AN ANALYSIS OF THE TIME‐RELATIONS OF ELECTROCARDIOGRAMS.

IN a preliminary attempt (which requires considerable modification) to determine from blood-pressure records the relative influence of the heart action and of vaso-canstriction, I suggestedS that it might be necessary to estimate the duration of ventricular systole for different heart rates. In order to obtain this information a number of measurements have been made of electrocardiographic curves, including some obtained by myself and a selection of curves from Dr. T. Lewis’s collection, which he very kindly put at my disposal. Electrical records have been preferred to mechanical, because it is easier to secure accuracy, and it has been shown by many workers that as a rub the electrical and mechanical changes correspond fairly closely. Lewis ,*7 in a comparison of the heart sounds with the electrical changes, found the first sound to commence 0.011 of a second to 0.039 of a second after the commencement of Q, while the second sound started either before or after the end of T but usually within 0.01 of a second of it. WiggersYS1 working with dogs, found the mechanical systole to commence 0.03 to 0.045 after the rise of R, and to terminate 0.034 to 0.048 after the end of T, so that as a rule the two changes corresponded in duration, but he found that adrenalin shortens the duration of the mechanical change more than the electrical, and under these conditions the ventrical contrsction ended before the end of the T wave. In considering, therefore, the relative duration of systole and diastole, both electrical and mechanical records are useful, if these differences be allowed for. Walleflsgivee the following values for the durebtion of mechanical systole with different heart rates, and it will be seen that almost exactly similar figures are obtained by calculation from the formula systole = K Vcycle, where K hae 8 value of 0.343.