Background: Controversy still surrounds the differential susceptibility of nerve fibers to local anesthetics and its relation to selective functional deficits. In the current study we report features of conduction blockade in different classes of rat sciatic nerve fibers after injection of lidocaine by a percutaneous procedure that closely resembles clinical applications. Methods: In 30 adult male Sprague-Dawley rats (weight, 300-400 g) during general anesthesia, impulses were recorded in different classes of sensory axons (large, Aα and β fibers; small, Aγ myelinated fibers and unmyelinated C fibers) and motor axons (large, Aγ fibers; small, Ay myelinated fibers) classified by conduction velocity. The sciatic nerve was stimulated distally, and impulses were recorded from small filaments teased from L4-L5 dorsal (sensory) and ventral (motor) roots sectioned acutely from the spinal cord. Lidocaine at concentration of 0.05-1% was injected percutaneously in 0.1-ml solutions at the sciatic notch. Both tonic (stimulated at 0.5 Hz) and use-dependent (stimulated at 40 Hz for Aδ and Aγ fibers and at 5 Hz for C fibers) impulse inhibitions by lidocaine were assayed. Results: Minimal effective (threshold) lidocaine concentrations (i.e., to block conduction in 10% of fibers) were, for sensory, 0.03% for AS, 0.07% for Aαβ, and 0.09-0.1% for C fibers, and for motor, 0.03% for Aγ and 0.05% for Aα fibers. The order of fiber susceptibility, ranked by concentrations that gave peak tonic fiber blockade of 50% (IC50s), was Aγ > Aδ = Aα > Aαβ > C. Faster-conducting C fibers (conduction velocity > 1 m/s) were more susceptible (IC50 = 0.13%) than slower ones (conduction velocity < 1 m/s;IC50 = 0.30%). At 1% lidocaine, all fibers were tonically blocked. Use-dependent effects accounted for only a modest potentiation of block (at a lidocaine concentration of 0.25%) in Aδ and Aγ fibers, and in C fibers phasic stimulation had even smaller effects and sometimes relieved tonic block. Conclusions: Susceptibility to lidocaine does not strictly follow the "size principle" that smaller (slower) axons are always blocked first. This order of fiber blockade is qualitatively consistent with previous reports of the order of functional deficits in the rat after percutaneous lidocaine, that is, motor = proprioception > nociception, if we assume that motor deficits first arise from conduction failure in Aγ fibers and that nociception relies on C fiber conduction.