METHODS OF MEASUREMENT. 25 



were used with all the subjects to prevent possible leakage around the 

 mouthpiece, but later these were omitted with W. K. and E. D. B. 

 after they had become accustomed to the conditions. The normality 

 of long-continued breathing through the mouthpiece, especially under 

 the conditions of the grade-walking experiments, was tested in a series 

 of experiments. The results of these tests are discussed on page 177. 



The valve-operating device. The method of operating the valve which 

 connects the subject with the air-circuit is shown in figure 5, page 28. 

 An arm on the stem of the valve is connected by a cord F to a bar K, 

 which is supported on a button L. (See, also, fig. 1, p. 19.) When the 

 operator pushes forward the lever J, L is forced from under the bar K 

 and the tension of the spring M turns the valve. Resetting the button 

 L and shifting the cord to the other side of the arm k (see fig. 1) permit 

 the closing of the valve at the end of the period by the same method. 

 This arrangement allows the operator to stand behind the subject, who 

 thus has no knowledge as to when the period is to begin and end. The 

 proper operation of the apparatus requires that the valve should be 

 opened and closed at the end of a normal respiration, which is noted by 

 the movement of a bit of goose-down affixed to the outlet of the valve. 

 This material was not affectecl by the moisture of the breath, and, being 

 light, instantly showed the point of change in the current of air at the 

 end of an expiration. 



By-pass. As in other experiments made in this Laboratory on muscu- 

 lar work, 1 the ventilating air-current was carried to within a few cen- 

 timeters of the mouthpiece by means of a supplementary pipe C 

 (fig. 1), operated by a by-pass valve B. This by-pass valve was 

 opened by a hand lever a few seconds after the beginning of an 

 experimental period and closed a few seconds before the end of the 

 period. This prevented any possible difficulty in obtaining sufficient 

 ventilation for the subject and obviated an excessive dead space. 



Rate of ventilation. The rate of ventilation during all of the walking 

 experiments was the maximum capacity of the blower, namely, 100 

 liters per minute, while for the standing experiments a rate of 65 liters 

 per minute was generally maintained. 



TESTS OF THE UNIVERSAL RESPIRATION APPARATUS. 



Although the efficiency of the absorbing system of the universa 

 respiration apparatus has been thoroughly and repeatedly demon- 

 strated, it seemed advisable to test this point further before beginning 

 the research, since in the walking experiments a ventilating air-current 

 with a rate as high as 100 liters per minute was to be used in place of 

 the current of 30 to 40 liters per minute employed in most of the 

 researches with this apparatus. If the water-absorbers were inefficient, 

 moisture would be carried over to the carbon-dioxide absorbers and 



^Benedict and Murschhauser, Carnegie Inst. Wash. Pub. No. 231, 1915, p. 33. 



