The Fagerström Tolerance Questionnaire (TQ) is widely used in both clinical and research settings as a measure of physiological dependence on nicotine. In the light of claims that the efficacy of nicotine gum is related to degree of dependence and that outcomes can be improved by tailoring gum strength to degree of dependence, the availability of a simple, noninvasive test of dependence assumes additional importance. Recently, a number of articles have expressed reservations about various aspects of the TQ: 1) Several TQ items suffer from flaws or ambiguities that undermine their usefulness, and 2) the TQ may not measure what it purports to measure. The present paper reviews these issues, assembles data and observations that bear on the problems, and makes suggestions that may help in refining the TQ or producing a better instrument.