The termination of U.S. treaties provides an especially rich example of how governmental practices can provide a “gloss” on the Constitution’s separation of powers. The authority to terminate treaties is not addressed specifically in the constitutional text and instead has been worked out over time through political–branch practice. This practice, moreover, has developed largely without judicial review. Despite these features, Congress and the President—and the lawyers who advise them—have generally treated this issue as a matter of constitutional law, rather than merely political happenstance. Importantly, the example of treaty termination illustrates not only how historical practice can inform constitutional understandings, but also how these understandings can change. Whereas it was generally understood throughout the nineteenth century that the termination of treaties required congressional involvement, the consensus on this issue disappeared in the early parts of the twentieth century, and today it is widely (although not uniformly) accepted that presidents have a unilateral power of treaty termination. This shift in constitutional understandings did not occur overnight or in response to one particular episode but rather was the product of a long accretion of Executive Branch claims and practice in the face of congressional inaction. This article examines the way in which historical practice has shaped the constitutional debates and understandings concerning this issue and is meant to help shed light on some of the interpretive and normative challenges associated with a practice-based approach to the separation of powers.