In normotensive, halothane-anesthetized rats, oscillations can be found both in the single-nephron blood flow and in the tubular pressure. Experimental data and computer simulations support the hypothesis that the oscillations are caused by the tubuloglomerular feedback (TGF) mechanism. Model studies show that the key parameters determining the stability of the TGF system are the open loop gain of the system and the time delays in the signal transmission through the various components of the feedback loop. Within a broad range of parameters, the system is unstable and has self-sustained stable oscillations. The parameter range where model studies show instability overlaps with the physiologic range for the values of the same parameters. The system appears to be poised on the border between stability and oscillation, and a small parameter change may cause the system to move from one state to the other. In renovascular and spontaneously hypertensive rats, regular oscillations give way to highly irregular, chaotic fluctuations. The chaotic fluctuations appear to have the same mechanism as the regular TGF-mediated oscillations. The irregular fluctuations most likely represent a parameter-dependent transition from a limit cycle (regular oscillation) to deterministic chaos. The key parameters causing the transition have not been identified. Associated with the difference in the dynamics of TGF between normotensive and hypertensive rats is a change in the dynamic autoregulation of total RBF. This is especially prominent in the frequency range in which TGF operates, and it is suggested that a causal relationship may exist between the two phenomena. This difference may play a role in the pathogenesis of hypertension by altering the renal response to the normal fluctuations in arterial pressure.