The aim of the present study was to evaluate the physiological significance of the rapid, short-loop, negative feedback of prolactin by passive immunization with antiserum to rat prolactin injected into the third cerebral ventricle (3V) of conscious, freely moving intact or castrated male rats. Blood samples for measurement of plasma prolactin concentrations were removed through implanted external jugular catheters. After injection of 3 μl of undiluted antiserum, plasma levels of prolactin decreased rapidly (within 5 min) to values undetectable by RIA. Further study revealed that this dose of antiprolactin serum had combined with circulating prolactin, thus rendering it undetectable by RIA. To overcome this problem, we repeated the experiment injecting 2 μl of diluted antiserum (dilution factors 20, 100, 200, 2,000) into the 3V. When compared to values of plasma prolactin in control rats injected with 2 μl of normal rabbit serum, none of the dilutions of antiserum induced a significant change in prolactin concentrations for as long as 4 h after injection. Since there was no effect of intraventricularly injected antiprolactin serum on basal prolactin secretion, in the next experiment, intact as well as castrated male rats were subjected to ether stress 30 min after intraventricular injection of antiserum (dilution factor 100). The elevation of plasma prolactin which followed ether stress was significantly higher in male rats pretreated with antiprolactin serum than that which occurred in control rats. A similar enhancement of the increase in plasma prolactin following ether stress, but of longer duration, was obtained in castrated rats injected with antiprolactin serum. The results indicate that the rapid, negative short-loop feedback of prolactin attenuates stress-induced prolactin release. By contrast, there was no effect of the intraventricularly injected prolactin antiserum on plasma LH concentrations of the castrated males.