Abstract
New data on mortality and voyage length in the nineteenth-century slave trade make possible further testing of hypotheses on why slaves died during the middle passage. Mortality rates (defined as death per slaves embarked/voyage length in days × 1000) were higher in the nineteenth century than in earlier centuries and varied markedly between regions of embarkation. In the high mortality regions, all ships in the sample appeared to have experienced a higher death rate, suggesting that epidemics were not of prime importance. Mortality rates do not appear to have fluctuated very much during the voyage nor does the slaves–per–ton variable have much explanatory power. The major explanation is probably endemic disease.