Muscle and adipose tissue from a total of 178 experimental bruises inflicted on sheep and aged from 1 to 72 h old were processed for light microscopic examination. Five observed histopathological features of inflammation and repair were scored semiquantitatively on a scale of 1-4 according to their degrees of change from the normal state. These data were evaluated mathematically using a Bayesian probability model designed for the purpose. The model was able to age bruises with an acceptable degree of accuracy only as either 1-20 h or 24-72 h old but within these constraints a degree of confidence in excess of 90% was achieved. The exact performance of the model depended on the nature and number of tissue samples examined but mathematical ageing was superior to interpretative ageing based on personal experience.