A critical evaluation of the measurement of the transthoracic basic impedance Zo according to Kubicek's method revealed inhomogeneous electrical field-distributions in the region of the neck and the shoulders and in the region of the mammae in female subjects. These inhornogeneties were found to introduce uncontrollable errors, like position dependency into the measurements. Also the sensitivity is deteriorated and the comparability between subjects is lost. A method for measuring thoracic impedances that avoids those disturbances is proposed, and some examples of registrations are presented. The measurements seemed to depend on thoracic fluid content, blood pressure, physiological variations and age. It was concluded that differential impedance plethysmography can be a useful tool in evaluating intra-thoracic processes, both in clinical care and physiological studies. The increased reliability might provide new insight into the determination of cardiac output by impedance cardiography.