The present study investigated physiological and pharmacological characteristics of socially "stressed" animals. Specifically, we examined (1) to what degree autonomic and behavioral "stress" reactions during intermittent confrontations between an intruder male adult Long-Evans rat with an aggressive resident undergo habituation, and (2) to what extent the defeat-experienced animal can be protected against these "stress" reactions with clonidine or metoprolol, two adrenergic agents with clinical anxiolytic effects. We developed an acute social stress situation that consisted of initially placing an experimental rat as an intruder into the homecage of a resident while the resident was not present, thereafter permitting brief physical agonistic interactions with the reintroduced resident until the intruder was forced into a submissive supine posture and emitted ultrasonic vocalizations (USV), and eventually exposing the intruder to the resident's threats for one hour, while being shielded from potentially injurious attacks ("threat encounter"). Over the course of the initial 4-weekly threat encounters the acute tachycardia but not the hyperthermic stress responses decreased in magnitude. Following the first three threat encounters core temperature (Tc) was significantly elevated for at least 3 h. The Tc was already elevated when the repeatedly defeated intruder was confronted with the olfactory cues of the resident's cage. This conditioned "anticipatory" hyperthermia developed in the course of the first three confrontations and was paralleled by a decrease in exploratory and motor behavior and by an increase in defensive behaviors and in both types of USV emitted in the "low" (20-30 kHz) and the "high" (31-70 kHz) frequency range. Clonidine (0.01-0.1 mg/kg, IP), an alpha 2-adrenergic agonist and metoprolol, a beta-adrenergic blocker (1.0-10.0 mg/kg, IP), dose-dependently prevented the tachycardic response to stress. Only clonidine, but not metoprolol, also attenuated the rise in Tc during the 1-h agonistic interaction. Clonidine decreased those aspects of motor behavior (e.g. rearing, walking) that are of lesser "cost" for the individual but maintained high levels of defensive reactions and increased the duration of "low" USV. The high doses of clonidine (0.06, 0.1 mg/kg) attenuated the homeostatic regulation and sedated the intruder while exposed to threats during a social confrontation. The absence of attenuation of the high level of defensive behavior and the prolonged "low" USV suggest a stress intensification by the higher doses of clonidine. In conclusion, after the fourth encounter, the autonomic, behavioral and vocal response pattern prior to and during repeated weekly confrontations show no evidence for habituation for the following 6 weeks.(ABSTRACT TRUNCATED AT 400 WORDS)