Six observers, experienced in telerobotic operations, were used across four replicated studies of remote performance of a simulated space station assembly task. An alignment/insertion task was performed with a remotely operated manipulator arm viewed either directly or through stereoscopic or monoscopic TV viewing systems. Target position, space light, and learning effects were also assessed by measures of task time and manipulator collisions. Performance with Stereo view was significantly superior to that with Mono in Experiments 1, 2, and 4. It's superiority fell below required significance levels in Experiment 3 due to of the accumulation of practice effects across the first two studies. Experiment 4, in which the left-right positions of manipulator arm and task element were reversed, reestablished the strong superiority of Stereo view over Mono. These results clearly show the superiority of Stereo TV over Mono viewing systems. They suggest that learning can also improve performance under Mono view when accompanied by Direct view and Stereo view experience, but such learning is specific to the perceptual and motor conditions that were present in practice. Space lighting was not significant in the two studies in which it was assessed.