The principal component analysis (PCA) is applied on juxtaposed average beats of unipolar and differential leads (DLs), obtained from 35-lead multichannel ECGs, for assessing proportions of cardiac electrical activity variance that the leads measure. It was shown that 93% of the total 35- lead measured variance can be explained by the first three principal components (PCs). An analysis of correlations between the ECG leads and thefirst three PCs shows that the leads employing electrodes further to the heart pick up less of the total ECG variance. It was also demonstrated that DLs with highest correlations to the first three PCs, coincide with the universal DLs selected by our previously published ECG synthesis algorithm: BDLSA. This enables the application of the PCA in the selection of optimal DLs combinations for the ECG synthesis.