The barometric method of Drorbaugh and Fenn for measuring respiratory rate and tidal volume in newborn infants has been compared by simultaneous recordings with an impedance pneumograph and a mercury-in-rubber strain gauge, using integrated flow rate measurements as the standard. The barometric method proved to be approximately as accurate as the others, but it allows the infant to be unrestricted with no interference by tactile stimulation during measurements, and it requires less instrumentation. Therefore a total body plethysmograph can be used confidently without additional equipment for intermittent recording of respiration in normal and diseased infants.