Abstract
Fossil assemblages of skeletal material are thought to differ from their source live communities, particularly in relative abundance of species, owing to potential bias from postmortem transport and time-averaging of multiple generations. However, statistical meta-analysis of 85 marine molluscan data sets indicates that, although sensitive to sieve mesh-size and environment, time-averaged death assemblages retain a strong signal of species' original rank orders. Naturally accumulated death assemblages thus provide a reliable means of acquiring the abundance data that are key to a new generation of paleobiologic and macroecologic questions and to extending ecological time-series via sedimentary cores.