We tested the hypothesis that sensory and nonsensory factors associated with birth stimulate respiratory behavior in the fetal and newborn rat. Late gestation (Day 21) rat fetuses were externalized from the uterus with intact umbilical connections to the dam and exposed to stimuli normally associated with labor and delivery. Onset and maintenance of respiratory movements were monitored. In the first experiment, rat fetuses were exposed to either: (i) simulated uterine contractions; (ii) cooling (26 degrees C); (iii) umbilical cord occlusion; or (iv) air heated to intrauterine temperature (37.5 degrees C). Subjects were videotaped for 1 h, and respiratory movements counted during tape review. Fetuses in each group showed some respiratory behavior although compression significantly elevated respiratory rate compared to other experimental conditions. All subjects in each group were respiring after 1 h, except for pups that received umbilical cord occlusion. The 100% attrition rate of the cord occlusion-alone group was reversed by combining cord occlusion with compression, or with compression and cooling, but not by combining cord occlusion with cooling. Simulated birth pups (i.e., those exposed to compression, cooling and umbilical cord occlusion) and normal, vaginally delivered pups breathed at identical rates and showed a similar pattern of postpartum breathing. These results suggest that mechanical stimulation of the fetus associated with uterine contractions plays a critical role in the maintenance of continuous respiration at birth. Possible mechanisms for the facilitative effects of compression on perinatal breathing are discussed.