Our small experiences with electrical stimulation in the VPL and VPM for dysesthetic pain show that it provoked only paresthesia and induced some relief of pain. It does not increase the beta-endorphin level in CSF. To clarify the anatomical substrata in VPL stimulation, neuroanatomical studies were done about the inputs to VPL in man, monkey and cat by the Fink-Heimer method. The spinothalamic tract terminates in VPL in a patchy fashion in the monkey. The corticothalamic fibers from SI and SII cortex project to VPL and VPM in somatotopical organization in the cat. SI and SII cortices have reciprocal connections, in addition to projections to area 5 or SIII cortex. The corticofugal fibers to the magnocellular and gigantocellular tegmental fields are suggested in addition to the dorsal column nuclei, spinal trigeminal nuclei and spinal posterior horn in cat. The medial lemniscus input to VPL and the above neural circuits are thought to be associated with VPL stimulation.