The theoretical background for the simulation of a hunting rate from age-composition data is discussed using excavated sika deer (* 2Cervus nippon* 1) remains. To compile life tables for the excavated collections, the number of live deer is calculated from the number of deaths in the age composition. Survival rate was defined as the total number of survivals after one year, as compared against the total number of live deer in the life table. Survivorship curves of a non-hunted population are simulated based on the known age composition of recent deer populations. The hunting rate is induced as the difference between the survival rate of a hunted population and that of a non-hunted population. Ages of excavated deer remains from 14 sites were determined both by the annulation of tooth cement and observation of wear conditions. Age compositions were classed into three patterns: adult-abundant, juvenile-dominant and fawn-abundant patterns, suggesting that the former two patterns were probably harvested by hunting, and that the last pattern was possibly a result of trapping. The calculated survival rates were all significantly higher than the simulated rate for the non-hunted population. Hunting rates were estimated at about 5% for the adult-abundant pattern, which was detected mainly during the Earliest and Early Jomon Periods and about 10% for the juvenile-dominant patterns, which were distributed throughout the later periods. -- AATA