The Messel Pit is an ancient maar lake that preserves an ecosystem from the last greenhouse phase of Earth’s climate with exceptional fidelity. There are two main competing hypotheses to explain the occurrence of animal fossils at Messel: asphyxiation in layers of heavy, toxic gases like carbon dioxide, and death following ingestion of surface water contaminated by cyanobacterial toxins. Both hypotheses imply mortality much higher than the null hypothesis of accidental death, for instance by drowning. We tested these hypotheses amongst bats by recourse to a unique survey on bats that drowned in modern swimming pools. We further conducted experiments on the fate of fresh bat carcasses that simulate drowning. Finally, we estimated the annual rate of bat fossil production at Messel. Overall, there are few good surveyed predictors of whether bats drown in swimming pools: most physical parameters like pool size or the height and distance of obstructions were not associated with mortality, although pool liner type would be a good candidate parameter for future study. In our experiments, carcasses with water-filled lungs tend to sink more quickly than those with air-filled lungs; all carcasses that initially sink “bloat and float” at shallow depths, but all floaters eventually disarticulate and sink within days to weeks. Our estimates of annual bat mortality at Palaeolake Messel are of the same order of magnitude as mortality in swimming pools. Mass mortality horizons for vertebrates are very rare at Messel. In sum, our quantitative analyses provide no evidence that bat mortality at Messel rises above background levels.