tudies of the the U.S. Courts of Appeals increasingly have moved beyond studying the voting behavior of judges in isolation from their panel colleagues and toward an approach that takes into account how panel composition can affect both individual judicial decisions and, as a result, the final decisions of three-judge panels. This paper presents the first assessment of the rates of various panel configurations over time on the Courts of Appeals, showing that while long stretches of single-party control of the presidency in the first half of the 20th century often produced a high rate of panels with three judges from the same party, frequent turnover of White House control in the last half-century has helped ensure that a majority of panels are composed of at least one judge from each party. Second, this paper presents the first longitudinal analysis of panel composition and judicial behavior, showing that the relationship between the two is a relatively recent phenomenon. These findings have important implications for the study of collegial behavior on the Courts of Appeals.