<p>The same axis of rotation can be defined in different coordinate systems, leading to different form and different interpretation of the rotation matrix. This definition is particularly important in case of <i>successive rotations</i> due to the wide use of finite orientation parameters to define rigid- and flexible-body kinematics as well as deformation measures. As discussed in this paper, in case of successive rotations, the rotation axes can be defined using a single frame or multiple intermediate frames. As it is known, the <i>single-frame method</i> (SFM) and <i>multi-frame method</i> (MFM) lead to different order of multiplication of the rotation matrices. It is demonstrated that the coordinate system in which the rotation axis is defined determines the physical interpretation of the rotation matrix and deformation measures. In the SFM approach, the single frame can be global, cross section, or body frame. Nonetheless, the SFM sequence does not allow easy interpretation of the rotations as torsion and/or bending measures. In the MFM approach, on the other hand, the rotations are defined relative to intermediate frames; and such relative rotations can only be given physical interpretation as deformation measures in case of infinitesimal rotations. Such a physical interpretation is not valid in case of using noncommutative finite rotations. Regardless of which method (SFM or MFM) is used, the sequence of rotations cannot, in general, be reversed. Using <i>Rodrigues formula</i> with the analysis and results of this study, it is demonstrated that the <i>noncommutativity problem</i> is not attributed to the choice of the spatial orientation parameters but to the noncommutativity of the rotation matrices.</p>