The effects of short- and long-lasting (2-min-long and up to 30-min-long) hypoxia episodes on the inspiratory activity (IA) recorded from then. phrenicus were tested in experiments on superfusedin situ semi-isolated medullo-spinal preparations (SIMSP) of newborn (the lst day of life) and 4- to 5-day-old rats. Hypoxia was provided by superfusion of the preparations with low-O2 solution. Short-lasting hypoxia evoked no significant modulation of the IA in preparations of newborn animals, while two-phase responses (an initial, up to 30 sec, increase in the frequency of inspiratory discharges, followed by their longer, up to 4 min, suppression) were observed in 4- to 5-day-old preparations. Long-lasting hypoxia suppressed activity in then. phrenicus of 1-day-old preparations, and this effect was replaced in five cases by the development of sporadic low-amplitude and short-lasting inspiratory discharges. These shortened discharges were qualified as gasps. The responses of 4- to 5-day-old SIMSP to long-lasting hypoxia episodes were more complex. An initial increase in the IA frequency lasted up to 30 sec, and in 4–6 min it was followed by complete suppression of the activity. In some of the SIMSP, permanent tonic activity appeared in then. phrenicus within the period of total absence of inspiratory discharges, which could be followed by generation of short low-amplitude gasping discharges. Against the background of gasping pattern, eupnoe-like discharges appeared in four preparations. Under control conditions, transerve section of the ventrolateral medullary regions (VLMR) at a middle level of then. hypoglossus root abolished respiratory activity in all studied SIMSP. Yet, in some of the SIMSP of both newborn and 4- to 5-day-old animals long-lasting hypoxia testing evoked weak tonic activity in then. phrenicus followed by the appearance of gasping discharges. After the transection of the VLMR at the caudal edge of then. hypoglossus root, long-lasting hypoxia evoked only weak tonic responses in some SIMSP of both age groups, and there were no phasic discharges in this case. The results of our experiments, first, show that the respiratory activity in newborn animals is more resistant to hypoxia than that in 4-to 5-day-old rats, and, second, they allow us to suppose that the gasp-generating medullary structures are localized in more caudal medullary regions. We discuss the questions of how the eupnoe-generating and gasp-generating medullary structures are formed in rats during their initial four to five postnatal days, and what specific features are typical of hypoxia-related respiratory responses in these animals.