The Appalachian Mountains (eastern United States) are the archetypal old, long-decaying orogen from which major theories for long-term landscape evolution have been derived. However, given the variability of relief and topographic correlation with geologic and tectonic history, it is difficult to describe the orogen as old and uniformly decaying. Long-term and short-term estimates suggest slow and steady erosion at ∼20 m/m.y.; however, intermediate-time-scale data like sediment accumulation rates and river incision suggest unsteadiness, which we assess using apatite (U-Th)/He thermochronology. All cooling ages from the central Appalachian hinterland in Pennsylvania and New Jersey and from the rugged Blue Ridge Mountains of western North Carolina are pre-Cenozoic, which places an upper limit on the volume of sediment that could have been sourced from these regions in connection with the documented large accumulation of Miocene siliciclastics offshore. Interpreting the timing and processes governing landscape evolution in these regions was hindered by complex age relations between neighboring samples and considerable age dispersion within individual samples. Through experiments with physical abrasion using two representative samples from the Blue Ridge Mountains, we find that variable zonation of U and Th in conjunction with radiation damage–induced differences in helium diffusivity is the source of age dispersion. Abraded grains produced a strong correlation between age and effective uranium concentration (eU) that was not observed for untreated grains and is expected as a result of grain-specific accumulation of radiation damage during slow cooling. Cooling histories derived from inverse modeling of the eU-age relationship of the abraded grains suggests that for a period of ∼60 m.y. during the Late Cretaceous, valley floors were exhuming at nearly twice the rate of neighboring ridge tops, generating relief equivalent to the modern landscape. This result illustrates that at least portions of the modern landscape are not a direct erosional remnant of long-dead orogenic processes and suggests that significant modifications of the Appalachian landscape can occur within the framework of slow long-term average erosion rates.