We used radiocarbon ages on dead Holocene shells of the venerid bivalve Chione spp. to investigate how time-averaging and taphonomy in shallow marine benthic assemblages vary with sedimentary and tectonic setting. We compared shells collected from the sediment surface in five depositional environments from two regions of the Gulf of California, Mexico: Bahia Concepcion, a young faulted rift basin with high rates of terrigenous and carbonate sedimentation; and Bahia la Choya, an intertidal system along a sediment-starved shelf. Frequency distributions of shell ages in all environments form a hollow curve, with a mode at young ages and a long tail toward older ages. This pattern suggests that shells are added to the taphonomically active zone (TAZ) at roughly constant rates (via continuous shell deaths), and removed from the TAZ at random, either through destruction or by achieving final burial. Shell half-lives (the amount of time to remove half the shells from the TAZ) provide a comparative measure of time-averaging (shell half-lives of 90 to 165 years) occur in Bahia Concepcion, where rapid rates of terrigenous sedimentation (on fan-deltas) and carbonate sedimentation (in pocket bays) bury shells rapidly. Time-averaging is higher in the sediment-starved environments of Bahia la Choya (shell half-lives of 285 to 550 years). The highest amounts of time-averaging occur the inner tidal flats of Bahia la Choya (shell half-life of 550 years). Here the conjunction of low sedimentation rates with low rates of shell destruction (due to periodic tidal emergence) permits shells to persist in the TAZ for very long time spans. There is no systematic relationship between a shell's age and its taphonomic condition (taphonomic grade) in any environment, probably because of the complex and random nature of burial-exhumation in the TAZ. Age variance tends to increase with increasing taphonomic alteration: highly altered shells range in age from young to several thousand years old, while less altered shells are mostly young. The correspondence between time-averaging and the taphonomic condition of entire shell assemblages is also weak, but might be resolved with further study. These results provide quantitative data on time-averaging in benthic assemblages as a function of sedimentary and tectonic setting, and suggest some guidelines for facies appropriate for particular studies. Shallow marine rift basins like Bahia Concepcion can potentially contain within-horizon fossil assemblages representing time spans of only a few hundred years--time resolution often beyond reach in paleontology. In contrast, sediment-starved shelf habitats like Bahia la Choya are unlikely to yield assemblages with time resolution finer than several thousands of years.