Lungs from near-term fetal guinea-pigs were supported in vitro for 3 h; lung liquid production was measured by a dye-dilution method using Blue Dextran 2000 (fetuses 62 +/- 2 days of gestation, 97.6 +/- 19.0 (SD) g body weight; n = 134). Untreated control preparations produced fluid at 1.30 +/- 0.22 ml/kg body weight per h, and showed no significant changes during incubation (n = 30). After 1 h of incubation, experimental lungs were expanded with Krebs-Henseleit saline in volumes estimated to be below or approximating those of the first breath (n = 30; first breath, 0.6-1.2 ml). Expansions were graded at 18 +/- 4%, 31 +/- 4%, 43 +/- 3%, 50 +/- 3% and 72 +/- 2% of lung volume (volumes used for expansion at the maximal level, 0.64 +/- 0.25 ml). All expansions of 31% or above produced reductions in fluid production significant by analysis of variance (P < 0.01-0.001); production halted at 50% expansion, and there was strong reabsorption at 72% expansion (-0.87 +/- 0.45 ml/kg body weight per h by the final hour). There was an exponential relationship between percentage expansion and percentage fall in production (r = 1.00). There was no evidence for excessive pressure, and no evidence for lung damage as judged by electron microscopy or entry of intracellular materials into the fluid (lactic dehydrogenase, protein, K+). In studies based on 36 preparations, 10(-6) M amiloride present in the lung lumen (apically) abolished the reabsorptions seen at 70 +/- 3% expansion, but not the arrest of production; it had no effect on control preparations. Studies based on 24 preparations showed that responses to 72 +/- 2% expansion were not affected by 10(-7) M propranolol placed in the outer saline. In studies of 14 fetuses of widely different body weights (68.3-124.9 g), responses to 74 +/- 2% expansion showed an exponential increase with increasing body weight (r = 0.96). Although caution is needed, the results suggest that expansion of the lungs at birth may induce fluid reabsorption by an action independent of tissues outside the lungs, probably involving both activation of a Na(+)-based reabsorptive system and arrest of production, but not requiring beta-receptor activation. The probability that the responses are maximal at birth is discussed, and it is suggested that the effect of expansion may be a specialization of the perinatal lung.