To decide whether tubuloglomerular feedback (TGF) can account for renal autoregulation, we tested predictions of a TGF simulation. Broad-band and single-frequency perturbations were applied to arterial pressure; arterial blood pressure, renal blood flow and proximal tubule pressure were measured. Data were analyzed by linear systems analysis. Broad-band forcings of arterial pressure were also applied to the model to compare experimental results with simulations. With arterial pressure as the input and tubular pressure, renal blood flow, or renal vascular resistance as outputs, the model correctly predicted gain and phase only in the low-frequency range. Experimental results revealed a second component of vascular control active at 100-150 mHz that was not predicted by the simulation. Forcings at single frequencies showed that the system behaves linearly except in the band of 33-50 mHz in which, in addition, there are autonomous oscillations in TGF. Higher amplitude forcings in this band were attenuated by autoregulatory mechanisms, but low-amplitude forcings entrained the autonomous oscillations and provoked amplified oscillations in blood flow, showing an effect of TGF on whole kidney blood flow. We conclude that two components can be detected in the dynamic regulation of renal blood flow, i.e., a slow component that represents TGF and a faster component that most likely represents an intrinsic vascular myogenic mechanism.