Summary By reviewing 1,000 electrocardiograms, configuration of QRS, ST and T, as well as QRS amplitude was compared between leads I, V6 and X, and between aVF and Y. Frontal QRS axis derived from leads I and aVF (conventional) was compared to that from X and Y (orthogonal). Morphologically, QRS in lead X resembled either one or both of leads I and V6 in 87.6%, while QRS in Y resembled that in aVF in 82.3% of cases. The net QRS amplitude and ST-T polarity showed good agreement between V6 and X, as well as between aVF and, Y, while their agreement was poorer between I and X. The frontal QRS axis in 12 leads showedthe following distribution normal (+90 to −30°), 87.1%; left axis: deviation (−30 to −180°), 7.7%; and right axis deviation (+90 to +180°), 5.2%. In orthogonal leads, the following criteria yielded similar incidence of these axis deviations: normal, between +60 and −15°; left axis, −15 to −180°; and right axis, +60 to +180°. The QRS axis from orthogonal leads showed lesser variation than that from 12 leads within the normal range, while it tended to show wider variations toward extreme left or right. When both lead systems showed abnormal axis deviation, possible clinical cause for such axis was found in 81–85% of cases, while when only one of the two systems showed abnormal axis, possible cause was found in only 50–60%. Hence, regarding abnormal axes, the conventional and orthogonal systems appear to supplement each other for a better sensitivity and specificity.