The influence of recording speed on apexcardiographic (ACG) measurements was determined by four observers who independently measured the timing of three standard ACG points (u, E and O) for five sub jects. Tracings were segmented according to speed, coded and randomly distributed for reading. Analysis of the data showed expected inter-subject variability. A significant observer difference occurred in the timing of u when measurements were compared over all four speeds (p < 0.05) but no significant observer differences in the timing of u could be demonstrated when each speed was considered separately. There were no observer differences nor significant observer interactions in the timing of E and O. Interspeed differences were highly significant (u: p < 0.05, E and O: p < 0.01) but the data showed no trend toward an optimal recording speed. Utilization of one speed, exclusively by all investigators recording the ACG is recommended to reduce variability in the timing of apexcardiographic points and, therefore, enhance intra- and interlaboratory comparisons.