Seventeen normal, healthy adults were trained to discriminate between orally administered d-amphetamine (AMP; 10 mg) and placebo. Standardized subjective effects questionnaires were used to examine the relationship between the subjective and discriminative stimulus effects of AMP. Seven of the subjects were able to learn the discrimination reliably. These seven "discriminators" did not differ from the ten "nondiscriminators" in their subjective ratings of mood in the absence of drug. Discriminators were generally more sensitive than nondiscriminators to the subjective effects of AMP, although this difference in sensitivity reached statistical significance only for ratings of "hungry." Stimulus substitution was tested in the the discriminators with other doses of AMP (2.5 and 5 mg) and with 10 mg diazepam. The discriminative stimulus properties of AMP were dose-dependent, with 5 mg being the threshold dose. In five of the seven subjects the discriminative stimulus properties of diazepam did not substitute for those of AMP. The results demonstrate that the experimental paradigm can be used successfully to study the discriminative stimulus properties of drugs directly in humans.