From the right side of the skirt near the front a 6 cm. pipe extended along the side of the skirt to a drier (H) 40 cm. high and 30 cm. in diameter, filled with 15 kg. of fused calcium chloride. The calcium chloride was in small lumps about 1 cm. in diameter, which allowed a free passage of the circulating air. It was fairly efficient in absorbing power, and kept the humidity within the chamber at a point varying from 20 to 50 per cent. The psychrometer, G, was inserted at the point where the 6 cm. pipe left the skirt.

From the drier a 10 cm. pipe, J, carried the dry air back to the chamber which it entered at the rear of the skirt. This return pipe had a total length of 250 cm. and the mill was so located in the room that 100 cm. of this pipe passed through and extended horizontally along the face of the doorway, I, of an adjoining room. In this part of the circuit were inserted the stopcocks through which air samples could be simultaneously withdrawn. The doorway was closed up so as to provide a room of reasonably constant temperature for the gas analysis. The last section of this return pipe, L, was made removable by means of two small water seals, K and K', so that, when this section was lifted from its connection with the chamber, the large cover could be lowered into or raised from the skirt without altering the tension of the air inside the chamber.

Ventilation Of Chamber

A "Sirocco" blower, F, within the chamber, attached to the circulating air pipe at the point where it left the skirt, maintained an air-current through the system of 1,000 liters a minute. An electric fan, E', was placed in the top of the cover in such a position as to send a current of air over the head and shoulders of the subject, while a similar fan, E, was placed under the belt between the two pulleys of the treadmill. These two fans, E and E', together with the circulating air-blower, F, previously referred to, were relied upon to keep the air well mixed within the chamber during the active periods of walking. In addition to these a large 15 cm. blower, capable of delivering 12 cubic meters of air a minute, was mounted outside on a table which could be rolled up to the side of the skirt. During the intervals between the experiments with the series of subjects, this blower was kept in operation to sweep out the air in both skirt and cover. Two windows on opposite sides of the room gave a rapid means of ventilation.

Tension Equalizers

On the left side of the skirt near the front was soldered a 2.5 cm. coupling, to which were connected two spirometers, M and M', with a capacity of 7.2 liters each. These spirometers were connected in series by a 3-way valve and were designed to allow for any change in volume of the chamber due to barometric or temperature conditions. Only one spirometer was regularly used, the second being kept in reserve for any sudden and unexpected changes which might occur. In practice it was found that any ordinary changes in volume could be controlled by changing the temperature of the room; at no time was it necessary to use the second spirometer.

The position of the spirometer was read on a millimeter scale at the beginning and end of each period under a uniform pressure as indicated by a Sonden oil manometer attached to the skirt of the chamber. The spirometer had been previously calibrated and each change of 1 mm. in the level of the spirometer bell was equivalent to a change in volume of 21.5 c.c.

Resistance Thermometers

The temperature of the chamber was measured by means of six resistance thermometers, made of silk-covered No. 33 copper wire which was wound on wooden frames 4 by 15 cm. These were fastened to the walls of the chamber by studs 35 mm. long. One thermometer, N1, was placed near the top, another, N2, near the bottom of the cover, one each (N3 and 2V4) was placed at the front and back of the skirt, respectively, about 20 cm. above the belt of the treadmill, while the remaining two (only one, N5, is shown in the figure) were placed below the belt near the rear and front pulleys of the mill. The six thermometers were connected in series and had a total resistance of 98 ohms. The leads left the chamber by means of brass rods, which passed through a hard-rubber plug, P, set into the front of the cover by means of gaskets and bushings. From this plug double leads of No. 14 copper wire carried the current to a galvanometer and Wheat-stone bridge in the adjoining gas-analysis room. Readings of 0.003 ohm were made, corresponding to a temperature difference of 0.01° C.

Barometer

The barometer was a standard observatory pattern graduated so that 0.05 mm. could be read by means of the vernier.