Abstract
Various extraction methods were evaluated to find a measure of soil Al that may be related to plant Al and yield in several crops grown on tropical soils. Maize (Zea mays L.) was grown in five unlimed soils collected in Hawaii. Maize, desmodium (Desmodium aparines L.) and Louisiana white clover (Trifolium repens L.) were also grown in Akaka soil limed to create pH differentials. Soil Al was extracted with 13 methods, using five extracting solutions, from soil samples collected after plant harvest. The amount of Al extracted by various solution decreased in the order: In NH4OAc = 0.2N BaCl2, pH 4.8 > 1N NH4OAc, pH 4.8 ? 1N BaCl2 ? 1N KCl ? 0.01M CaCl2. Aluminium extracted from five soils was in the order: Akaka (Typic Hydrandept) > Halii (Typic Gibbsihumox) > Wahiawa (Tropeptic Eutrustox) ? Lualualei (Typic Chromustert) ? Kawaihae (Ustollic Camborthid). Aluminium extracted was related to the degree of weathering (the amount of rainfall) affecting the soil. No single method of extracting soil Al was best correlated with plant Al or yield in the four test crops studied. However, the overall average relationship between soil Al extracted with 0.01M CaCl2 and plant Al was slightly better than that extracted by other solutions. The results show that the multiple correlation coefficients for the soil Al-yield relationships were generally lower than those for the soil Al-plant Al relationships, suggesting that while soil Al has a strong influence on plant Al concentrations, it has considerably less effect on yield in tropical soils.
