Production of sclerotia of Rhizoctonla solani on infected rice plants and their population in different soil types were evaluated during the year of 1992/1993 and 1993/1994. The production of sclerotia was estimated on 20 diseased rice plants and plant debris (rice straw) placed on soil surface, in 10 cm depth, and in 20 cm depth. The population of sclerotia in the soil was estimated by separating the sclerotia from soil samples collected from different soil previously planted with different crops. Data indicated that during the rainy season of 1992/1993, the mean sclerotia produced were 14.85 and 10.95 per hill on the variety of IR64 and non-lR64, respectively. While during the dry season of 1993 the mean sclerotia produced on these varieties were 7.50 and 7 .25 per hill. On both varieties, the production of sclerotia was positively correlated with disease severity of sheath blight, as indicated by the correlation coefficient of 0.90 and 0.70, for the variety of IR64 and non-IR64, reepectively. Their close relationship was estimated by the model of Y=-29.00+1.16x (R^2=0.82) and Y=-2.94+0.35x (R^2=0.45), for the variety of IR64 and non-IR.64, respectively. The production of sclerotia on the infected rice straw was significantly affected by the soil depth where the diseased straw were kept. On the straw of IR64, the sclerotia produced were 7.00, 5.25, and 1.25, when the straw were kept in the depth of 0, 10, and 20 cm, respectively. While on the straw of non-IR.64 variety, the sclerotia produced were 7.75, 5.25, and 0.50. when the straw were kept in the depth of 0, 10, and 20 cm, respectively. Highest number of sclerotia was observed in Ultisol soil previously planted with corn, while the smallest was in Ultisol previously planted with mungbean.