AIM: Preterm infants begin the transition from gastric tube feeds to sucking feeds around 34 weeks' postmenstrual age. We compared physiological stability in two bottle feeding positions, cradle hold versus side lying in preterm infants. METHODS: Randomised crossover trial of infants <34 weeks' gestation at birth, ≥34 weeks' postmenstrual age at study and receiving at least two sucking feeds/day. Two feeds were studied on successive days. A pulse oximeter measured oxygen saturation (SpO₂) and heart rate (HR) before, during and 30 min after feeds. Continuous data were compared using paired t-tests and proportions using chi squared. RESULTS: Twenty-five study infants were mean (standard deviation (SD)) 37 (2.4) weeks' post-menstrual age and 2740 (589) g at study. There was little difference in mean (SD) SpO₂ during feeds between the cradle-hold and side-lying position 94 (6) % versus 95 (6) %, respectively (P = 0.55, confidence interval (CI) -1.4, 5.4). During feeds, 17/25 (68%) experienced a period of SpO₂ <80% in the cradle-hold position compared with 14/25 (56) % in the side-lying position (P = 0.26, CI 0.68, 4.10). There were no significant differences in the mean HR or number of episodes of bradycardia HR <100 bpm. There was a trend towards infants consuming a smaller mean (SD) proportion of their feed in the cradle-hold position compared with the side-lying position, 82 (25) % versus 87 (20) % (P = 0.08, CI -0.64, 10.00). CONCLUSIONS: There was little difference in infants' physiological stability between the two bottle feeding positions. Both methods may be appropriate for the transition from gastric tube to sucking feeds in preterm infants.