The rates of the select sample of Alzheimer's disease cases are adjusted according to age and sex in order to facilitate fair spatial comparisons between the twelve geographical regions. Although there are small deviations in patterns of significance with time, the patterns revealed generally are consistent.

The t-test of significance is applied to log transformations of thirty-one parameters contained in the drinking water quality data set. It is found that those areas that correspond to the significant deviations in Alzheimer's disease occurrence also tend to reveal significant deviations in average water quality at the 95 percent and 99 percent levels.

Simple linear regression reveals that the essential trace element, arsenic, maintains the only significant inverse spatial relationship with Alzheimer's disease. Factor analyses and stepwise multiple regression/correlation procedures reveal that three qualities of drinking water which include essential trace elements, total water hardness, and salinity, account for 93.30 percent of the total explained variance in disease occurrence. There is a significant inverse spatial association between disease occurrence and concentrations of essential trace elements and the degree of water salinity. A direct association is discovered pertaining to disease and total water hardness. (Abstract shortened by UMI.)