Theoretical and experimental model studies on the effects of intracardiac blood have indicated that voltages due to a radial dipole are smaller than have been found in in-vivo experiments. The possible reasons for these differences have been examined. It is concluded that the lungs may tend to increase voltages over the anterior and posterior thoracic walls because of a current-channeling effect. This effect depends on the direction of cardiac excitation relative to the lungs. Voltages on the lateral and postero-lateral sides of the thorax are reduced because of the relatively large mass of the lungs in these regions. Changes in intracardiac blood volume for a constant hematocrit have less effect than changes in hematocrit per se.