The effects of large ramp and hold stretch of the gastrocnemius muscles were studied on dorsal horn nociceptive-responsive neurons (DHN) in decerebrated cats. The stretch consisted of 3 subsequent plateaus, ending with a release phase from a near maximal physiological length of the muscle. This was done in order to mimic the "therapeutical stretches" used in clinical settings to alleviate muscle tenderness and stiffness. Stretch was applied before and after i.a. injection of bradykinin (50 microg.) into the gastrocnemius muscles. Changes in DHN activity were extracellularly recorded with high impedance glass microelectrodes. Only neurons up to approximately 1 mm were recorded. Neurons were identified as nociceptive on the basis of their response to an identification procedure applied on beforehand. Responsiveness of the DHN to stretch was variable. If responsive, a typical behavior was an increased firing rate during the ramp phases, followed by a gradual decrease during the hold phases and a reduction in firing rate or a complete silence during and after the release phase, usually lasting up to 50 s. Bradykinin injections induced either excitation or inhibition of DHN.. Stretch applied directly after the bradykinin injection in most cases elicited changes in the bradykinin-induced response of the DHN, changing both response profile as well as its strength. Results indicate that nociceptive-responding DHN could be influenced by innocuous ramp and hold stretch both in conjunction with and without chemically induced activation. The behavior of the DHN to stretch in the presence of the painful stimuli may be one of the possible mechanisms that could explain the short term relieving effects of muscle stretching when applied on stiff and tender muscles.