MIXED ELECTORAL SYSTEMS seem to be in vogue. In recent years Ukraine, Russia, Hungary and Japan have adopted electoral systems in which one portion of the legislature is elected in single-member district elections and another portion is elected in a nationwide proportional representation election. Significant work has been done analysing the effects of mixed electoral systems on party systems,1 but little work has focused on the effects of mixed systems on voting patterns in legislatures. This is surprising given the fact that the vast literature on party systems highlights significant differences between the interests of deputies elected in single-member district (SMD) and proportional representation (PR) systems. SMD deputies are thought to be more beholden to local, parochial interests than PR deputies are. Conversely, PR deputies, who are dependent on the party leadership for placement on the list to obtain seats, are more closely tied to the party than are SMD deputies. If this is true, mixed electoral systems should produce legislatures with cleavages that partition the legislators along mandate lines. I begin with a brief review of the theoretical literature on which voting differences between SMD and PR legislators are hypothesised. This review will be followed by an analysis of Duma roll-call voting data, which will demonstrate that, even controlling for other factors, mandate has an independent effect on legislative voting. Finally, the conclusion will discuss the implications of the arguments made here.