Detection of conduction block (CB) has important clinical implications because it suggests segmental demyelination potentially reversible following treatment. There are no universally accepted criteria to define CB. We tested the sensitivity of two diagnostic criteria of partial motor CB in patients with chronic inflammatory demyelinating polyneuropathy (CIDP) and hereditary motor-sensory neuropathy type I (HMSN I) which are thought to be the prototypes of segmental and uniform demyelination respectively. Criterion I requires > 20% drop in negative peak amplitude and area and < 15% change in duration between proximal and distal compound muscle action potentials (CMAPs). Criterion II requires > 50% drop in negative peak amplitude and area of proximal CMAP independently from temporal dispersion. Twenty-eight percent of CIDP nerves and 65% of CIDP patients had CB according to criterion I. However, 29% of nerves and 61% of patients with HMSN I also fulfilled the same criterion. Thirty-four percent of CIDP nerves and 78% of CIDP patients and none of HMSN I patients fulfilled criterion II. Criterion II has the same sensitivity as criterion I, but seems to be highly specific in diagnosis of the segmental demyelination characteristic of CIDP. Therefore this criterion should be employed, as the diagnostic tool of CB, in chronic neuropathies presenting high stimulation threshold and coexisting axonal loss.