Pallidotomy paradoxically reduces the intensity of levodopa-induced dyskinesia without worsening motor symptoms. The reasons for this are not clear and no experimental study has investigated this phenomenon. The objective of this investigation was to evaluate the effects of unilateral pallidotomy on locomotor activity, motor disability and levodopa-induced dyskinesia in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated levodopa-primed common marmosets. Animals were primed to exhibit dyskinesia by daily administration of levodopa until stable dyskinesia was evoked by each dose. Locomotor activity, motor disability and dyskinesia were assessed weekly at baseline and following an acute levodopa challenge. Prior to pallidotomies, two distinct groups of animals emerged: poor responders to levodopa with mild dyskinesia (Group 1) and those exhibiting a marked increase in motor activity and pronounced dyskinesia (Group 2). Electrolytic lesions were placed in the left internal segment of the globus pallidus. Pallidotomy had no effect on basal or levodopa-induced motor activity in either group but significantly improved basal motor disability in Group 2. Following pallidotomy, the ability of levodopa to reduce motor disability was significantly increased in both groups. Pallidotomy improved dyskinesia in both Groups 1 and 2 but it was more effective in reducing dystonia compared with chorea. The effect of pallidotomy on dyskinesia in Group 2 was transient, with the intensity of involuntary movements reverting to presurgery levels 4 weeks later. This study shows that in levodopa-primed, parkinsonian marmosets, placement of discrete globus pallidus lesions can ameliorate levodopa-induced dyskinesia but not akinesia. This model allows the evaluation of pallidotomy-induced biochemical changes in dyskinetic primates.