In this study, gypsum, a common amendment for sodic soil reclamation, was compared with langbeinite, a lesser used and known mineral. A column leaching experiment using sodic water was conducted on a sodic, non-saline soil (fine, montmorillonitic, thermic Ruptic Vertic Albaqualf) dominated by smectitic clays. Soil was amended with gypsum and langbeinite at rates equivalent to exchangeable Na at soil depths of 0.15 and 0.30 m. The soil water at depths of 0.75, 0.15, and 0.225 m and the effluent from each column were collected at intervals of 12 h and analysed for soluble bases. Sodium adsorption ratio (SAR) was calculated from soluble salts. Saturated hydraulic conductivity (K_sat) was calculated. At the end of the experiment, soil samples were removed from each column in 4 depth increments. Significantly less exchangeable Na and lower SAR of the soil water was found in the lower sections of the soil columns, and K_sat was greater for the amended treatments than for the control. High solubility of the langbeinite resulted in the highest K_sat value, with possible increase in electrolyte concentration and reduction of clay swelling and dispersion in the first 12 h. However, there was no significant difference in reclamation efficiency between equivalent rates of 2 amendments throughout the experiment. This experiment indicated that factors influencing the decision about using either amendment should be availability of the product, the seasonal fluctuation in price, required reclamation time, and the crop needs for Ca or Mg and K.