The McGill Pain Questionnaire (MPQ) was administered to 102 'toothache' patients to determine whether it was sufficiently sensitive to distinguish between dental patients whose pain was clinically diagnosed as originating from a reversibly inflamed tooth pulp (group I) and those whose pain was diagnosed as originating from an irreversibly inflamed or necrotic pulp (group II). Scores for Total Pain Rank Index (PRI(T)), Sensory Pain Rank Index (PRI(S)), Evaluative Pain Rank Index (PRI(E)), Miscellaneous Pain Rank Index (PRI(M)), and Number of Words Chosen (NWC) were significantly higher (p less than 0.05) for group II patients. The PRI differences between both groups were attributed mainly to the more frequent selection by group II patients of 8 of the 20 subclasses of words and/or of words with higher rank values within the 8 subclasses. A significantly greater degree of sleep disturbance, nausea, headache, drowsiness and/or dizziness was also found in group II patients. Discriminant analysis using the 20 subclasses and 4 supplementary questions related to sleep disturbance, changes in food intake or activity levels, and accompanying symptoms, indicated that the MPQ, when used alone, correctly predicted diagnosis and treatment outcome in 73% of patients. Therefore, our findings indicate that the MPQ can distinguish between the two types of toothache and suggest that, especially when used along with other standard diagnostic tests, it may be a useful clinical adjunct in the diagnosis of dental pain.