When the eyes fixate a point in the median plane of the head, the physical surface to which the stereoscopic disparities of “nearer” and “farther” points are referred is the so-called frontal plane horopter. When, however, a point is fixated with the eyes in asymmetrical convergence, the horopter is now rotated with respect to the objective normal plane-that is, the surface that is normal to the direction of gaze. This rotation is believed to pose a problem with veridical stereoscopic localization. It has, therefore, been postulated that in asymmetrical convergence some physiological compensation takes place in order to preserve accurate stereoscopic perception. Some experimental evidence apparently supports this view. The logic of the arguments and the experiments are critically assessed with the conclusion that there is little evidence of any visual modification, nor would any seem likely. A reinterpretation of the stereoscopic axes of reference does occur, but this can be accounted for purely by innervational factors, the visual aspects remaining unchanged.