Electrically elicited long-latency reflexes (LLRs) were obtained from thenar muscles by either fully rectified or non-rectified raw recordings in 10 healthy volunteers. The LLR II onset latencies were significantly (P < 0.0001) delayed on rectified (mean +/- SD: 49.8+/-2.9 ms) compared to raw (45.3+/-2.3 ms) recordings, with a mean difference of 4.4 ms. These data show that, according to the recording technique employed, the LLR II onset latencies can change considerably. The possible implications on cortical relay time (CRT) calculation and the understanding of the intracortical connections physiology are discussed.