1. In dogs that were supported in the upright position and that were breathing spontaneously acute pulmonary oedema was induced either by extracellular fluid volume expansion (n = 5) or by increasing pulmonary microvascular permeability with alloxan (n = 5).
2. Before oedema was induced, the bronchi of the right lung were outlined with tantalum powder.
3. During a baseline period, standard chest radiographs were taken at inflation pressures from 0 to 2 kPa and the distributions of perfusion (with radioactive microspheres) and ventilation (with 133Xe) were measured. Arterial and mixed venous blood samples were taken to estimate the degree of venous admixture (physiological shunt) and pulmonary arterial and wedge pressures were measured.
4. After oedema had developed the radiographs and observations were repeated at a time when the pulmonary vascular pressures were insignificantly different from the baseline state.
5. With induction of oedema, particularly when due to volume expansion, the lower-lung zones decrease in volume. There were no significant changes in the upper zones.
6. Abnormal venous admixture occurred only in dogs with >20% loss of lower-zone volume; volume loss and physiological shunt were significantly correlated.
7. The distribution of perfusion changed little with induction of oedema. Volume-expanded dogs showed a slight diversion of perfusion away from the bases and towards the upper zones.
8. There was an approximately 30% reduction of ventilation to the lung bases and a corresponding increase to the upper zones.
9. With induction of oedema, bronchi were narrowed when there was a reduction of lung volume. There was a significant linear correlation between volume change and narrowing.