1Chemistry Department, Himachal Pradesh University, Simla
2Central Potato Research Institute, Simla, Himachal Pradesh
*Present address: Indian Grassland and Fodder Research Institute, Jhansi, U.P.
The efficiencies with which biological and chemical methods estimate plant available phosphorus in eight Indian soils were compared. Biological methods were in general better predictors than chemical ones, when the absolute yield of potato of the control treatment was adopted as the criterion and among them the cunning hamella plaque technique had the highest reliability, followed by the biological slope-ratio method and the L value technique. But the per cent yield criterion was useful preferably with soil chemical tests, of which only Schofield's mono-calcium phosphate potential could predict reliably the phosphorus availability status in these soils. On compressing the widely divergent test values by a log-transformation, Spurway's test turned out to be extremely efficient as a predictor.
With the simultaneous use of phosphate potential and the log-transformed Spurway test value as conjoint predictors, an R2 value of 0.92 was obtained, which thereby improved the value of soil tests as predictors of plant available phosphorus. It is shown that this approach takes into account the Q/I ratio as a measure of soil P availability.
Plant available phosphorus, predictors, criteria biological methods, multiple-predictor technique
