This study quantifies the fine structure of time-averaging by using large samples of dated shells collected from within individual strata. Time-averaging results in both good and bad news for interpreting bioclastic deposits. Nine samples of shells were collected from four Holocene cheniers on the Colorado Delta (Gulf of California) and 165 shells of the bivalve Chione fluctifraga were dated using 14 C calibrated amino acid racemization (D-alloisoleucine/L-isoleucine). The age range of shells within samples averages 661 years and, in seven out of nine samples, exceeds 500 years. The sample standard deviation ranges from 73 to 294 years and averages 203 years, far exceeding the dating errors (<100 years) and potential variation in the life span of Chione (<10 years). Time-averaging is homogeneous among strata within cheniers but varies significantly among cheniers. Age-distributions of dated shells indicate that at 50-year resolution, the samples provide a continuous and uniform record for the entire interval. The actual sample completeness (63.6%) is very close to that predicted by simulations of sampling a 100% complete, uniform record (67.3%). The bad news is that, no matter how carefully collected, data from shell beds may not be suitable for studying processes on timescales shorter than 10 2 to 10 3 years; explanations for faunal change that invoke reasoning or models derived from a strictly ecological point of view may rarely be justifiable. Also, notable differences in temporal resolution between the shell beds of seemingly identical origin imply that paleontological patterns (e.g., species diversity) may be affected by cryptic variation in time-averaging. The comparison of our data with time-averaging estimates obtained from other cheniers at coarser sampling resolutions indicates that pooling of samples (analytical time-averaging) can significantly reduce the temporal resolution of paleontological data. The good news is that shell beds can record the optimal type of time-averaging: where paleobiological data are a time-weighted average of the faunal composition from the spectrum of environments that existed during the entire interval of time. Samples from single strata provide a long-term record that is representative of the predominating environments. Within the range of 14 C dating, shell beds can provide a complete, high-resolution record, and thus may offer exceptional insights into the environmental and climatic changes of the last 40 thousand years.