Previously it has been reported that horizontal disparity vergence is strongly influenced by subject instructions to vary attention or tracking effort. This paper describes experiments which compared these instruction effects on horizontal and vertical disparity vergence. Within-trial comparisons were made possible by use of oblique (combined horizontal and vertical) disparity modulation. Subjects viewed a flat, fully correlated, dynamic random noise stereogram pattern through stationary circular apertures, with a small stationary fixation cross superimposed in the center. The disparity of the noise pattern was either modulated sinusoidally or changed abruptly. Subjects were instructed either to (1) hold fixation on the cross and ignore the disparity modulation of the noise pattern; or (2) follow the movement of the noise pattern as accurately as possible. Subjects showed clear effects of instruction on the horizontal component of tracking, but showed little or no effect on the vertical component. Horizontal and vertical components of oblique vergence tracking appear to be largely independent, and vertical vergence is affected minimally, if at all, by an effort to track.