As may be noted in figure 6, air is taken from outdoors and delivered by means of a rotary air-impeller, k, through an opening, 6, near the top of the chamber. Since on many days in winter the outdoor air is extremely cold, it has been necessary at times to warm the air. This is done by means of a Bunsen burner, m, with a small hood, attached to the outside of the sheet metal pipe carrying the air from the blower to the chamber. By regulating the size of this flame, any degree of temperature may be secured for the air entering the chamber. Butterfly valves, n, in the air-pipe control the amount of air delivered to the chamber and may be adjusted at will, irrespective of the speed of the blower. For practically all experiments thus far made we have used only a fraction of the possible discharge from the blower and the cross section of the pipe has always been considerably reduced by turning the butterfly valve.

Incoming Air

The incoming air is taken from a point outside the north window of the building. The pipe on the intake side of the blower enters the calorimeter laboratory through a board fitted into one of the windows and has a diameter of 6 inches. The pipe between the blower and the chamber has a somewhat smaller diameter, 4 inches. No provision is made for noting directly the amount of air entering the chamber or its degree of humidity. In fact, as will be seen later, no analysis is made of the ingoing air. This is one of the simplified features of the apparatus. Care is taken to deliver to the chamber only uncontaminated outside air. Repeated analyses of air in the neighborhood of the Nutrition Laboratory have shown that, for all practical purposes, the carbon-dioxide content is constant, irrespective of weather conditions, temperature, or season.1

1Benedict, Carnegie Inst. Wash. Pub. No. 166, 1912, p. 114.

Outgoing Air

From the experience of more than a decade with the ventilation of various types of respiration chambers, it was clear that a provision for the rapid renewal of air inside the chamber was unnecessary. Experiments at Wesleyan University for periods of 2 to 13 days have shown that if the ventilation is sufficient to maintain the carbon-dioxide content of the air inside the chamber at not more than from 0.5 to 1 per cent, no discomfort is experienced by human subjects. Indeed, in one instance a human subject lived in an atmosphere containing approximately 2 per cent of carbon dioxide for more than 24 hours without discomfort. Without doubt the greatest factors in determining so-called "bad" air are temperature and humidity; we firmly believe that carbon dioxide, per se, has no influence.

Until the oxygen content is lowered to considerably below 15 per cent there is no evidence of labored respiration or indication of oxygen-want. On the other hand, since moisture in the air and especially a high temperature are extremely disagreeable, it became necessary to remove the moisture by a sufficient flow of air in the chamber or to condense it by use of brine, which likewise provided for temperature control. Finally, we decided that the safest procedure would be to adjust the rate of ventilation so that the percentage of carbon dioxide residual in the air in the chamber would be not far from 0.5 per cent, and to control the total amount of the ventilating air-current so that this adjustment could be made within wide limits. In other words, when but a small amount of carbon dioxide is being produced, the total ventilation would be low; when a large amount, it would be very high. This fluctuation in the possible ventilating capacity of the chamber was readily secured by the use of the rotary air-impeller, referred to in a foregoing paragraph. These impellers, the discharge of which may be cut down at will by butterfly valves, permit the movement of a very large amount of air through the chamber in a very short time. Indeed, with the impeller used here, with a discharge of 97 mm. and with a speed of the armature shaft of 1,700 revolutions per minute, it has been computed that the entire volume of the chamber can be swept out in a very few minutes with full and free discharge of the blowers. On the other hand, by reducing the delivery by butterfly valves or other suitable device, the total ventilation of the chamber can be brought to as small a volume as desired.

It is thus apparent that the carbon-dioxide content of the outcoming air will, under all circumstances, be very considerably greater than that of normal air; hence the problem of measuring exactly the total carbon dioxide removed from the chamber assumes grave importance. To pass the entire air-current through soda-lime would, for reasons outlined previously, be wholly impracticable, as it would require an absorption system of purifiers which would be difficult to maintain, a very large amount of soda-lime, and provision for taking care of the intense heat of reaction between carbon dioxide and soda-lime. It became necessary, therefore, to provide for an accurate aliquoting of the main air-current leaving the chamber; this aliquoting device forms the chief feature in our description of the group respiration chamber.

Aliquoting And Analysis Of Ventilating Air-Current

On the general fundamental principle that the total ventilating air-current passing through the chamber should be of such a magnitude as to maintain a carbon-dioxide content in the air inside the chamber of not far from 0.5 per cent, several possible methods for the determination of the carbon dioxide in the outcoming air presented themselves. Thus, one could use the long-established method of Pettenkofer and Voit1 of determining the carbon-dioxide content of both the incoming and outgoing air, noting the total amount of air leaving the chamber through a series of meters, and computing from these the carbon dioxide produced inside the chamber. This would require complicated gas analysis which, if possible, it is desirable to eliminate. Our success with the universal respiration apparatus and its train of purifiers, consisting of soda-lime bottles and sulphuric-acid bottles of the Williams type, naturally led to an attempt to employ this thoroughly-tested train in connection with the analysis of the air leaving the chamber.