Four methods of assessing oestrogen receptor (ER) status were compared in 33 patients with operable primary breast cancer. The methods used to assess the ER status were immunocytochemical assay (ER-ICA) of frozen sections, fine needle aspirates and imprint material and enzyme immunoassay (ER-EIA) of tumour tissue. A mean overall ER positivity of 45% (15 out of 33), 41% (13 out of 32) and 21% (six out of 29) was obtained by immunocytochemical (ER-ICA) staining of frozen sections, fine needle aspirates and tumour imprints, respectively, and a mean overall ER positivity of 42% (14 out of 33) was obtained by ER-EIA. The concordance of ER positivity in pairs of data obtained from different method combinations was found to range between 72 and 91%. However, statistically there was no significant difference between the four methods on the basis of the data obtained. Good comparability has been shown between the three tissue analyses and therefore the method of choice is technically not immediately apparent.