Some time before the man enters the chamber, an electric lamp of from 16 to 24 candle-power (depending upon the size of the subject) is placed inside of the chamber as a substitute for the man, and the cooling water-current is started and the whole apparatus is adjusted to bring away the heat prior to the entrance of the man. The rate of flow with the chair calorimeter is not far from 350 cubic centimeters per minute with a resting man. The proper mixture of cold and warm water is made, so that the electric reheater can be controlled readily by the resistance in series with it. Care is taken not to allow the water to enter the chamber below the dew-point and thus avoid the condensation of moisture on the absorbers. The thermal junctions indicate the temperature differences in the walls and the different sections are heated or cooled as is necessary until the whole system is brought as near thermal equilibrium as possible. After the man enters, the lamp is removed and the water-current is so varied, if necessary, and the heating and cooling of the various parts so adjusted as to again secure temperature equilibrium of all parts. When the amount of heat brought away by the water-current exactly compensates that generated by the subject, when the thermal-junction elements in the walls indicate a 0 or very small deflection, when the resistance thermometers indicate a constant temperature of the air inside the chamber and the walls of the chamber, the experiment proper is ready to begin. The physical observer keeps the chemical assistant thoroughly informed as to the probable time for the beginning of the experiment, so that there will be ample time for making the residual analyses of the air. After these analyses have been made and the experiment is about to begin, the observer at the table calls the time on the exact minute, at which time the blower is stopped and the purifying system changed. The physical observer takes the temperatures of the wall and air by the electric-resistance thermometers, reads the mercury thermometers, records the rectal thermometer, and at the exact moment of beginning the experiment the current of water which has previously been running into the drain is deflected into the water-meter. At the end of the period this routine is varied only in that the water-current is deflected from the water-meter into a small can holding about 4 liters, into which the water flows while the meter is being weighed. MANIPULATION OF THE WATER-METER.