Eighteen 200-ha study plots were established in the Monongahela National Forest, West Virginia, and in the George Washington National Forest, Virginia. The overall purpose of the study was to determine nontarget effects of biological insecticides used to control gypsy moth, Lymantria dispar (L.). From 1995 through 1998 blacklight traps, foliage pruning, and canvas bands were used to collect lepidopteran adults and larvae to determine population abundance 2 yr before and during two consecutive years of treatment applications. During 1997 and 1998, six plots were each aerially treated with nucleopolyhedrosis virus Gypchek (GC) (US Forest Service, Ansonia, CT) and Bacillus thuringiensis variety kurstaki (Btk). The remaining six plots were left untreated (control). Counts of adults and larvae of 19 species from five families were tallied to assess treatment effects. Both spring- and summer-defoliating larvae were sampled. Analysis of variance indicated a significant interaction between pretreatment/treatment years and the three treatment groups for the adults of two species and the larvae of three species. Fewer larvae were collected from Btk plots than from GC and control plots during treatment years, but not during pretreatment years. Gypchek was not determined to adversely affect nontarget species. Adults of 10 species and larvae of four species were more numerous during treatment years than pretreatment years. Adults of four species and larvae of five species were more numerous during pretreatment years than during treatment years.