Path dependence and imprinting are two important but distinct theoretical lenses to understand persistent effects of the past. While both concepts focus on how history has a lasting effect, they differ substantially in terms of how the past effect is formed, in particular (a) the importance of initial conditions (weak versus strong) and (b) the mechanism sustaining the historical effect (self-reinforcement versus inertia). To differentiate the two perspectives theoretically and empirically, we conduct a study of firm innovation—a context in which prior research has examined both path dependence and imprinting processes—with a longitudinal sample of U.S. publicly traded companies from 1981 to 2010. We find that prominent features of the founding environment and firms’ founding inventors strongly and enduringly affect patenting behaviors, and that after including initial condition variables that tap imprinting processes, effects of variables representing path dependence become substantially weakened. Our results suggest that imprinting and path dependence should be better differentiated in sociological research, and that prior research on path dependence may be mis-specified because of ignoring founding conditions.