The errors involved in pH determination in soils.

Errors of replication of pH values of 1: 5 soil-water suspensions are shown to differ significantly between routine observers and to be larger when duplicate determinations are made upon different days rather than on the same day. For the routine technique employed in these laboratories the 5 per cent. fiducial limits of a single determination do not rise above ±0.09 pH units due to the above causes. Errors due to soil variation over small distances in the field may show 5 per cent. limits as high as ±1.3 pH units. The causes of the errors which may arise within the measuring system are considered and details are given of errors in soil systems due to the suspension effect and to lack of equilibrium between the soil and aqueous phases. Absence of equilibrium may give differences as high as 1.0 unit when measurements are made upon sedimenting alkaline suspensions; no errors occur due to this cause below pH 5. The presence of salts does not modify the differences observed. The suspension effect is relatively small. It is recommended that where possible, pH measurements be made upon soil systems with the glass electrode in the suspension and the reference electrode in the dialysate or supernatant liquid. The description and use of a suitable electrode arrangement is given in an appendix. Generally pH measurements can be considered to no greater accuracy than ±0.1 unit and quite often circumstances do not justify this precision.