SUMMARY1. Baroreflex activity was assessed in nine fetal, four new-born and six adult sheep, using the relationship between heart period and arterial pressure. Arterial pressure was raised either by inflating a balloon in the dorsal aorta, by rapid intravenous injection of phenylephrine or methoxamine, or by slow intravenous infusion of methoxamine.2. In the fetus the three methods gave different estimates of baroreflex sensitivity (balloon, 1-3 + 0 7 mmHg; injections, 5-4 + 0.5 msec mmHg; infusions, 7-2 + 0 9 msec/ mmHg) whereas they were-comparable in the new-born and adult. 3. Estimates of baroreflex sensitivity were significantly lower in the fetus and new-born than in the adult whichever method was used.4. In the fetus there were variable changes of heart period when arterial pressure was raised by inflation of the balloon. The responses to injection of phenylephrine or methoxamine were also variable.5. This variability was not associated with changes in electrocortical activity, the presence or absence of breathing movements or limb movements, or changes of blood gases.6. In the fetus the heart period frequently did not change unless the arterial pressure was raised by approximately 15 mmHg (to 61 mmHg), suggesting that the threshold for baroreflex activity is above the normal range of arterial pressures before birth.