The life table studies were carried out on the first and the second generations of Papilio xuthus population at Omura, Nagasaki. Natural and semi-natural (caged in the field) populations were used in order to detect the cause and the rate of mortality in each of the developmental stages, such as egg, five larval instars and pupa. The effect of predation was measured by comparing the two survival curves obtained from the caged (5 mm net) and the uncaged populations. By this treatment, it was made clear that small predators (spiders, etc.). attacked the young larvae and large predators (polistes wasps, etc.) attacked the later larvae. The high mortality rate during the later larval instars was mainly due to the predation. Egg mortality was mainly due to the egg parasite, Trichogramma dendrolimi. Three species of parasitic wasps, Pleromalus puparum, Trogus mactator, and Brachimeria obscurata emerged from the pupae. These egg and pupal parasites killed 24.3 per cent of the eggs and 83 per cent of the pupae in the first generation, though with only slight effect on the second generation. The difference in egg population density between the first (12.7 eggs per 10 plants) and the second generation (3.3 eggs per 10 plants) was partially explained by the difference in the mean fecundities of the adults. The mean fecundity per female was estimated as, M.F.=(L-length of pre-ovipositional period)×Mean number of eggs laid per day. The mean length of the adult life span (L) was given as follows, L=1/(1-K) where K is the daily survival rate of the adult under the natural condition, and K is given as the solution of the following equation, N'/N=K^α where N is the total number of adults in a generation and N' is the number of adults living longer than α days.