1. Monkey interpositus neurones show large discharge modulations durin
g reaching to grasp, however, the same neurones show little or no modu
altion during operation of devices that exercise individual forelimb j
oints. We tested the hypothesis that grasping during the reach-to-gras
p is necessary for eliciting high discharge modulation. 2. Three monke
ys (Macaea mulatta) moved an articulated lever between low and high ta
rget zones. While in the lower zone the monkey's hand was at its waist
, in the upper zone its hand was in a position that required forelimb
extension at right-angles to the body axis. Small drawers adjacent to
the target zones contained raisins, and the drawers could be remotely
opened. Thus, we could elicit two types of reaches having similar traj
ectories: one reach involved limb transport while holding the lever ha
ndle, and the other involved limb transport while forming the hand to
grasp a raisin. 3. Eighty-one neurones from two monkeys, mostly from i
nterpositus with some from adjacent regions of dentate, were tested du
ring device use and reaching to grasp: 93 % of the neurones discharged
at high rates during at least one of the tasks. Of these, about half
increased discharge rate solely during reaching to grasp; the other ha
lf showed some increase during device use but only discharged strongly
during reaching to grasp. Overall, discharge modulations during the r
each-to-grasp averaged twice as high as during the corresponding devic
e movement (112 versus 56 impulses s(-1)). 4. Individual neurones cons
istently discharged with characteristic patterns during the reach-to-g
rasp with rates often exceeding 300 impulses S-1 Discharge during the
a reach-to-grasp was independent of reach trajectory: discharge patter
ns and amplitudes were similar when reaching from either the lower or
upper target zone to the upper raisin drawer as when reaching from the
upper target zone to the upper raisin drawer. Reach direction also ma
de little difference: reaches from the upper target zone to the lower
drawer typically elicited similar discharge modulation as those from t
he lower target zone to the upper drawer. 5. Righ discharge rates asso
ciated with grasping were independent of the item being grasped: typic
ally, grasping the device handle elicited as high discharge rates as g
rasping a raisin. 6. The hypothesis was confirmed that grasping is cri
tical for eliciting high discharge modulation in interpositus during r
eaching to grasp. Discharge pattern and modulation do not vary with re
ach direction or amplitude of the reach and, therefore, it is unlikely
that intermediate cerebellum controls these features of the reach-to-
grasp. We hypothesize that intermediate cerebellum is specialized for
the control of specific movements that involve the hand: grasping; in
the reach-to-grasp is a movement that heavily involves intermediate ce
rebellum. The results suggest that major divisions of the cerebellum h
ave highly specialized functions in movement control.