Although exercise-induced changes in electrocardiographic R-wave amplitude have been ascribed to changes in left ventricular (LV) size, QRS axis, heart rate and ischemia, the physiologic mechanism remains unclear. To clarify the relation between R-wave amplitude and changes in LV size and position, simultaneous 9-lead electrocardiograms and targeted M-mode echocardiograms were recorded from 15 normal subjects. Recordings were made at rest, during Valsalva maneuver and during methoxamine infusion. LV diastolic dimension increased with methoxamine and decreased with Valsalva maneuver (p less than 0.001). R-wave amplitude in leads V5 and V6 varied directly with LV dimensions (p less than 0.001). The correlation coefficient between the change in R-wave amplitude in V5 or V6 and the change in LV dimension was 0.81 (p less than 0.01). No significant changes in R-wave amplitude were seen in electrocardiographic leads I, II, III, aVR, aVL, aVF or V1. Distance from the chest wall to the LV posterior wall correlated with change in R-wave amplitude (r = 0.79, p less than 0.001). Change from supine to left lateral position moved the left ventricle closer to the lateral chest wall in association with a 41 +/- 8% increase in R-wave amplitude in V5 and V6 (p less than 0.001). In conclusion, there is a direct and a dynamic relation between R-wave amplitude and LV chamber size. Chamber size and distance from the left ventricle to leads V5 or V6 interact as major determinants of R-wave amplitude.