Abstract
A potassium persulfate‐sulfuric acid autoclave digestion was used to determine phosphorus (P) in 10–50 mg soil and sediment samples. This safe, convenient digestion method released a large fraction of the P released by the hazardous perchloric acid digestion method. Total P in stormflow sediments by perchloric acid digestion can be predicted with acceptable accuracy from P by the acid‐persulfate digestion provided that homogeneous, replicate subsamples are analyzed. Nonrepresentative subsampling is the main source of analytical error in the P results by either digestion method.