Studies on surgery for temporal lobe epilepsy are hampered by lack of information about the actual surgery that has taken place. A method is described for accurately measuring the volumes of resection by MRI after surgery. Ten cases of surgically treated temporal lobe epilepsy (nine non-tailored resections, one selective amygdalohippocampectomy) are presented to show the technique. Indices of extent of resection in both the mesiobasal and lateral temporal lobe compartments have been measured, compared, and evaluated. By comparison with identical preoperative volumetric MRI the hippocampal resections have been correlated with the demonstrated hippocampal volume loss, thought to be of relevance in the aetiology of temporal lobe epilepsy. Detailed postoperative audit in this manner is vital in providing a rational basis for follow up studies of outcome.