The orofacial formalin test in the rat is a valid and reliable model of nociception and is sensitive to various classes of analgesic drugs. The noxious stimulus consists in an injection of diluted formalin (2.5% in saline) into the upper lip. The behavioural nociceptive response is measured in terms of the amount of time the animal spends rubbing the injected area. Two distinct periods of intense rubbing activity can be identified, a first phase occurring in the first 3 min and a second phase lasting from 12 to 39 min after formalin injection. The present study verified the peripheral origin of the first phase of the formalin response and examined whether the second phase is produced by peripheral activation of afferent fibres and/or by a phenomenon of central facilitation induced by the neural activity that occurs during the first phase. This was determined by assessing the effect of a local anaesthetic agent (lidocaine) administered into the formalin injection site, before or after the first phase of the formalin response. Local injection of 50 microliters of lidocaine prior to formalin completely abolished the first phase of the formalin response but this blockade did not significantly influence the appearance and development of the second phase. Thus, the primary afferent activity that normally occurs during the first phase of the formalin response is not a prerequisite for the expression of the second phase. A higher dose of lidocaine (150 microliters) induced, in addition, inhibition of the first part of the second phase.(ABSTRACT TRUNCATED AT 250 WORDS)