The heating system consists of a water pan inside the box near its floor, connected with a boiler on the right outside the chamber. This is in all respects similar to any hot water heating system. The premature infant requires a warm, moist atmosphere, of a definite tension, and also a sure and constant amount of radiant heat. The air must be 86° to 93° F, the bed must be 94° to 100° F, and these must vary proportionately as the incubator is set at one or the other degrees of temperature. These figures were obtained by thermometric observations. The radiant heat is supplied from the top of the pan and controlled by the distance between it and the bed. The air is heated by the under surface of the pan.

Both must be automatically regulated, which is obtained thus. In the upper right half of the chamber are three biconvexed hollow discs containing ethyl chloride these expand and contract with heat and cold. The slight motion thus obtained is multiplied by a system of levers thirty times, and in this way the cover of the air flue of the hot water boiler is raised and lowered. When the cover is raised the heat escapes, the water cools, the temperature in the incubator sinks. When the incubator grows too cool, the discs contract, the cover falls, the hot air is confined, the water is heated and then the temperature is raised. This regulator can be set at any degree from 70° to 100° F. The difficulties encountered with the heating apparatus were, first, to get a thermoregulator that would work. This one is the result of some experiment, much study, and the investigation of the instruments in the practice of poultry raisers. Mercury and gas regulators are not practical for infant incubation; metallic coils did not function regularly, or for any length of time, and electric thermostats were a complete failure. Second, the gas pressure varied throughout the day. This is obviated by attaching a gas pressure regulator to the burner; before we did this the work of preventing irregularities of the heat production and loss incident to the height of the flame had all to be borne by the thermoregulator; now this is not the case. Third, changes in the temperature of the room. A cold draught will cool the incubator before the regulator can supply the heat. This is obviated with this instrument, by protecting it from such chilling. In a larger system which we are going to install the incubators will be placed in a room, the temperature of which is to be even. Fourth, extreme changes of temperature outside. Of course, no one would expect any thermoregulator to equalize the extremes of our Chicago temperature. A change of 50 degrees is not impossible here, and to anticipate these variations the hospital is supplied with the daily weather forecast, and a thermometer is placed outside the nursery window, which the nurse consults, and regulates the admission of cold air to the incubator by a damper. Even without such attention, this incubator took care of a change of 47 degrees. For below zero weather we will install a preparatory warming chamber for the air. To show how perfectly this thermoregulator functionates, I have had hourly observations made during periods of two days. The incubator was set at 90° F, and in two days did not vary more than three degrees; that is, from 88° to 91°, while the outside temperature, given by the official weather bureau, varied 16 degrees, and the wind veer around the points of the compass.