Loss of alveolar barrier function is important in the development of pulmonary edema, but quantitation of its integrity has been difficult in the intact lung. We report a new non-radioactive method to assess paracellular and transcellular permeability of alveolar barrier in buffer-perfused rabbit lungs. Changes in alveolar barrier parameters were then correlated with different types of lung edema formation. The paracellular and transcellular barrier function was quantified by calculating the apparent epithelial permeability-surface area products (PS) for a fluorescent hydrophilic solute, FITC-dextran (FD-4), and a hydrophobic solute, rhodamine B, respectively. In control lungs, the apparent epithelial PS for FD-4 and rhodamine B were 0.7+/-0.05 x 10(-4) and 40.0+/-4.1 x 10(-4) ml/sec, respectively. The apparent epithelial PS of FD-4 could be increased 25-fold by inhibition of epithelial anion exchange with 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) without affecting the PS of rhodamine B. The apparent epithelial PS of FD-4 could be increased 6- and 1.7-fold by disrupting microtubules with nocodazole and colchicine respectively, but microtubule agents decreased PS for rhodamine B. A pattern similar was produced when ATP production in the lung was inhibited by 2-deoxyglucose or when oxidative injury was induced by ischemia-reperfusion. Neither DIDS nor nocodazole altered endothelial permeability to albumin. DIDS, but not nocodazole, increased transcapillary liquid filtration and calculated interstitial compliance of the lung during hydrostatic challenge. We conclude that epithelial permeability in the intact lung can be assessed using fluorescent solutes, and that increased permeation of hydrophilic solutes may enhance lung edema formation.