The detection of premalignant lesions of the oral mucosa allows for treatment that may be sufficiently early to prevent their progression to invasive carcinoma. Although a number of techniques for supplementing clinical examination have been described, on the whole they rely on laboratory-based investigations with variable false-negative rates. Toluidine blue vital staining, on the other hand, has been advocated as a simple, inexpensive, and sensitive chairside test. Most of the clinical series investigating its efficacy, however, have relied on clinically abnormal mucosa or tissue that stained. For the purpose of quantifying the true sensitivity of toluidine blue, the present study involved the examination of clinically normal mucosa as well as abnormal tissue and compared it with the presence or absence of staining. With this technique the false-negative rates of 42% and 58% obtained for carcinoma-in-situ and mild-to-moderate dysplasia, respectively, suggest restricting the use of vital staining to selective cases.