Laboratory procedures for estimating plant-available Al and Mn in soils were compared on 33 Canadian surface soils having a pH range of 3.65 to 4.98 (measured in 0.01 M CaCl2). Barley, rapeseed and buckwheat were grown on the soils in the greenhouse, and correlation coefficients were calculated between laboratory soil measurements and the percent yields (yield unlimed × 100/yield limed) of barley, Al contents of rapeseed, and Mn contents of rapeseed and buckwheat. The quantities of soil Al soluble in 0.01 M and 0.02 M CaCl2 were highly correlated with the Al contents of rapeseed and percent yields of barley. Levels of soil Mn soluble in the same reagents were highly correlated with the Mn contents of rapeseed and buckwheat. The CaCl2-soluble Al was better correlated with the crop data than was exchangeable Al (2 N NaCl). The CaCl2-soluble Mn was better correlated with the crop data than was 0.1 N H3PO4-soluble, H2O-soluble, reducible or exchangeable Mn. The following method is recommended for determining plant-available Al and Mn in acid soils: Shake the soil at a 1:2 (w/v) ratio with 0.01 M CaCl2 for 5 min and measure Al and Mn in the supernatant solution by atomic absorption spectrophotometry.