In this study we extend previous research on the impact of constituency preferences on the roll-call behavior of senators during Supreme Court confirmations by examining the 1967 vote on Thurgood Marshall. In contrast to the later case of Clarence Thomas, we find that constituent racial characteristics (as measured by African-American percentage of state population) had a significant negative impact in the Marshall case. We conclude with a discussion of the theoretical implications of these findings, and what they illuminate about the changing nature of the Southern Democratic coalition.Recent research on the Senate confirmation of Supreme Court nominees has suggested that constituency concerns can exert a substantial influence on senators' voting behavior In an earlier work (Overby, Henschen, Walsh, and Strauss 1992), we tested the impact of constituency preferences on senators' roll-call behavior in the recent case of Clarence Thomas, concluding that state African-American constituency size and reelectoral proximity were positively and significantly related to the probability of a &dquo;yes&dquo; vote. Here, we compare the Thomas vote with Senate action on the earlier Supreme Court nomination of Thurgood Marshall, a case that reflects similar constituency concerns. Our findings demonstrate not only the relevance of constituency preferences in the Marshall vote, but in comparison with the Thomas case speak to the changing demographics of Southern politics.