Two models of magnetization reversal for the case of films with biaxial anisotropy are considered: 1 ) the coherent rotation model of Stoner and Wohlfarth (S-W) and 2) the noncoherent or wall-motion model of Kondorsky (K). The former has been considered by several authors previously while the latter is developed in the present paper. The (K) model presumes that the anisotropy field is much larger than the critical field for wall motion and that only the component of the applied field perpendicular to the bisector of the initial and final positions of the magnetization is effective. It also presumes that whenever the effective component exceeds a certain critical field determined by the structure of the sample, the magnetization will switch. Rotational hysteresis, torque curves, and hysteresis loops observed on single-crystal cobalt films exhibiting biaxial crystalline anisotropy are compared with the predictions of both models. It is found that at low fields the (K) model is a much better description than (S-W) while at high fields the reverse is true. This behavior is analogous to the behavior of uniaxial polycrystalline films with low values of . This similarity and others suggest that the true reversal process is as complicated in single-crystal films as in polycrystalline films, and that dispersion in both the magnitude and direction of the anisotropy is present. Because the samples are single-crystal, single-element films, the possible causes of dispersion are reduced. The most likely cause is locally varying stresses arising from uneven substrate bonding.