BACKGROUND: The motor unit number index (MUNIX) provides an index value relative to motor unit number from surface electromyography (EMG). It has been used in investigations of larger muscles, although very large ranges of values have been reported in the biceps brachii (BB) (Neuwirth et al, 2011). The MUNIX calculation relies on the surface interference pattern (SIP) from voluntary contractions and the compound muscle action potential (CMAP) from a supra-maximal stimulation of the nerve, both of which may vary when measured at different sites across a muscle. AIM: Investigate the effects of recording electrode position on SIP and CMAP area and power, and resulting MUNIX values from the BB muscle in healthy adults.METHODS: Participants gave written informed consent, and a 4-channel monopolar linear array surface EMG electrode was placed on both the medial and lateral heads of the BB of 10 participants, nine young; 24.8(4.2) yrs and one old; 70 yrs. The linear array on the medial head was placed over the motor point, defined as the area of muscle with the highest excitability from the smallest electrical stimulus. The remaining linear array was placed in line and parallel to this on the lateral head. EMG signals were recorded during maximum voluntary isometric contractions and at 5, 10, 20, 30, 40, and 60% of the maximum. The CMAP was obtained with a supramaximal stimulation of the musculocutaneous nerve. SIP and CMAP area and power, and MUNIX values were calculated for each of the 8 channels. RESULTS: MUNIX values differed significantly across different proximal-distal locations of the BB when analysed with a two-way repeated measures ANOVA (p<0.001), but did not differ between sites side by side on medial and lateral heads (p=0.532). Mean MUNIX across all channels was 131.CONCLUSIONS: Site selection is important when obtaining a MUNIX from the BB. Medial-lateral variations are less important than proximal-distal variations when considering a recording location. Similar values can be obtained from the same locations on the medial and lateral heads. A value recorded at a given location is unlikely to be representative of the entire muscle. This consideration should be applied when studying other larger muscles with the MUNIX method.