The pars compacta and pars dissipata of the pedunculopontine nucleus contain cholinergic cell group Ch5, and the laterodorsal tegmental nucleus contains cholinergic cell group Ch6. The pedunculopontine nucleus has been implicated in a variety of functions, including mediation of rapid eye movement sleep and in extrapyramidal motor function, although the role of cholinergic and non-cholinergic neurons is unclear. Quantitative neuroanatomical techniques were used to map the distribution of cholinergic neurons in the mesopontine nuclei of the adult human brain. In addition, the number and distribution of comparably sized non-cholinergic neurons at selected anatomical levels were compared. An antibody raised against human choline acetyltransferase was used to stain immunohistochemically the mesopontine neurons in six brains, ranging in age from 28 to 60 years. The rostrocaudal length of the Cb5/Ch6 cell complex was approximately 10 mm. The estimated total number of cells was similar for all brains, and varied by less than 7%. The estimated average number of cholinergic cells in the combined pedunculopontine and laterodorsal tegmental nuclei was approximately 20,000, with 30% of the cells in the pedunculopontine nucleus pars compacta, 57% in the pedunculopontine nucleus pars dissipata and 13% in the laterodorsal tegmental nucleus. There was no correlation between cell number and age. Within areas of mesopontine tegmentum occupied by the Ch5 cholinergic neurons, there were often more noncholinergic neurons than comparably sized cholinergic neurons. The present study provides detailed maps of the distribution and number of mesopontine cholinergic neurons in the normal human brain. Many non-cholinergic neurons are intermixed with the cholinergic pedunculopontine neurons. One region of the pedunculopontine nucleus pars dissipata containing few cholinergic neurons, located adjacent to the ventral border of the pedunculopontine nucleus pars compacta, may correspond to the midbrain-extrapyramidal area as defined previously in rodent and in non-human primate. These data will be useful for quantitative neuropathological studies concerning the role of both cholinergic and non-cholinergic mesopontine neurons in diseases proposed to affect these neurons, including Parkinson's disease, schizophrenia and progressive supranuclear palsy.