Surfactant-deficient premature newborn rabbits obtained by hysterotomy on day 27 of gestation were tracheostomized at birth, kept in a multichambered pressure-constant body plethysmograph, and subjected to intermittent positive pressure ventilation using a standardized insufflation pressure (30 cm H2O). A sequence of various frequencies (20, 40, and 60/min) and inspiration:expiration (I:E) ratios (1:4, 1:2, 1:1, 2:1, and 4:1) were applied, and the effect of these various respirator settings on tidal volume (VT) and flow was evaluated by means of a Fleisch tube connected to the body plethysmograph. Irrespective of respirator frequency, increasing I:E ratio from the basic setting of 1:1 resulted in increased VT; decreasing I:E ratio had the opposite effect. There was a nearly linear relation between variations in I:E ratio and VT, but the slope of the regression line was steeper at frequencies 40 and 60/min than at 20/min. At a frequency of 20/min, values (X +/- S.D.) for lung-thorax compliance at I:E ratios of 1:4 and 4:1 were 0.50 +/- 0.26 and 0.99 +/- 0.42 ml/cm H2O.kg, respectively. Corresponding figures for frequency of 60/min were 0.21 +/- 0.10 and 0.73 +/- 0.36 ml/cm H2O.kg. At the two higher frequencies and low I:E ratios (less than or equal to 1:1), the duration of the inspiration phase was not sufficient to provide the maximal degree of lung expansion attainable with this insufflation pressure under static conditions. At frequency of 60/min and I:E ratio of 4:1, the expiration was short enough to prevent zero flow before onset of inspiration.