Previous findings on the role of visual contact with the hand in the control of reaching and grasping have been contradictory. Some studies have shown that such contact is largely irrelevant, while more recent ones have emphasised its importance. In contrast, information arising from the surrounding environment has received relatively little attention in the study of prehensile actions. In order to identify the roles of both sources of information, we made kinematic comparisons between three conditions. In the first, reaching was performed in a dimly lit room and compared with a second condition in which reaches in the dark, but with the thumb and first finger illuminated, were made to a luminous object. This contrast allows the effects of environmental context to be identified. A comparison between the second and a third condition, in which both vision of the hand and the environment was removed, but the object was still visually available, enabled the assessment of how and when vision of the hand plays a role. Removing environmental cues had effects both early and late in the reach, while vision of the hand was only crucial in the period after peak deceleration. In addition, removal of both sources of information resulted in larger grip apertures. Differences and similarities between our findings and those of other studies are discussed, as is the ongoing debate about the relative importance of visual feedback of the hand in the control and co-ordination of prehensile actions. We conclude with suggestions for further research based on the set-up used in the present study.