Serum prostatic specific antigen (PSA) and ultrasound-determined prostatic volume (UPV) were measured in 50 patients with histologically proven benign prostatic hyperplasia (BPH) and in 40 patients with histologically proven prostatic cancer of whom 17 had evidence of distant metastases (M1) and 23 did not (M0). A good correlation between log PSA and UPV was demonstrated in the BPH group and rearrangement of the linear regression equation allowed calculation of a single variable--the log PSA corrected to a standard prostate volume for any given individual. A volume-corrected PSA correctly identified all patients with M1 disease and greatly improved but did not eliminate overlap of M0 disease with BPH. Reduction of serum PSA to a single volume-corrected variable will allow the introduction of practical and optimum protocols for the management of patients with prostatic enlargement.