A recent publication from this laboratory (1) described an enzymatic assay for inorganic phosphate (Pi) which eliminates the need for standards and, through mild reaction conditions, avoids the hydrolysis of labile organic phosphates. Those features are advantageous, particularly for Pi measurements in biological samples. However, subsequent inquiries from other laboratories and our own experience indicated that the assay, as described (1), does not perform well. Specifically, it was found that the assay range was 10-fold narrower than that reported, completion times were 3- to 5-fold longer, and the reaction with Pi standards was only 90-95% complete. Because of these deficiencies we have systematically evaluated every aspect of the assay and have found that the difficulties are eliminated and the assay is improved and simplified by the following changes; (i) Triethanolamine is used in place of Tris as the assay buffer; (ii) triose phosphate isomerase is eliminated and the levels of other enzymes are adjusted to obtain optimum reaction conditions; (iii) ammonium sulfate is removed from the analytical enzymes. The modified procedure described below is more convenient, is linear up to Pi concentrations of 0.1 μmol/ml in the assay, gives complete reaction of Pi standards and quantitative recovery of Pi added to biological extracts, and comes to stable endpoints in 30 min or less (depending on the amount of Pi in the assay). © 1979.