The microsporidium Thelohania solenopsae Knell, Allen, and Hazard was released into colonies of red imported fire ants, Solenopsis invicta Buren, at four field sites in Louisiana. Social form of the ant affected the establishment of the microsporidium; long-term epizootics developed at two sites with predominantly polygyne populations, whereas the pathogen produced a low disease prevalence before it disappeared or did not infect ants at all at two monogyne sites. This study is the first report of artificial infection of monogyne S. invicta field colonies. When the microsporidium became established in mixed monogyne/polygyne ant populations, prevalence rates peaked at >75% in both social forms. In mixed ant populations, the social form suffering higher prevalence of the disease decreased proportionally to the other form, possibly indicating a competitive disadvantage. Host population density or site characteristics may have influenced spread of the disease; the rate of spread was 0.9 m/mo at one epizootic site and reached 9.4 m/mo at the other. There was little seasonal variation in prevalence, which averaged 47% in February, 51% in April/May, and 57% in October/November at the two epizootic sites. The strongest evidence of microsporidian impact on S. invicta populations was a negative correlation between colony numbers versus percentage infection and a sporadic decrease in the number of foragers. There was some evidence of a decrease in the size and number of colonies at one epizootic site and a decrease in brood at the other. This sporadic weakening of the S. invicta populations did not lead to significant immigration of other ant species.