The causes of recent increase of the population of the southern green stink bug, Nezara viridula, were analyzed by the construction of life tables of three consecutive generations in 1961.<br>236 plants of potato, 1, 250 plants of early planting rice, and two plots of late planting rice, each contained 500 plants, the one as the untreated and the other as the spider-erradicated where spiders were erradicated by hands every other day, were used for investigations in the first, the second and the third generations, respectively. Census was made twice a week for the first (7th May to 17th July) and the second generations (5th July to 16th Aug.), and every other day for the third generation (5th Sept, to 29th Oct.).<br>Actual numbers were used for each stage of the first generation and for egg, the first instar and the second instar of other consecutive twogenerations. Corrected value of the population of each stage less the third instar was obtained by the following formula: A⋅P/I=N…(1), where A=total counts of incidence of each instar, P=the mean interval of surveys for the period of incidence of the instar concerned, and I=the mean developmental duration of the instar. Duration of each stage was assessed at 30°C on pods of the common haricot for the second generation and directly in the census paddy-field for the third generation. Total durations from egg to adult are 40.1, 34.7 and 42.5 days at 25°, 30°C and under natural conditions of the third generation, respectively. Period of egg laying was two to three weeks. Mean densities of egg masses per plant were 0.10, 0.10 and 0.07 and the mean number of eggs per mass 74.1, 82.5 and 97.9 for each successive generation.<br>Survivorship curves for the period egg to adult emergence exhibit moderately concave curves, somewhat between the curves of Type II and III in Deevey's sense. Mortality curves for the generations of the first and the second have a peak in the period of early stages, two peaks can be expected for the third generation which should have another peak of mortality in adult stage before egg laying caused by overwintering death. The 100qx of the first generation from egg to the second instar decreases, and increases as the stage advances, on the contrary, in the generations of the second and the third. These differences among generations are largely due to the high percentage of parasitism among eggs of the first generation.<br>Principal mortality factors in egg stage are egg-parasites, physiological death and climatic factors. Asolcus mitsukurii is the commonest species throughout seasons, but Telenomus nakagawai rarely appears in the third generation. Other species, namely T. gifuensis and a Eulophid species almost confined their activities to the second generation. Total percentage parasitism was 74, 26 and 22 per cent. for three consecutive generations. A clear relationship was observed for A. mitsukurii that eggs laid in the later period of egg laying are attacked more intensively than those in the early period. But this relationship does not hold for T. nakagawai.<br>For larvae of young stages predation by spiders or destruction by weather factors usually resulted in the disappearance of larval colony as a whole due to the gregarious habit of young larvae. Studies for 95 egg-masses of both the first and the third generations, and calculation of the disappearance rate for the second generation revealed that the larvae of the second instar are most vulnerable to the predation of spiders. The rôle of spiders in the third generation estimated as percentage reduction of larval loss which are attributable to erradication of spiders was at least 2.30% in terms of the number of larvae that hatched.<br>Weather as direct death factor seems to be immaterial to egg and mature larvae under usual conditions.