The ability for adaptive increase in heart rate following exercise is of prime importance to the organism. In normal persons the circulation adjusts itself very rapidly to the muscular demand. The cardio-inhibitory center is very sensitive. Investigators have observed that with subjects sitting quietly the pulse-rate changes in the next one or two cycles following the movement of the arm or any slight change in the bodily position. The quickness with which the change occurs from the resting pulse to the rate for exercise, provided the exercise begins promptly, is a measure of vagus tone. In certain physiological conditions, notably that of heart block, the adaptive increase in heart rate during exercise will not take place. It is stated by Wiggers1 that this is even more characteristic of incomplete heart block.

Since it therefore happens that with many individuals showing a slow radial pulse-rate, the circulation is adequate for conditions of muscular repose but decidedly inefficient during exertion, and since the subjects in the low-diet research began somewhat early to exhibit radial pulse-rates which were notably below those of normal, it appeared to us particularly important to examine the change in heart rate with exertion. An accurate and at the same time the most convenient method of securing records which will provide information of this sort is to take electrocardiograms by the technique described on page 152, i. e., with body electrodes connecting the subject to the string galvanometer for the taking of continuous records while he is quiet, active, and again at rest. The sample records, which are illustrated in figure 24 (see page 152), demonstrate clearly that any considerable irregularity in the rhythm or conduction of the electrocardiogram under these conditions could be readily discovered, except perhaps during the actual moments of exertion, when only the prominent R deflection is legible in the tracing. The auricular wave P is usually very small, partly because of the capacity and resistance in series in the circuit reducing the amplitude of the deflections. The P wave is, however, usually indicated in the quiet pulse prior to exertion. It also becomes visible in the latter part of the recuperation period, and since the ventricular complex (R and T waves) appears with increased frequency between the two points in the record where the P wave is not evident, it is probable that all of the impulses are of sinus origin and proceed in the normal way. No stress can be laid upon the shape of the waves. Only their order and frequency concern us here.

1 Wiggers, Circulation in health and disease, Philadelphia, 1915, p. 279.

To secure an accurate measure of the length of the pulse cycle, we measured from the sharp point R to R.1 A table for all the pulse-cycle data shown in these records would be very large and unwieldy. Data of this character for a series of similar electrocardiograms taken on one subject have been published elsewhere.2 Each individual pulse cycle (R to R distance) is measured, the unit being 0.01 second. In discussing such pulse changes, it is more logical to use pulse-cycle length than pulse-rate per minute. The two statements must not be confused. The data can be conveniently presented in the form of curves. An illustrative set of curves for an individual subject is given in figure 92. Records were taken for Mon on the five days, October 28, November 11, December 9, January 13, and January 27. In order to avoid a confused diagram, curves for only three dates are given, those for November 11, December 9, and January 27. In each case the curve is the average for two similar and consecutive records, separated by one minute or more of rest. The portion of the curve to the left of the heavy vertical line represents the period of quiet rest in the steamer chair. The pulse-cycle length varies somewhat, as is to be expected in any normal individual, due to the respiratory changes and other influences on the vagus. No signal or indication was given to the subject until the exact moment when he was to grasp the bar and "chin" himself, the beginning of which event is represented by the vertical heavy-line. Immediately at the beginning of exercise the cycle length is shortened and continues to decrease during the 5 seconds of muscular tension, the end of which is indicated in each curve by a short vertical line.

1 Miles. Carnegie Inat. Weah. Pub. No. 266, 1918, p. 95. 2 Miles. ibid., p. 98. table 17.

The curve for December 9 in figure 92 is at a higher level - i. e., a faster pulse-rate - particularly in the period of quiet and of rest. It is also to be observed in the case of this curve that at the end of the exertion the pulse continues to rise during about 5 cycles, after which it shows a steep decline. December 9 was not long after the Thanksgiving vacation, and the pulse-rate had risen on this date to more nearly its normal level. On November 11 and January 27 the reduced diet had been in force for several weeks continuously in each case and the pulse was at a lower level. The difference in level between December 9 and the other two dates is particularly prominent in the quiet pulse, that is, preceding the activity. It is less prominent in the rest pulse following the activity and rather slight during the activity. It is significant that the cycle length during exercise is so nearly the same, even though the resting pulse is at rather widely different levels. It would appear that in the case of this individual the given amount of exercise required a pulse-rate of about a certain level, and that in the case of the lower resting rate the tone of the vagus is higher or is somewhat more affected at the time of the exercise.

Fig. 92. - Changes in the pulse-cycle duration with exertion, Kirk G. Montague.