The transfer of 90Sr to rice plants following its acute ground deposition was examined experimentally in a greenhouse. Lysimeters were flooded after being filled with the soil monoliths from 12 paddy fields. A solution of 90Sr was applied to the standing water in the flooded lysimeters at the pre-transplanting stage or booting stage. Applied 90Sr was mixed with the topsoil only after the pre-transplanting application (PTA). The transfer was quantified with the areal transfer factor (TF(a), m2 kg(-1)-dry) defined as the ratio of the plant concentration to the initial ground deposition. In the PTA, the first-year TF(a) values in the 12 soils were in the range of 8.2 x 10(-3) -2.1 x 10(-2) and 1.7 x 10(-4) -3.6 x 10(-4) for the straws and hulled seeds, respectively. The TF(a) values from the booting-stage application (BSA) were higher than those from the PTA by a factor of up to four. The ratios of the seed TF(a) to the straw TF(a) were, on the whole, higher in the BSA. The 90Sr TF(a) in the PTA was negatively correlated with the soil pH and, to a lesser degree, the exchangeable Ca content. In the second year, the TF(a) in the PTA reduced to 53-90% of that in the first year. A more significant reduction, in general, occurred in a sandier soil. Based on the four consecutive years' transfer data, an overall half-time of the 90Sr TF(a) was estimated to be 2.2 years.