Our initial experience with diagnostic ultrasound mammography (UM) showed it to be of high diagnostic accuracy and improved specificity when compared with x-ray mammography (XRM) in certain categories of patients. To evaluate this modality further, we reviewed our experience with 600 consecutive patients who underwent ultrasound mammography as part of their initial evaluation for breast disease. Five categories of patients were reviewed: I: age 35 or less--129; II: DY or P-2 pattern on XRM--174; III: negative XRM but symptoms requiring additional evaluation--81; IV: high risk remaining breast in patients previously having mastectomy--48; V: refusal of repeat XRM because of pregnancy, lactation, or fear of radiation exposure--168. Fifty-five solid lesions were diagnosed, including 36 carcinomas. Seventeen of these were not diagnosed on x-ray mammography, giving an improved pick-up of carcinoma in this overall group of 2.83%. Of particular interest was that, of the 27 carcinomas that were found in the group having XRM, 8 or 29.6% were missed on that modality alone. Five of these (62.5%) were patients having a DY pattern. All patients have been followed carefully and re-examined at 3-month intervals, with no evidence of a false negative examination. Overall, we feel diagnostic ultrasound mammography is a valuable adjunct to x-ray mammography and recommend its continued use in selected groups of patients where it has a superior diagnostic accuracy over x-ray mammography alone.