Phosphorus, potassium, calcium, magnesium, manganese, and zinc extracted by three soil extractants at different soil-solution ratios

Sixty soil samples, representing 20 soil series, were analyzed by each of three soil testing procedures (Double Acid, the Tennessee soil testing procedure, and a newly proposed procedure by Mehlich) at six soil to solution extracting ratios (1:4, 1:6, 1:8, 1:10, 1:15, and 1:20) in triplicate for P, K, Ca, Mg, Mn, and Zn. The soil test procedures were compared to each other and to other standard extraction procedures that are generally used for the tested elements. The extraction procedures used were Bray 1 for P, 0.1N̲ HCl for Zn, and neutral normal ammonium acetate for K, Ca, Mg, and Mn. Phosphorus was determined colorimetrically on a Technicon AutoAnalyzer, potassium was determined by flame emission on a Technicon Flame Photometer III, and calcium, magnesium, manganese, and zinc were determined by atomic absorption spectrophotometry. The experimental design used was a nested design. Regression and correlation were performed using the amount of the test elements extracted by each soil test procedure as the dependent variable. Duncan's New Multiple Range Test was used to separate significantly different means. The chemical extractant, soil-solution extracting ratio, soil sample, and all interactions affected the amount of P, K, Ca, Mg, Mn, and Zn measured. The coefficients of determination (R²) were high, indicating that most of the variance in the tested elements could be explained by the analysis of variance model chosen. The coefficients of variation (C.V.) differed only slightly among chemical extractants for the same element indicating similar data dispersion for the chemical extractants. In general, except for Ca and Mg, each increase in soil-solution ratio from 1:4 to 1:20 resulted in significant increases in the amount of P, K, Mn, and Zn measured for all extractants. The data for Ca and Mg were more variable and the same trend was not evident. Across soil solution ratios, the Tennessee procedure extracted significantly more P, K, Mn, and Zn, and the Double Acid procedure extracted significantly more Ca and Mg than the other extractants. Although there were differences in the amounts of the test elements measured, the amount of P, K, Ca, Mg, Mn, and Zn extracted by each chemical extractant was highly correlated to the amount of the elements extracted by the other generally used extraction methods. Regression analyses indicated a linear relationship between P, K, and Zn extracted by all three chemical extractants and the soil-solution extraction ratio used. Only for the Tennessee extractant was the soil solution extraction ratio a significant predictor for Ca. There was a curvilinear relationship for Mg and Mn with soil-solution extracting ratio with all of the chemical extractants tested. Based on the data collected and the statistical analyses performed, there appears to be little, if any, real difference in the three tested procedures. The Tennessee and Double Acid extractants in most cases extracted more of the tested elements, but correlation analysis indicates all procedures to be highly correlated to standard extraction procedures. The data dispersion, as shown by the coefficients of variation, was very similar for all chemical extractants for each tested chemical.