In a pot experiment the effect of silicate on P availability for rice grown in a P-deficient soil with and without flooding was analyzed. Treatment were designed as follows: C (control: Yakuno soil), SS (sodium silicate application, at 0.47 mg Si g-1 soil) and SC (sodium carbonate application). In order to separate pH effect from Si effect, SC was adjusted to the same pH as SS. Soil pH of SS and SC increased by 1.0 unit. Shoot dry weight of SC plants, and more so of SS plants, increased under both nonflooded and flooded conditions. P concentrations in the shoots were not increased under either condition of SS and SC. With SS, Si concentration in the shoots significantly increased, Mn concentration significantly decreased, resulting in a higher P/Mn ratio in the shoot, but not with SC. Both SS and SC increased N concentration in the shoots nearly two times compared with control under both conditions. Adsorption experiments showed that neither SS nor SC decreased P adsorption by soil. SS also could not displace the adsorbed P in soil samples which had previously either received P or not. These results suggest that the beneficial effects of silicate on rice growth do not result from increasing P availability in soil. The Si effect may be attributed to decreasing Mn uptake, thus indirectly improving P utilization in the plant.