The resting 12 lead electrocardiogram and vectocardiogram were reviewed in 47 patients with the Wolff-Parkinson-White syndrome (a) who had pre-excitation on the resting 12 lead electrocardiogram, (b) who had a single anterograde conducting accessory pathway assessed and located during preoperative electrophysiological study and during epicardial mapping at operation, and (c) in whom surgical division of the accessory pathway resulted in loss of pre-excitation. The site of the accessory pathway established during operation was compared with that established by evaluating the polarity of the delta wave and QRS complex on the resting 12 lead electrocardiogram. The electrocardiogram was assessed by the Rosenbaum criteria (Wolff-Parkinson-White type A, left-sided pathway; or type B, right-sided pathway), the Gallagher criteria (atrial pacing resulting in maximal pre-excitation), and the World Health Organisation criteria (a composite of previous studies). The Gallagher and World Health Organisation criteria were derived from patients demonstrating maximal pre-excitation that often required atrial pacing. The present study was designed to determine whether these criteria could be accurately applied to the resting 12 lead electrocardiogram on which the degree of pre-excitation was variable. The Rosenbaum criteria correctly identified a left sided accessory pathway in 26 of 34 patients and a right-sided accessory pathway in nine of 13 patients. The Gallagher and World Health Organisation criteria correctly identified the location in only 15 (32%) of the 47 patients. The resting vectorcardiogram was inaccurate for locating the accessory pathway. Although published criteria are useful for identifying the site of the accessory pathway from an electrocardiogram obtained when rapid atrial pacing is being used to achieve maximal pre-excitation, they are not suitable for identifying the exact site of an accessory pathway from the resting 12 lead electrocardiogram.