The P status of the soils of Geneva is routinely determined by two independent chemical extractions. The P amounts extracted independently by an aqueous solution of ammonium acetate 0.5 M + EDTA 0.02 M (pH 4.65) (1:10 solid weight : solution volume, 60 minutes) and by H₂O (1 : 10 solid weight : solution volume, 60 minutes) were assumed to represent mobilizable and mobile phosphorus pools. The best correlations with the data from 1023 crop field soil samples in Geneva canton during the years 1994 to 2000 were obtained by power functions if the data were grouped in two distinct pH ranges:

pH 4.6–6.5: (P-AAE) = 16.436 (P-H₂O) ^0.6083; R² = 0.625; n = 166,

pH 6.6–8.6: (P-AAE) = 25.582 (P-H₂O) ^0.6518; R² = 0.47; n = 857.

The derivatives of these functions are representative of apparent P-buffering power. Environmentally sound threshold values for P-H₂O and P-AAE are estimated by calculating the change points first of these P-AAE (quantity) against P-H₂O (intensity) curves and second P-buffering power against P-H₂O or P-AAE curves.

The change points obtained from the smoothed quantity-intensity data were:

pH 4.6–6.5: P-H₂O = 3.8 mg kg^–1 and P-AAE = 16.436 (3.8)^0.6083 = 37.0 mg kg^–1; n = 166,

pH 6.6–8.6: P-H₂O = 2.5 mg kg^–1 and P-AAE = 25.582 (2.5)^0.6518 = 46.5 mg kg^–1; n = 857.

The change-points for P-buffering power curves occurred at

pH 4.6–6.5: P-H₂O = 1.3 mg kg^–1; P-AAE = 19.7 mg kg^–1; n = 166,

pH 6.6–8.6: P-H₂O = 1.3 mg kg^–1; P-AAE = 20.6 mg kg^–1; n = 857.

The threshold buffering powers were:

pH 4.6–6.5: d(P-AAE)/d(P-H₂O) = 9.998 (P-H₂O) ^–0.3917 = 9.0,

pH 6.6–8.6: d(P-AAE)/d(P-H₂O) = 16.674 (P-H₂O) ^–0.3482 = 15.2.

These results show that presently only about 25 % of the crop soils of Geneva canton were within these safe limits. The possibility of applying values of this magnitude which are significantly lower than officially recommended reference levels for phosphorus fertilizer recommendations is supported by recent agronomic investigations.