Radiocarbon dates provide a means for estimating the time a shell may persist in active sedimentary environments and the actual temporal extent of time-averaging in marine deposits. Information compiled from the published literature on the radiocarbon age of marine shells gave information on a total of 734 radiocarbon dates from 276 localities from nearshore (< 10 m depth) and shelf (> 10 m depth) habitats. The median age of 128 nearshore shells is 2,465 years; that of 158 shells from the shelf is 8,870 years. The distribution of shell ages in both nearshore and shelf environments is strongly skewed: most dates are in the 0–3,000 range, and the number of shells in older age-classes falls off rapidly. The maximum age of a shell in an active sedimentary environment is a measure of time-averaging, because it estimates the amount of time represented in the deposit. The median duration of time-averaging in 63 nearshore deposits is 1,250 years; the median duration of time-averaging in 129 shelf deposits is 9,190 years. Radiocarbon-dated shells from fossil deposits confirm our estimate of time-averaging in nearshore environments: the median difference between maximum and minimum ages in 49 inactive beach ridges is 1,390 years; the median difference in other, predominantly nearshore deposits, is 830 years. Greater shell survival and longer durations of time-averaging in shelf settings may result from lower rates of sedimentation, lower rates of taphonomic destruction, greater rates of bioturbation, the history of post-glacial sea level, sample bias, or some combination of these factors. Our results may estimate the actual magnitude of time-averaging of shelly deposits forming under conditions of low net sediment accumulation. Such extensive time-averaging may confound attempts at detailed paleoecological and paleoenvironmental reconstruction. □Time-averaging, benthic assemblages, taphonomy, radiocarbon, paleoecology.