There is debate whether the terminal latency index (TLI) is a sensitive marker for polyneuropathy with anti-myelin-associated-glycoprotein antibodies (anti-MAGP). We examined 6 patients with an anti-MAGP and 6 patients with a chronic inflammatory demyelinating polyneuropathy (CIDP). The electroneurographic features studied were: distal compound motor action potential (CMAP), distal motor latency (DML), motor conduction velocity (MCV) elbow to wrist (distal MCV), MCV axilla to elbow (proximal MCV), MCV distal/proximal, terminal latency index (TLI), residual latency (RL), F-wave, and modified F ratio. We found significant differences between anti-MAGP and CIDP for DML and for RL. No significant differences were found for TLI and the other measures. The TLI values were not significant probably because our patients had a longer duration of disease,which supports the hypothesis of a distal to proximal progression of conduction slowing over time. We propose that a residual latency >4.0 and a distal motor latency >7.0 are strongly suggestive for an anti- MAGP.