Changes in transthoracic electrical impedance (T.E.I.) due to high-altitude hypoxia (3,658 m) have been measured in 20 young, healthy Indian soldiers. They were first studied at sea level (198 m) and then rapidly transported by air to 3,658 m, where they were studied daily from day 1 to day 5 and then on days 8 and 10. The mean (+/-S.D.) T.E.I. at sea level (34.6+/-0.6Omega) fell sharply to 29.6+/-0.8Omega, 30.3+/-0.9Omega, and 30.5+/-1.1Omega on days 1, 2, and 3 (P <0.001) and levelled off at 31.5+/-0.7Omega on day 10, which was comparable to the mean value obtained in 13 persons permanently resident at high altitude (32.2+/-0.7Omega). Five sea-level residents who had acute mountain sickness (A.M.S.) or high-altitude pulmonary oedema (H.A.P.O.) had a still lower mean value (22.5+/-1.1Omega). One normal healthy subject who at sea level had a T.E.I. of 34.7Omega developed H.A.P.O. when the T.E.I. fell to 21.1Omega. Ninety minutes after the administration of 80 mg of intravenous frusemide the value increased to 35.5Omega. In another subject with A.M.S. who received 40 mg of frusemide intravenously the T.E.I. rose from 21.9 to 33.2Omega.Since the study was non-invasive the changes in impedance could not be correlated objectively with alterations in either pulmonary blood volume or pulmonary extravascular water space. In the subject, however, with x-ray evidence of H.A.P.O. and a low T.E.I. intravenous frusemide produced a marked rise in T.E.I. together with clearing of the chest x-ray picture within 24 hours, indicating an inverse relationship between impedance and thoracic fluid volume. It is suggested that with further objective verification in man the measurement of T.E.I. may be a potentially promising technique for the early detection of increased pulmonary fluid volume.