A preliminary analysis is presented of whether the isotropic temperature factors derived from refinement of the tetragonal crystal structure of hen egg-white lysozyme at 2 Å resolution can be interpreted in terms of molecular motion. If the contributions to the temperature factors from experimental errors, crystal disorder and imperfections of the molecular model are neglected, the apparent thermal motion is found to be compatible with the pair of molecules that have the strongest interactions in the crystal either moving as a rigid-body with libration about a common axis, or vibrating in an intramolecular mode with the principal amplitudes radial to this axis. There is only weak evidence for hingebending, i.e. that the two lobes of lysozyme vibrate so that their separation varies. Side-chains that are exposed to the solvent and some segments of the main polypeptide chain (including residues 70-73, 82-84, 101-103) have greater apparent thermal motion than the remainder of the molecule. Although this analysis at a single temperature cannot establish whether thermal motion or static disorder (including conformational variability) underlies the observed effects, it suggests that the accurate determination of temperature factors will be useful in detailed studies of the dynamic properties of macromolecules. © 1979.