Grey-scale ultrasonography will detect reliably the presence of clinically significant popliteal cysts, fluid collections which do not fill by arthrography, and will frequently demonstrate ruptured cysts and the soft tissue changes resulting from a recent leak. Forty-eight knees, in 25 patients with popliteal and/or calf pain were examined by ultrasonography followed by arthrography. Popliteal cysts were demonstrated in 40% (19/48) by ultrasound and in 46% (22/48) by arthrogram. For comparison between arthrography and ultrasonography chi2 = 8.58 and contingency coefficient, phi = 0.42 (p less than 0.01). Acute cyst rupture was shown in 2 patients (8%) by both arthrography and ultrasound. In a further study ultrasonography demonstrated popliteal cysts with a prevalence of 31% (22/72) in 36 patients with definite or classical rheumatoid arthritis compared with 4% (3/72) in controls closely matched for age and sex. This difference in prevalence between the rheumatoid patients and controls was highly significant chi2 = 17.48, p less than 0.001. Ultrasonography, therefore, will demonstrate noninvasively the presence of popliteal cysts, may assist in the diagnosis of rupture, and furthermore will assist in quantitative, sequential assessment of patients with painful knees and calves.