To assess the potential of the Cd hyperaccumulator Thlaspi caerulescens (Ganges ecotype) for Cd phytoremediation in Japan, we compared the changes in the soil Cd concentration between a Fluvisol and an Andosol and the efficiency of Cd removal under different planting frequencies in a pot experiment. The soils were artificially contaminated with Cd(NO3)2 to the level of about 5 mg Cd kgâ1. The cultivation system of Thlaspi caerulescens applied to both soil types continued for 4 months (4 Ă 1 treatment) and the other system only applied to the Fluvisol consisted of two plantings, at the beginning and after 2 months (2 Ă 2 treatment). Control without plants was also prepared. The cumulative amount of Cd absorbed by shoots for 4 months was 385 ÎŒg potâ1 for the 4 Ă 1 treatment of the Fluvisol, 582 ÎŒg potâ1 for the 2 Ă 2 treatment of the Fluvisol, and 112 ÎŒg potâ1 for the Andosol. In the Fluvisol, Thlaspi caerulescens plants in the 2 Ă 2 treatment could remove 1.5 times more Cd than in the 4 Ă 1 treatment. A much higher Cd concentration in the shoots of the second cropping in the 2 Ă 2 treatment mainly led to this difference in efficiency. The total Cd concentration of the unplanted Fluvisol and Andosol was 5.0 and 6.1 mg kgâ1, and after 4 months, that of the Fluvisol in the 4 Ă 1 and 2 Ă 2 treatments and the Andosol reached 4.0, 2.9 and 5.1 mg kgâ1, respectively. The difference between the rhizosphere and the nonârhizosphere soils was not evident in the Fluvisol, while the concentration in the rhizosphere was lower than that in the nonârhizosphere soil in the Andosol. In the Fluvisol, the amount of available (1 M NH4NO3âextractable) Cd decreased only in the rhizosphere, while in the Andosol, it was halved in both rhizosphere and nonârhizosphere soils. Furthermore, the ratio of the amount of available Cd to total Cd did not decrease in the Fluvisol, whereas it decreased in the Andosol, which indicated the presence of a smaller pool of potentially available Cd in the Andosol. Based on the shoot biomass and bioaccumulation factor obtained in the 4 Ă 1 treatment, it was estimated that only about 2 crops for the Fluvisol and 6 crops for the Andosol could decrease the Cd concentration to 50%. In conclusion, 1) shortârotation is recommended because of the high Cd concentration in the shoots of the second cropping and the increase in the volume of the rhizosphere soil, 2) the amount of available Cd in soil should be monitored in addition to the concentration of total Cd to evaluate the effectiveness of phytoremediation, and 3) the potential of Thlaspi caerulescens was clearly demonstrated and the use of this hyperaccumulator could become a promising alternative for the remediation of Cdâcontaminated soils in Japan.