Recently, Wong et al. (Wong KA, Bano A, Rigaux A, Wang B, Bharadwaj B, Schurch S, Green F, Remmers JE, and Hasan SU, J Appl Physiol 85: 849-859, 1998) demonstrated that fetal lambs that have undergone vagal denervation prenatally do not establish adequate alveolar ventilation shortly after birth. In their study, however, vagal denervation was performed prenatally and the deleterious effects of vagal denervation on breathing patterns and gas exchange could have resulted from the prenatal actions of the neurotomy. To quantify the relative roles of pre- vs. postnatal vagal denervation on control of breathing, we studied 14 newborn lambs; 6 were sham operated, and 8 were vagally denervated below the origin of the recurrent laryngeal nerve. Postoperatively, all denervated animals became hypoxemic and seven of eight succumbed to respiratory failure. In vagally denervated lambs, expiratory time increased, whereas respiratory rate, minute ventilation, and lung compliance decreased compared with the sham-operated animals. In the early postoperative period, the frequency of augmented breaths was lower but gradually increased over time in the denervated vs. sham-operated group. The dynamic functional residual capacity was significantly higher than the passive functional residual capacity among the sham-operated group compared with the denervated group. No significant differences were observed in the prevalence of various sleep states and in the amount of total phospholipids or large- and small-aggregate surfactants between the two groups. We provide new evidence indicating that intrauterine actions of denervation are not required to explain the effects of vagal denervation on postnatal survival. Our data suggest that vagal input is critical in the maintenance of normal breathing patterns, end-expiratory lung volume, and gas exchange during the early neonatal period.