The quantitative relation between the concentration of local anesthetic (LA), the length of nerve exposed, and severity of conduction blockade was studied with use of a chamber where exposure length was varied as the concentration of lidocaine was held constant. Recordings of the compound action potential and of single axons established that small variations in the length of nerve exposed to LA strongly modulate conduction block even at exposure lengths in excess of 2 cm. Therefore, exposure length is a significant factor in determining blocking potency, and only at very high concentrations of LA, where voltage-dependent Na conductance is almost completely blocked, is the critical exposure length less than three nodes of Ranvier. The concentration required for 50% block of impulses in single fibers (that is, where 50% of the impulses would fail to propagate through the exposed region of the nerve) diminished as the exposed length of nerve increased, approximately halving as exposure length was changed from 6 mm to 15-25 mm. Conduction latency increased with the exposure length becoming sharply more variable as the critical exposure length for conduction block was approached. The results are consistent with the hypothesis of decremental conduction, where a partial active response in nodes exposed to marginal blocking concentrations extends the decay of the action potential along the axon, and do not support the interpretation that lengths of several centimeters affect blocking concentration because such distances increase the probability that three nodes will be blocked in succession. This study contradicts the broader common assumption that beyond three nodes, the length of nerve exposed is not a factor in nerve block with local anesthetics.