In testing several of these apparatus, Mr. Louis E. Emmes, of the Laboratory staff, has had excellent success in determining the oxygen consumption by using a modification of the method outlined in that he reads the spirometer after the subject has been connected with the air current. Noting the height to which the spirometer is raised at the end of a series of regular normal expirations, he starts a stop-watch. The experimental period then continues about 12 minutes, when again the height of the spirometer at the end of a series of 3 or 4 normal expirations is noted and recorded, and simultaneously the watch is stopped. Using the difference between the two readings as an index of the oxygen consumption, the usual calculations for temperature and pressure are then applied.

The obvious advantage of the Emmes method lies in the fact that while the goose feather over the opening in the 3-way valve gives a fair indication of the end of a normal expiration for determining the moment to begin the experimental period, this point is much better indicated by the actual excursions of the spirometer bell. Mr. Emmes thus uses the excursions of the bell to begin as well as to end the period. The two methods may be simultaneously applied and one period be made to include another. This involves only two additional readings of the height of the spirometer bell and two additional time records with a second stop-watch. The use of the modified method is recommended in all cases as a desirable check. A series of tests shows that a somewhat better agreement of duplicate periods may be secured with the Emmes method.

A still further check may be obtained by making the usual records before and after the valves are turned at the beginning and end of the period, and employing the Emmes method for two additional readings 10 or 20 seconds apart at the beginning of the period after the subject has been connected with the air current and again before he is disconnected at the end of the period. Two stop-watches are used for these intermediate records. Thus two sets of readings are secured for control within one 12 to 13 minute period, and in one 15-minute period it is possible to secure readings for three determinations of the oxygen consumption.

At the conclusion of a period, oxygen is again admitted, the initial position of the spirometer read, connections made with the subject as before, and a new period begun. All of this can be carried out without stopping the motor. If, during the first period, a larger amount of oxygen has been introduced than is actually consumed, a little more air can be rejected by turning the 3-way valve and lifting slightly the counterpoise. It is thus seen that the oxygen consumption may be approximately measured by the fall of the spirometer and the actual computation of this contraction in volume be completed by using the data obtained regarding the temperature and barometric pressure.