Soybean does not respond well to high levels of K, because its availability has been related to interactions with Ca and Mg in the soil and, therefore, to liming. In order to study soybean K nutrition related to soil levels of Ca and Mg, a greenhouse experiment was carried out in pots with a sandy loam Typic Hapludox. A 5 x 7 factorial randomized block experimental design was adopted with four replications. Forty days before sowing, five levels of burned dolomite lime (0, 500, 1000, 1500, 2000 mg dm-3) were applied, mixed with seven K levels (0, 15, 30, 45, 60, 75, 90 mg dm-3) in the form of KCl, before sowing. Four plants were grown in each pot. Plants of the early blooming soybean cultivar, IAC-17 with 112-day cycle, were inoculated with Bradyrhizobium japonicum. Two plants were evaluated at the R2 stage (full bloom) and the other two at the R8 stage (full maturity). When the soil ratio of exchangeable (Ca+Mg)/K was greater than 36 or leaf (Ca+Mg)/K concentration greater than 3.6, the plants presented lower dry matter yield, K deficiency symptoms and lower leaf K concentration. Otherwise, greater dry matter yield, related to the optimum equilibrium of leaf K, Ca and Mg concentrations, were verified when the soil ratio of exchangeable (Ca+Mg)/K was between 20 to 30. The soil ratio of exchangeable (Ca+Mg)/K presented an important index to evaluate soil K availability for the soybean crop. Recommendations regarding K fertilization to the soybean crop must also consider the applied liming rates.
Glycine max; potassium; dry matter; (Ca+Mg)/K ratio