We tested two patients with posterior cerebral lesions on two pointing tasks. In the first task, the patients pointed to targets presented on a touch screen monitor and pointing accuracy was recorded. One patient (JR) demonstrated good localisation of targets presented to her blind field while the other patient (YP) did not. Movement kinematics were measured in the second task to compare the kinematics of movements made to sighted field targets with those made to blind field targets. For this version of the task both patients demonstrated above chance localisation of blind field targets although the slope of the relationship between the end of pointing movements and the target locations was significantly steeper for JR than for YP. Furthermore, JR showed a kinematic profile for movements made to blind field targets that mirrored the profile of kinematics to sighted field targets. That is, both peak velocity and time to peak velocity increased with increasing target eccentricity for movements made to blind and sighted field targets alike. Although patient YP now showed more reliable spatial localisation on this pointing task when compared with the touch screen task, his kinematics for movements made to targets in his blind field were quite different from those made to targets in his sighted field. Based on the patients' CT scans, we suggest that the superior performance of patient JR is a consequence of greater sparing of her parietal cortex in the damaged hemisphere.