In computing the results obtained with a brake ergometer, such as the Gärtner ergostat, still another method has been used for finding the net efficiency as it is customary to deduct from the total energy output during work the energy output required to turn the ergostat the same number of revolutions without external friction. Various other methods of computation have suggested themselves to different writers, with a consequent lack of uniformity in the base-lines used. The difficulties incidental to the computation of the results obtained during mountain-climbing and marching are also found in connection with our bicycle ergometer experiments, but, as has previously been pointed out, an attempt has been made to secure as many base lines as possible. Unquestionably some of these base-lines are wholly fictitious and their use in the computations is not justifiable. On the other hand, with all the possible combinations and conditions, it is more than probable that certain of these calculations do indicate truly the increased metabolism incidental to a measured amount of external muscular work. It is our purpose to recognize these various base-lines, and use them, so far as possible, in computing the results of our experiments.

Confining ourselves for the moment exclusively to the question of the method to be used in computing the results obtained in experiments with the bicycle ergometer, it can be seen that not only may we deduct the metabolism for the equivalent period when the subject is lying quietly in bed, but we may also deduct the metabolism incidental to sitting quietly upon the bicycle without rotating the pedals. This is an upright position that must be assumed by the subject before work can be begun and which calls for an increased metabolism over the lying position. The work incidental to this can not be measured on the ergometer; indeed, it is doubtful if there is any measurable external muscular work performed, although sustaining the body in an upright position does call for a considerable increase in internal muscular work.

Furthermore, in rotating the pedals, there is a movement of the legs with consequent muscular work which, for the want of better phraseology, may be termed the internal friction of the legs. A value may be obtained for this activity and used as a base-line by rotating the wheel of the bicycle by means of an electric motor, so that the legs are made to move up and down as in riding while no actual work is done upon the ergometer itself. Although the ergometer has but little friction, considerable internal work of the legmuscles is required in rotating the wheel of the ergometer, even when no current is passed through the magnet; hence in computing one of the various grades of efficiency the energy output incidental to rotating the ergometer under these conditions may properly be deducted from the total energy output, thus forming a third base-line.

Even though measurable, the friction of the ergometer is so small in amount that it would not justify a computation based upon the ground that overcoming the friction of the bicycle requires a given amount of external muscular work. The efficiency of the body in performing a very small amount of work may be studied, however, by using a slight measurable brake-effect upon the wheel. The amount of work thus done is so small in comparison with the total metabolism of the body that it may be classed with such minor muscular movements as the moving of a fan back and forth or rocking

in a rocking-chair. Such work does, however, bring the body-muscles into play, and theoretically, at least, forms an excellent base-line for the superimposition of more severe loads; by increasing the resistance of the ergometer step by step, various loads may be superimposed upon this base-line and very satisfactory computations made from the increments in work.

Finally, since many investigators have found that the efficiency varies with variations in the speed, the work may be done both at slow and at high speeds, all with the same resistance on the wheel. As a matter of fact, inasmuch as for comparison experiments wide variations in speed should be employed, with this particular form of ergometer there exists an inherent difficulty in carrying out such a study, for, as will be seen by the calibration curves, with wide differences in speed, there is a marked difference in the brakeeffect at any part of the curve. Nevertheless, experiments have been made on this basis and the results will be subsequently discussed.

COMPARISON OF BASAL METABOLISM IN BICYCLE ERGOMETER EXPERIMENTS. INCREMENT IN METABOLISM DUE TO CHANGING FROM A LYING TO A SITTING POSITION.

The influence of minor muscular movements and of changes in bodyposition on metabolism have received especial attention in the last few years. In a long series of experiments carried out in the laboratory of Wesleyan University, alterations in the body-position were studied by Benedict and Carpenter. Later a series of observations was carried out in the Nutrition Laboratory by Emmes and Riche," who noted a definite increase in metabolism on changing the subject from a lying to a sitting position, the increase being less marked when the sitting body was supported by pillows or otherwise. In one experiment made in this laboratory and still unpublished, there was no measurable increase in metabolism following a change to the sitting position, this result being probably due to the fact that the body, shoulders, arms, neck, and head were thoroughly supported by pillows. Nevertheless, in changing the position of the subject from lying on a couch to sitting on the bicycle seat, it is reasonable to expect considerable increases in metabolism, and as the values obtained in the latter position were likely to be used as a base-line for the computation of the net efficiency, it seemed desirable to secure evidence as to the increase in metabolism incidental to the change in position. Accordingly, such studies were made with four of our subjects including one experiment with each of three subjects and several experiments with the fourth subject. Nearly all of the observations were made with ergometer I. The results have been collected and presented in abstract in table 112, the carbon-dioxide production, oxygen consumption, and the pulse and respiration rates being recorded, as well as the increments in these factors in the form of per cent. Turning our attention particularly to the oxygen consumption per minute, it will be seen that in the experiments made with ergometer I the increments for all the subjects ranged from 6.1 per cent to 24.8 per cent, and, indeed, with the subject M. A. M., the ranges are nearly as great, i. e., from 6.1 to 23.1 per cent. On the other hand, with the two subjects E. P. C. and K. H. A., the increments are essentially the same, 8.2 and 9.1.

a Benedict and Carpenter, Carnegie Institution of Washington Publication No. 126, 1910, p. 243. Emmes and Riche, Am. Journ. Physiol., 1911, 27, p. 406.

TABLE 112.-Metabolism in experiments without food with subject lying on couch compared with metabolism in experiments with subject sitting inactive on ergometer. (Average results.)

1 Average pulse-rate for first two periods.

2 Aside from a preliminary experiment on Dec. 6, this was the first experiment with this subject.

3 In averaging, the oxygen consumption during the first two periods was omitted.

4 Sitting periods were run before lying periods on this day.

5 In averaging, the oxygen consumption during the first period has been omitted.

It is important to note here that the increased oxygen consumption incidental to sitting was accompanied by an increased pulse-rate. The increments in these two factors, it is true, were not invariably parallel, especially in the experiment with M. A. M. on December 11, in which the oxygen increment was but 12.2 per cent, while the increase in the pulse-rate was 40 per cent. Generally, however, the increment in the pulse-rate was roughly proportional to the increased oxygen consumption. The respirationrate, on the other hand, was not materially altered by changing from a lying to a sitting position. The record obtained on December 7 with M. A. M. is worthy of especial note; the low value for this subject of 12 respirations per minute, while sitting, is unquestionably due to an altered type of respiration, since the average value for a large number of experiments made with this subject throughout the experimental year was 20 respirations per minute. It should be said, however, that this was one of the earlier experiments; the subject was therefore unused to the apparatus and the routine, and consequently somewhat apprehensive. In general it may be inferred that there are practically no alterations in the respiration-rate due to changing from lying on a couch to the position of sitting on a bicycle seat.

Our attention was chiefly given to the subject M. A. M., and his personal observations on these tests are of unusual interest. This subject invariably complained that the experiments were long and tedious and that he found it irksome to sit still for so long a period. Although he was allowed to select the form of bicycle seat he preferred, as well as to adjust the angle, height, and position of both seat and handle bars, and every effort was made to secure for him the most comfortable position, yet reference to the protocols of the experiments shows that apparently sitting quietly on the ergometer was for him a great strain. As any position he might assume on the ergometer ul

timately became strained and uncomfortable, it was obvious that this subject was in a wholly abnormal condition throughout practically all of this class of experiments. With the subject M. A. M. no foot-rests were used, the feet being placed upon the pedals and allowed to remain there. With the other subjects, J. E. F., K. H. A., and E. P. C., a foot-rest was usually placed on the front of the machine or else a wooden block was placed upon the floor on which the feet could rest. J. E. F. also found the sitting position very tiresome and accompanied by strain.

It will be seen, therefore, that there is a complete lack of harmony in the results obtained from experiments in which the subject changed from a lying to a sitting position. It was particularly unfortunate that with ergometer II, with which the largest number of work experiments was made, the values for sitting indicate clearly that the discomfort experienced by the professional subject unquestionably increased his metabolism, a fact which must be taken into consideration when this abnormally high base-line is used in any way.

INCREMENT IN METABOLISM DUE TO A CHANGE IN POSITION FROM LYING ON A COUCH TO SITTING On a Motor-DRIVEN ERGOMETER.

In certain experiments, when the subject was sitting on the bicycle ergometer with the feet on the pedals and the body in a riding position, the pedals were rotated by means of an electric motor belted to the rear wheel. Under these conditions the legs moved up and down in the usual manner, but no external muscular work was performed," the legs being raised and lowered by means of the electric motor; the work done, therefore, was essentially that of the internal friction of the leg-muscles. Five of these experiments were made with slightly varying rates of speed; the results are given in table 113, together with the results for the comparison experiments made on the same days in which the subject lay quietly upon a couch. In one instance, namely, March 8, the average value for the whole experimental year was used for comparison, as no couch experiment preceded the motor-driven ergometer experiment for that day. By deducting the metabolism of the subject when lying upon the couch from that during the ergometer experiment, the increment due to the muscular activity could be easily obtained. The increases found for the carbon dioxide and the oxygen are essentially alike, the increases in the oxygen consumption ranging from 162 to 207 per cent according to the speed. The pulse-rate increments were likewise considerable, ranging from 42 per cent to 51 per cent. As in the series of experiments given in table 112, the respiration-rate showed but slight alteration. From the results of these experiments it will be seen that when the subject rode on an ergometer with the wheel rotated by a motor, the metabolism was practically 3 times that obtained when he lay on a couch.

In connection with these motor-driven ergometer experiments many interesting observations were made regarding the muscular control of the subject. Throughout the entire series he objected to them, stating that it was hard for him to relax completely and allow his legs to swing idly and be rotated by the motor. From a consideration of the construction of the apparatus it can be seen that there should always be a tension on the sprocket

• See Berg, du Bois-Reymond and L. Zuntz for a discussion of this motion. Archiv f. Anat. u. Physiol., Physiol. Abth. Suppl., 1904, p. 42.

chain, with the lower part taut. With a steady load or steady resistance on the pedals, the pull on the lower part of the chain would be constant. It was noticed, however, that the lower part was frequently slack and that the upper part was taut, showing that the subject at times involuntarily overcame the inertia of the machine and perhaps contributed slightly to the

(Subject

TABLE 113.—Metabolism experiments without food with subject lying on a couch compared with the metabolism in no-load experiments with motor-driven ergometer. M. A. M.; ergometer II.)

1 There was but one record of pulse in cach no-load period on this day.

2 This figure is a general average of results with this subject while lying on couch before work in respiration experiments without food. (See table 91.)

movement of the pedals. While the results are reasonably uniform, the five experiments being made inside of a month, they doubtless represent a larger amount of energy output than would normally be expected from the simple friction of the leg. Unquestionably there were slight involuntary compensatory movements of the legs, tending to oppose the work of the motor at one time and again tending to assist it, and this interplay of muscles may have caused a not inconsiderable proportion of the total increase in the metabolism. While at first sight it would appear as if in such experiments equilibrium was obtained by the balancing of one leg against the other as it went down, the fact that the legs are not in equilibrium has been interestingly brought out by Berg, du Bois-Reymond, and L. Zuntz."

Unfortunately, the difficulties incidental to the calibration of the ergometer by means of an electric motor belted to the rear wheel made it likewise impossible to measure exactly the amount of energy required to drive the motor when the legs of the subject were moved freely by the motion of the pedals. We believe it especially significant that the professional subject M. A. M. found these experiments difficult; consequently, while the results obtained with the motor-driven ergometer have been computed and used as a base-line, we seriously question whether these values are not considerably higher than they should be. If they are too high, the use of them as a baseline in computing the increment due to severe work would obviously lead to erroneous results.

INCREMENT IN METABOLISM DUE TO A CHANGE IN POSITION FROM LYING ON A COUCH TO RIDING ON AN ERGOMETER WITH NO LOAD AND NO MOTOR.

In these experiments the subject sat upon the ergometer in the usual riding position, and rotated the pedals by the feet at varying speeds, the only

a Berg, du Bois-Reymond, and L. Zuntz, loc. cit.

b See p. 26 of this report.