The purpose of this investigation was to determine the specificity of hypergraphia to temporal lobe epilepsy using the paradigm developed by Sachdev and Waxman (1981). One hundred and thirty-eight patients were sent a standard stimulus letter and the response rates were examined for the patient groups with temporal lobe epilepsy (N = 80), and without (N = 31), and mixed seizure types (N = 13). For those patients who responded, additional analyses involved the number of words per response and the presence/absence of any qualitative indices of the postulated interictal behavioural syndrome of temporal lobe epilepsy. The results were interpreted in the light of the contemporary definitions of hypergraphia and were related to the larger literature concerned with personality and behavioural change in temporal lobe epilepsy.