Oestrogen receptors were measured by a histochemical technique in tumour tissue from 45 patients with breast cancer, and the results compared with receptor levels measured by a standard biochemical dextran-charcoal assay. Histochemical measurements were made by incubating frozen sections of tumour tissue with 17-fluoresceinated oestrone, using dithiothreitol to inhibit uptake by low affinity binding sites, and estimating receptor content by the extent of fluorescence seen microscopically. Results from the two assays gave varying estimates of receptor concentration, with 66% of patients classed as receptor-positive by the biochemical assay, and 44% of patients classed as receptor-positive using the histochemical technique. Because of these discrepancies the histochemical assay cannot be considered a direct substitute for biochemical receptor measurements. Progesterone receptor was detected by biochemical measurement in 80% of the subgroup of patients who were positive for oestrogen receptor in both assays. The histochemical assay may thus identify a subset of those patients considered oestrogen receptor-positive by conventional receptor assays, and this group may be more likely to respond to hormonal therapy. However, whether the histochemical method provides a predictive measurement for endocrine responsiveness which is superior to standard progesterone receptor assay remains to be determined.