TECHNIQUE FOR METABOLISM DURING WORK. 121 



enough to contain a treadmill and a subject walking on it. Thus all 

 objections to the use of the mouthpiece and nosepieces would be avoided 

 and the subjects would have much greater freedom of head and body 

 movements while walking than when joined to a circulating air system 

 by a mouthpiece. In fact, the subjects were as free as when walking in 

 the open room. The temperature within the chamber could be con- 

 trolled by cooling the air of the room by open windows and electric fans 

 and the humidity kept at a low point by circulating the air of the chamber 

 through a drying system. Under these conditions the gaseous ex- 

 change could be computed from the volume of the chamber, the percent- 

 age increase of the carbon dioxide, and the decrease in the oxygen in the 

 air of the chamber during the period of walking. 



It was also considered possible by this method to make one 20- 

 minute walking period suffice for each subject. By taking the samples 

 of air for analysis in duplicate the possibility of error from this source 

 would be shght. As a further check, however, it was planned to draw 

 air samples for carbon-dioxide determinations at 10-minute intervals, 

 thus dividing the walking period into two 10-minute periods, which 

 should show agreement in the carbon dioxide present. 



The walking during the experiments was done on a treadmill which is 

 described briefly on page 126, and more in detail in another publica- 

 tion.^ This treadmill was placed inside the specially constructed 

 chamber; both treadmill and chamber are shown in figure 14. 



THE TREADMILL CHAMBER. 



The treadmill chamber was, for convenience in handling, constructed 

 in three parts; the base A, the skirt B, and the cover C. These parts 

 were made of Nos. 24 and 28 galvanized sheet iron. 



The sheet iron base, A, was formed over a platform 12.5 cm. high, 

 223 cm. long, and 88 cm. wide, made of spruce strips and matched floor 

 boards to give the necessary support and rigidity for the heavy mill. 

 The edge of the base was shaped into a trough, a, 4 cm. wide and 8 cm. 

 deep, which was filled with a light non-viscous oil. The skirt, B, fitted 

 into this trough and the oil made an air-tight seal between the base and 

 the skirt. 



The skirt, B, was of such a shape that it fitted rather snugly over the 

 mill, having a clearance of 6 cm. at the ends and 5 cm. at the sides, and 

 rising towards the center in the general form of a blunt wedge with an 

 opening at a height of 62 cm. above the leather belt. The upper open- 

 ing of the skirt was approximately 81 cm. square and was shaped with 

 a trough 6 cm. deep and 4 cm. wide, thus forming a second seal (&) 

 similar to that of the base; into this the cover (C) fitted. The seal b 

 was filled with water, for if oil were used the frequent raising and 

 lowering of the cover and consequent drip from the edges would make 



1 Benedict and Murschhauser, Carnegie Inst. Wash. Pub. No. 231, 1915, p. 34. 



