This study is part of a long-term analysis of nontarget effects of microbial pesticide application in the George Washington (Augusta County, VA, USA) and Monongahela National Forests (Pocahontas County, WV, USA). Ants were collected using pitfall traps to assess the effect of Bacillus thuringiensis Berliner variety kurstaki (Foray 48 F) and gypsy moth nuclear polyhedrosis virus (Gypchek) application on ant communities. Ant samples were also compared by sampling years. Pitfall traps were operated for 45 wk during summers of 1995–1997. A total of 31,732 ants was collected from pitfall traps; they belonged to four subfamilies, 17 genera, and 31 species. The ant species richness, diversity, abundance, and species composition did not change as a result of the treatments. Further tests of ant abundance were suggested because the test power was low. Comparisons between sampling years showed a very similar species composition and species evenness. There was a significant decrease in ant abundance in the third year of sampling, which might have been caused by over-trapping. Some rare species did not appear in the second and third year of sampling.