Heat losses from babies in incubators.

Incubators are now widely employed for nursing premature and ill newborn infants, because they provide a shielded and warm environment in which temperature, relative humidity, and oxygen content can be conveniently controlled. Infants are frequently nursed naked, which makes accurate observation of them easier and diminishes the amount of handling necessary. In these circumstances it is clearly important to be able to assess the adequacy of the warmth provided by the incubator. Two types of incubator have been investigated, one depending on natural convection for the internal distribution of warmed air, and the other depending on forced convection. Natural convection is the term applied to the movement of air produced by heat; the upward current of air over a hot stove is a familiar example. Forced convection takes piace when an external force, such as a fan, produces an air current. Natural convection depends on localized density changes in the air; forced convection depends on outside means being used to control air movement. The main results are presented and discussed below, but, for convenience, details of the methods employed and certain technical arguments have been relegated to an Appendix. A brief preliminary account of this work has already been published (Hey and Mount, 1966). Heat exchange occurs through four main channels: conduction, convection, radiation, and evaporation. The measurement of air temperature within an incubator provides some indication of the likely heat loss by convection, but no indication of the losses by the other channels, and to determine the nature of the thermal environment provided by an incubator it is necessary to assess the heat exchange from each of these four channels separately. When this is done it becomes clear that the radiant heat loss suffered by the infant can be substantial. In order to keep the infant warm, the incubator air is heated and its temperature monitored with a thermometer. The incubator walls, however, are