Abstract
Results of analyses of variance, linear regressions, and a pot trial are presented in the comparison of a resin method (resin K) and two acid-extraction methods (Kc and step K) of determining non-exchangeable K in suils. All three methods were significantly correlated with each other, although the correlatiOn between Kc and resin K appeared to operate indirectly through their separate correlations with step K. The three methods also exhibited similar correlations with crop yield, K uptake, and % K in herbage during the pot trial. However, step K was the only one to show a significant reduction in its level over all soils as a result of the trial. This reduction when summed with the reduction in exchangeable K was significantly correlated against total K uptake. Of the three methods step K was considered best suited to the prediction of crop yield and K uptake, although it was noted that the resin K method could be preferred because of its practical advantages. Results of sand, silt, and clay contributions to whole-soil resin K values are also reported.