A local cutaneous injury can produce primary hyperalgesia within the injured area and secondary hyperalgesia in the normal surrounding skin. An intradermal injection of capsaicin in humans causes intense pain and hyperalgesia to heat and to mechanical stimuli in the surrounding skin. Psychophysical studies in humans supported the conclusions that the hyperalgesia was predominantly the secondary type and depended on one set of neurons sensitizing another ('neurogenic hyperalgesia') and that the latter set of neurons is located in the central and not the peripheral nervous system. To further test this hypothesis and to search for peripheral neural mechanisms contributing to the pain and neurogenic hyperalgesia from a local injury, we performed neurophysiological experiments in the monkey (Macaca fascicularis) and recorded the responses of cutaneous primary afferent fibers to an intradermal injection of capsaicin and to mechanical and heat stimuli delivered before and after the injection. Most C- and A-fiber mechanoheat-sensitive nociceptive afferent fibers (CMHs and AMHs, respectively) responded too weakly or transiently to capsaicin to account quantitatively for the magnitude of capsaicin pain. Of the known primary afferents tested with capsaicin injections, only the responses of heat-selective nociceptors could potentially account for the pain measured psychophysically in the human. In addition, a novel type of primary afferent-tentatively termed 'chemonociceptive'-may have contributed as well. Nociceptive fibers did not become sensitized to either mechanical or heat stimulation after an injection of capsaicin either outside, adjacent to, or inside the receptive field (RF); any changes that occurred could not explain the hyperalgesia to mechanical or heat stimuli observed in humans. The depressed responsiveness ('desensitization') of both myelinated and unmyelinated nociceptive fibers in the monkey to heat and/or mechanical stimulation of the injection site after capsaicin was injected inside their RFs correlated with the analgesia observed at the capsaicin injection site in the human. Capsaicin, topically applied to the RF in a vehicle of dimethyl sulfoxide or alcohol, excited CMHs and AMHs and enhanced the responses of some of these fibers to heat and/or to stroking the skin. In some cases, similar results were produced by the vehicle alone. However, capsaicin and not the vehicle lowered the thresholds of some CMHs to heat. Thus the sensitization of CMHs contributes to the primary hyperalgesia known to occur within the area of skin directly exposed to topically applied capsaicin. We conclude that 1) the desensitization of CMHs and AMHs contributes to the cutaneous analgesia produced at the site of injection of capsaicin and 2) the prolonged neurogenic hyperalgesia to heating and mechanical stimulation of the skin surrounding an intradermal injection of capsaicin is not accounted for by the sensitization of peripheral nociceptors. It is hypothesized that the sensitized neurons contributing to neurogenic hyperalgesia are located in the CNS.