Upper extremity (UE) neurodynamic tests are used to examine neural tissue in patients with neuro-musculoskeletal disorders. Although comparisons between involved and uninvolved limbs are made clinically, minimal data exist reflecting the normal variation between sides. The purpose of this study was to determine if within-subject differences exist between limbs in the UE component of neurodynamic tests of the median, radial, and ulnar nerves. Sixty-one healthy subjects were examined. Difference between limbs for the median nerve-biased test was significant (right=16.4° ± 11.4°, left=20.1° ± 13.7°; p=0.045). There was no significant difference between limbs for the radial or ulnar nerve-biased tests. Correlation between limbs was poor for all tests (median r(2) =0.14; radial r(2) =0.20; ulnar r(2) =0.13). Lower-bound scores were calculated to determine the amount of difference needed to consider asymmetry beyond measurement error; the scores for each neurodynamic test were as follows: median 27°, radial 20°, and ulnar 21°. The results of this study show that between-limb values have low correlation and that it may be normal for an individual to have range of motion differences between limbs with neurodynamic tests.