Instantaneous helical axes (IHAs) and screw displacement axes (SDAs) are commonly used to investigate joint functional axes of rotation. In the wrist, these have often been obtained through in vitro motion analysis. These definitions are then employed for in vivo applications, such as the design of implants or the development of musculoskeletal models. However, functional unguided joint motions are, by definition, affected by the activity of muscles. Previously published data has disagreed on the relative position and orientation of the two primary axes of rotation of the wrist, i.e. the radioulnar deviation (RUD) axis with respect to the flexion-extension (FE) axis. An in vivo study comparing the FE and RUD IHAs and SDAs of guided motions, to replicate in vitro conditions, and unguided motions of 23 healthy participants was conducted using optical motion capture. Guided motions were performed with the hand and forearm flush against a flat surface. The relative position and orientation of the RUD SDAs with respect to the FE SDAs differed (p = 0.019, p = 0.001) between unguided FE and guided RUD (0.1 ± 4.3 mm, 93.5 ± 16.0°) and guided FE and RUD (1.6 ± 4.0 mm, 107.8 ± 17.7°). This indicates that the use of different constraints, and not physiological differences, is the primary cause of differences in the relative positions and orientations of the FE and RUD axes in the literature. Thus, the practice of using in vitro definitions of the axes of rotation of the wrist for in vivo applications, especially involving FE, may be inappropriate and care must be taken to account for any constraint on wrist motion. It is recommended that investigators define the axes of rotation specifically for their study or refer to literature featuring the desired levels of constraint.