Neuropeptide immunohistochemistry was used to test several hypotheses of the anatomical bases of chorea and rigidity-akinesia. To test the hypothesis that elevated concentration of striatal somatostatin causes chorea, we visually compared the density of striatal neurons containing somatostatin and neuropeptide Y in brains affected by choreic or rigid-akinetic Huntington's disease (HD). The density of these neurons was elevated in both rigid-akinetic and choreic HD specimens with an apparently normal total number of these neurons, indicating that elevated somatostatin concentration, by itself, does not lead to chorea. We tested the hypothesis that rigid-akinetic HD results from deficient dopaminergic nigrostriatal neurotransmission by examining tyrosine hydroxylase-immunoreactive (TH-IR) neurons in the substantia nigra. In rigid-akinetic HD brains, there was no obvious reduction of nigral TH-IR neurons, indicating that rigid-akinetic HD is probably not due to loss of nigral dopaminergic neurons. Finally, we also examined the status of striatal projection neurons and found near total loss of all striatal neurons projecting to the lateral globus pallidus, medial globus pallidus, and substantia nigra in brains affected by rigid-akinetic HD in contrast to the preservation of neurons projecting to the medial globus pallidus in choreic HD. These results are consistent with the hypothesis that chorea results from preferential loss of striatal neurons projecting to the lateral globus pallidus and that rigid-akinetic HD is a consequence of the additional loss of striatal neurons projecting to the medial segment of the pallidum.