¹ See, among others, Mayhew, David R., ‘Congressional Elections: The Case of the Vanishing Marginals’, Polity, VI (1974), 295–317CrossRefGoogle Scholar; and Fiorina, Morris P., ‘The Case of the Vanishing Marginals: The Bureaucracy Did It’, American Political Science Review, LXXI (1977), 177–81CrossRefGoogle Scholar; Fiorina, Morris P., Congress: Keystone of the Washington Establishment (New Haven: Yale University Press, 1977).Google Scholar

² Buchanan, Christopher, ‘Senators Face Tough Re-election Odds’, Congressional Quarterly Weekly Report, 38 (5 04 1980), 905–9.Google Scholar Buchanan shows (p. 908) that the success rate for incumbents seeking re-election to the Senate did not fall below 70 per cent from 1960 to 1974; except for the landslide year of 1974 (when the success rate was still 88 per cent), House incumbents seeking another term have not fallen below 90 per cent success rate since 1966. Also cf. Kaiser, Robert G., ‘Anxious About '80: Wind from Right Chills Senate Liberals’, Washington Post (10 04 1979)Google Scholar, A1, A7; and Barone, Michael, ‘A Republican Senate?’, New Republic (13 10 1979), 8–12.Google Scholar The liberals defeated in 1978 were: Wendell Anderson (Minnesota), Dick Clark (Iowa), Floyd Haskell (Colorado), William Hathaway (Maine), and Thomas McIntyre (New Hampshire); the liberals re-elected were Joseph Biden (Delaware), Claiborne Pell (Rhode Island), and Jennings Randolph (West Virginia). The liberal Democrats defeated in 1976 were Vance Hartke (Indiana), Gale McGee (Wyoming), Joseph Montoya (New Mexico), Frank Moss (Utah), and John Tunney (California). The classification was done by comparing scores from groups such as the Americans for Democratic Action with Congressional Quarterly conservative coalition and party unity scores. The classifications of Randolph, McGee and Montoya are somewhat controversial and were made at least partially in the light of their challengers.

³ On life-style issues, see Miller, Warren E. and Levitin, Teresa E., Leadership and Change (Cambridge, Mass.: Winthrop, 1976)Google Scholar, Chaps. 4, 7, and 8. For the evidence on issues of spending, see Watts, William and Free, Lloyd A., State of the Nation III (Lexington, Mass.: Lexington Books, 1978), pp. 62–4Google Scholar and especially Table 3–4, p. 63, where Watts and Free show a decline in public support for spending on fourteen items and increased support for only two, both related to defence, over the 1972–76 period.

⁴ Hucker, Charles W., ‘Senate: Slightly More Conservative’, Congressional Quarterly Weekly Report, 36 (11 11 1978), 3244–7.Google Scholar

⁵ See Light, Larry, ‘For Many Incumbents, Running for Re-election is Now a Full Time Job’, Congressional Quarterly Weekly Report, 37 (7 07 1979), 1350–7.Google Scholar Light presents data which indicate that Mayhew's trend of ‘vanishing marginals’ in House contests is increasing: the percentage of incumbents for the lower chamber winning with margins exceeding 60 per cent of the vote rose from 692 per cent in 1976 to 766 per cent in 1978.

⁶ Cf. Uslaner, Eric M., ‘Ain't Misbehavin’: The Logic of Defensive Issue Voting Strategies in Congressional Elections', American Politics Quarterly, IX (1981), forthcoming.Google Scholar

⁷ On position-taking, see Mayhew, David R., Congress: The Electoral Connection (New Haven, Conn.: Yale University Press, 1974), pp. 61–73.Google Scholar

⁸ Fenno, Richard F. Jr., Home Style: House Members in Their Districts, (Boston: Little, Brown, 1978), especially Chaps. 3–4.Google Scholar

⁸ Parker, Glenn R., ‘Incumbent Popularity and Congressional Elections’, in Maisel, Louis and Cooper, Joseph, eds., Sage Electoral Studies Yearbook (Beverly Hills, Calif.: Sage Publications, 1980)Google Scholar, forthcoming; and Hinckley, Barbara ‘House Re-elections and Senate Defeats’, British Journal of Political Science, X (1980), 441–60.CrossRefGoogle Scholar

¹⁰ Supporting this argument are data I analysed from the 1978 National Election Study of the Center for Political Studies which indicate that 80 per cent of those who had some type of contact with the incumbent only in Senate races voted for the incumbent compared with 63·8 per cent who had contact with both the Senator and his opponent (and none of the four cases who only had contact with the challenger).

¹¹ Cf. Mann, Thomas E. and Wolfinger, Raymond E., ‘Candidates and Parties in Congressional Elections’, American Political Science Review, LXXIV (1980), 617–32.CrossRefGoogle Scholar They report (Table 12) that both Senators and their challengers made contact with most voters (80 and 70 per cent respectively) through the medium of television; newspapers and magazines rank second (73 per cent and 63 per cent); only mail from incumbents among the other six types of contact reached the majority of Senate voters.

¹² Fenno, , Home StyleGoogle Scholar, especially Introduction and Chap. 1.

¹³ Fenno, , Home Style, p. 36Google Scholar; and Frantzich, Stephen E., ‘The “Perking” Up of Electoral Fortunes: The Use of Perquisites by Congressmen’Google Scholar (United States Naval Academy, mimeo, 1979).Google Scholar

¹⁴ See Parker, , ‘Incumbent Popularity and Congressional Elections’.Google Scholar

¹⁵ Fenno gathered his House data for an off-year and thus it seemed logical to follow in his footsteps. Trip data for 1978 were also gathered, and the overall pattern did not change markedly. Fenno is currently working on a study of the home style of Senators and will present a more complete picture of Senators' trips in that work. I am grateful to the many Senate staffers who provided the data and other forms of encouragement. Every office I contacted was helpful in this regard.

¹⁶ This is essentially the same argument that Fiorina, makes in ‘The Case of the Vanishing Marginals’, p. 181.Google Scholar

¹⁷ These figures are discussed in Fiorina, , Congress: Keystone of the Washington Establishment, p. 61.Google Scholar

¹⁸ Barkley, Alben recounts the following story in his That Reminds Me (Garden City, N.Y.: Doubleday, 1954), p. 165Google Scholar: ‘I would relate to the crowds how I called on a certain rural constituent and was shocked to hear him say he was thinking of voting for my opponent. I reminded him of the many things I had done for him as prosecuting attorney, as county judge, as congressman, and senator. I recalled how I had helped get an access road built to his farm, how I visited him in a military hospital in France when he was wounded in World War I, how I had assisted him in securing his veteran's benefits, how I had arranged his loan from the Farm Credit Administration, how I got him a disaster loan when the flood destroyed his home, etc., etc. “How can you think of voting for my opponent?” I exhorted at the end of this long recital. “Surely you remember all these things I have done for you?” “Yeah,” he said, “I remember. But what in hell have you done for me lately?”’

¹⁹ Perry, James M., ‘Gordon Humphrey: Copilot to Senator in One Short Hop’, Wall Street Journal (22 01 1979), 1, 31.Google Scholar

²⁰ Fenno, , Home Style, Chap. 1.Google Scholar

²¹ Fenno, , Home Style, especially Chaps. 2–4.Google Scholar

²² Fenno, , Home Style, p. 38.Google Scholar The regions employed in Table 2 are those used by Fenno.

²³ This suggestion was made by the anonymous reviewers, one of whom recommended gathering trip data for House members as well. This was simply not feasible. Other trip data on the House provided problems as well. The Frantzich data (see fn. 11) comprise a limited sample for the height of the 1978 campaign; the Parker figures are more complete but are more dated and were not available for analysis. For the Parker data, see his ‘Sources of Changes in Congressional District Attention’, American Journal of Political Science, XXIV (1980), 115–24Google Scholar; and ‘Congressional District Attention: Trends, Cycles, and Cohorts’ (presented at the annual meeting of the Midwest Political Science Association, Chicago, April, 1979). In any event, there are grave problems in comparing trip data gathered by office surveys (such as Fenno and I have done) and by examining travel vouchers from the Clerk of the House (Parker and Frantzich). Discussions with both Fenno and Parker have been helpful in this regard, although they fail to resolve the issue; I, not surprisingly, lean toward the office surveys.

²⁴ Mann, and Wolfinger, , ‘Candidates and Parties in Congressional Elections’Google Scholar, discuss some of the limitations of this survey for analysing Senate races. Controls for party are not presented below because any breakdown beyond simple cross-tabulations for these variables literally produces ‘vanishing marginals’. What few results I have obtained indicate few differences between the parties.

²⁵ Averaging the small number of cases (N = 27) yields 81·5 per cent in the combined category voting for the Senate incumbent.

²⁶ Cf. Hinckley, ,‘House Re-elections and Senate Defeats’Google Scholar; and Jacobson, Gary C., ‘Congressional Elections, 1978: The Case of the Vanishing Challengers’, in Louis Maisel and Joseph Cooper, eds., Sage Electoral Studies Yearbook (Beverly Hills, Calif.: Sage Publications, forthcoming).Google Scholar

²⁷ I owe this idea to Malcolm E. Jewell. A similar statement is found in Hinckley, Barbara, ‘The American Voter in Congressional Elections’, American Political Science Review, LXXIV (1980), forthcoming.Google Scholar

²⁸ Perry, , ‘Gordon Humphrey’, p. 1.Google Scholar

²⁹ It was Bruce Barr, a former member of Senator Pete Domenici's staff (R., New Mexico), who first suggested to me that Friday sessions were becoming less frequent following the defeat of eight Senators in 1978. The figures are 34 Friday meetings out of a total of 178 days in the 1977 session, 26 out of 158 (16·5 per cent) in 1978, and 20 out of 168 in 1979. These data were graciously provided by Beth Shotwell of the Senate Democratic Policy Committee. Senate Majority Leader Robert W. Byrd (D., West Virginia) has told Democratic Senators that no votes would be scheduled on Fridays during the 1980 election year. Senators could thus skip Friday sessions without missing roll calls.