The dynamics of heartbeat interval time series over large time scales were studied by a modified random walk analysis introduced recently as Detrended Fluctuation Analysis. In this analysis, the intrinsic fractal long-range power-law correlation properties of beat-to-beat fluctuations generated by the dynamical system (i.e., cardiac rhythm generator), after decomposition from extrinsic uncorrelated sources, can be quantified by the scaling exponent (alpha) which, in healthy subjects, for time scales of approximately 10(4) beats is approximately 1.0. The effects of chronic hypoxia were determined from serial heartbeat interval time series of digitized twenty-four-hour ambulatory ECGs recorded in nine healthy subjects (mean age thirty-four years old) at sea level and during a sojourn at 5,050 m for thirty-four days (EvK2-CNR Pyramid Laboratory, Sagarmatha National Park, Nepal). The group averaged alpha exponent (+/- SD) was 0.99 +/- 0.04 (range 0.93-1.04). Longitudinal assessment of alpha in individual subjects did not reveal any effect of exposure to chronic high altitude hypoxia. The finding of alpha approximately 1 indicating scale-invariant long-range power-law correlations (1/f noise) of heartbeat fluctuations would reflect a genuinely self-similar fractal process that typically generates fluctuations on a wide range of time scales. Lack of a characteristic time scale along with the absence of any effect from exposure to chronic hypoxia on scaling properties suggests that the neuroautonomic cardiac control system is preadapted to hypoxia which helps prevent excessive mode-locking (error tolerance) that would restrict its functional responsiveness (plasticity) to hypoxic or other physiological stimuli.