1. Four male monkeys (M. fascicularis) were trained in a movement performance task which involved pulling a horizontal lever into a target zone and then collecting, from one of a variety of positions, a small food reward. The same animals were also trained to sit quietly and accept passive manipulation and natural stimulation of the arm and hand while remaining relaxed. 2. After complete bilateral section of the cuneate fasciculi or division of a major part of these dorsal column afferents at C1-C2 or at C5 level, the animals were still able to perform movement tasks normally. Disturbance of discrimination ability was revealed after vision was occluded it the animal was required to detect differences in texture with only a small cutaneous area in contact with the object (e.g. using only the tip of the index finger). Contactual-placing reactions could be performed in the absence of vision and the movements the animal made in these reactions were well controlled and appropriately directed. Minimal disturbance of contact placing was noticed if the surface touched was on the hand or fingers or if the reaction involved crossed placing. 3. An examination of the natural discharges of 342 percentral neurones revealed that the patterns of activity exhibited in relation to complex movements were indistinguishable from patterns recorded in normal monkeys carrying out similar tasks. 4. Discharges of ninety-one of 321 precentral neurones could be produced by appropriate natural stimuli delivered within the cell's afferent input zone at the periphery. The zone from which a given cell could be influenced was usually limited and its location could be on any part of the contralateral forelimb. However, all but nine of these responses were found in animals in which a small proportion of the cuneate fibres remained intact. In an animal with histologically proven complete section of the cuneate faciculi very few (nine of 171) precentral neurones were influenced by natural activation of peripheral receptors in the forelimb. The zones from which these few afferent inputs were found could all have been proximal to the level of the cuneate lesion. 5. The very small number of responding pre-central neurones found in an animal with complete section of the cuneate fasciculi made it likely that the dorsal columns provide the major pathway for effects from circumscribed peripheral receptors in the forelimb to influence precentral neurones. However, even in an animal with complete interruption of cuneate fibres, a proportion of post-central neurones could still be influenced by natural activation of peripheral receptors within restricted regions of the forelimb. Hence the 'sensory' cortex was still in receipt of afferent projections which could be revealted readily by the tests used. 6...