe's, “Parties, Partisanship, and Public Policy in the Pennsylvania Legislature,” this Review, Vol. 48 (1954) p. 450Google Scholar, in which roll-call analyses are used. Young, G. D., The Role of Political Parties in the Missouri House of Representatives (unpublished Ph.D. thesis, University of Missouri, 1958), pp. 86–92Google Scholar. Dr. Young (D-Howard County) was a member of the Missouri House in the 1955 and 1957 sessions.

¹¹ Jewell, M., “Party Voting in American State Legislatures,” this Review, Vol. 49 (1955), p. 773Google Scholar; Lockard, W., “Legislative Politics in Connecticut,” this Review, Vol. 48 (1954), p. 166Google Scholar; Lowell, A. L., “The Influence of Party Upon Legislation,” Annual Report of the American Historical Association, Vol. 1, 1901Google Scholar; MacRae, D., “Relation Between Roll Call Votes and Constituencies in the Massachusetts House of Representatives,” this Review, Vol. 46 (1952), p. 1046Google Scholar; Turner, Julius, Party and Constituency (Baltimore, 1951)Google Scholar.

¹² While thousands of roll-calls occurred in the Missouri Senate over the five-session period, few demonstrated enough conflict to be subjected to the analyses made of other roll-calls in this study. The conventions of the Missouri Senate apparently operate to reduce recorded conflict. Committees function more meaningfully in the Missouri Senate than they do in the House, and since roll-calls are more difficult to take than in the House much more of the Senate business is conducted by viva voce or show-of-hands votes. Undoubtedly the size of the group (34) facilitates pre-voting compromise and consensus. An examination of the contested roll-calls which did occur in the Senate suggests that there was no more, and possibly less, metropolitan conflict than data included here indicate for the House. This is substantiated by Young, George D. in “The 1958 Special Session of the Missouri General Assembly,” Missouri Political Science Association Newsletter, No. 3 (May, 1958), p. 3Google Scholar.

¹³ In order to give concrete meaning to “support” or “opposition” in this study, a delegation is considered to have voted with a high cohesion if the absolute test showed a percentage of at least 67 or more. Most of the findings are presented on the basis of this definition, but findings for other degrees of cohesion are included for purposes of comparison. A vote on which less than half of the delegation was in agreement must be eliminated on the basis that a minority expression should not be imputed to the majority. On the other hand, to demand a cohesion of 100 per cent would eliminate all contested roll-calls from the analysis. Interviews with observers of the Assemblies indicated that, if anything, this 67 per cent level of significance may be too low.

¹⁴ The relatively higher incidence of voting cohesion from 67 to 91 per cent in the 1955 and 1957 sessions of the Senate may be accounted for by the fact that Democrats comprised 83 per cent of the Chicago delegation in 1957, and 89 per cent in 1955. In all other House and Senate sessions, some Republican votes were needed to push delegation cohesion above 67 per cent. The 1955 and 1957 Senate data reflected party conflict.

¹⁵ For analyses of the cumulative voting system and its political implications see Blair, George S., “Cumulative Voting,” this Review, Vol. 52 (1958), p. 123Google Scholar; and Hyneman, Charles S. and Morgan, J. D., “Cumulative Voting in Illinois,” Illinois Law Review, Vol. 32 (1937), p. 12Google Scholar.

¹⁶ Edward D. Banfield has pointed out the inherent political difficulties of metropolitan area organization stemming from control of the central city and the suburbs by two different political parties. “The Politics of Metropolitan Area Organization,” Midwest Journal of Political Science, Vol. 1 (1957), p. 77CrossRefGoogle Scholar. He states that “it seems likely that the central cities will become more and more Democratic” and that “the Republican suburban vote has in general suffered little from the increase in population.” As a result “these facts suggest that for many years to come it will be difficult or impossible to integrate local governments where the two party system operates” because “in effect, advocates of consolidation schemes are asking the Democrats to give up their control of the central cities or, at least, to place it in jeopardy.” It may be that Banfield's argument can be broadened to say that the same forces which bring the central city Republicans and the suburbs into collision with the Democrat central city administration in metropolitan area organization can be expected to appear in legislative consideration of expanded “home rule” powers for the central city.

¹⁷ An example of the opposition of St. Louis County to the St. Louis City earnings tax is provided by a series of 18 House votes on H.B. 50 (1951) which authorized extension of the tax. St. Louis City and St. Louis County delegations were in high opposition on 12 roll-calls, in moderate opposition on 3 roll-calls, and in agreement on 3 roll-calls. Ten of the roll-calls were chamber party votes under the definition of 67 per cent of the elected party groups in opposition to each other, and fifteen were chamber party votes under the definition of 80 per cent of the voting members of the party groups in opposition to each other.

¹⁸ Young, George D., “The 1958 Special Session of the Missouri General Assembly,” Missouri Political Science Association Newsletter, No. 3 (1958)Google Scholar. My emphasis.