The utility of measuring tissue estrogen receptors (ERs) to assist in the determination of appropriate therapy for patients with breast carcinoma is widely recognized. Cytosol assays for ER are being replaced by immunohistochemistry (IHC) of histologic sections. We compare the cytosol assay, manual IHC with antibody H222, and automated IHC with two new antibodies CC4-5 and 6F11. In 29 cases of breast carcinoma, cytosol ER, manual MC with H222 and automated IHC with CC4-5 were compared. The cytosol ER, automated CC4-5, and manual H222 had complete agreement in 82% (14/17) cases. CC4-5 and H222 agreed in 95% (20/21) cases. CC4-5 correlated with H222 (p < 0.0001) and the cytosol assay (p < 0.0012). Automated IHC was performed on 35 cases using CC4-5 and 6F11. In 97% (34/35), concordant results were obtained. The proportion of cases with strong nuclear staining was significantly greater for 6F11 (53% of cases) than for CC4-5 (11%) (p = 0.0013). These results indicate that automated IHC can be used to assay for ER using these new monoclonal antibodies.