A battery of tests of peripheral and central nervous system function was administered to 205 former workers of a large heavy industrial plant, 104 of whom were previously exposed to inorganic mercury. The mean age of those examined was 71 years. Exposed subjects had participated in a urine-mercury exposure monitoring program during the time of operation of a process that required the use of mercury and its subsequent clean-up. Mercury exposure had been high (mean peak urine mercury concentration was >600 microg/l) and had ended 30 years or more prior to the investigation. Peripheral nerve function outcomes that were statistically significantly associated with cumulative mercury exposure after controlling for covariates included classification as having peripheral neuropathy, peroneal motor nerve conduction velocity, ulnar motor nerve conduction velocity, and peroneal motor nerve F-wave latency. Quantitative assessment of resting tremor was nearly significantly associated with cumulative mercury exposure (p=0.07). Among tests of central nervous system function, results of the Handeye Coordination test were significantly associated with cumulative mercury exposure after controlling for covariates. Cumulative mercury exposure was not observed to be associated with a quantitative measure of dementia or with a number of cognitive neurobehavioral test outcomes. The statistically significant associations with mercury exposure were observed in spite of greater mortality among the exposed group than the unexposed group. These results suggest that substantial occupational mercury exposure can have long-term adverse effects on the peripheral nervous system detectable decades after cessation of exposure. Such long-term adverse effects were not observed for a measure of dementia or other measures of cognitive function.