A laboratory bioassay was developed to evaluate miticides to control Varroa jacobsoni (Oudemans), an important parasite of the honey bee, Apis mellifera L. Bees and mites were exposed to applications of essential oil constituents in petri dishes (60 by 20 mm). The registered mite control agents tau-fluvalinate (Apistan) and formic acid also were evaluated as positive controls. Treatments that caused high mite mortality (> 70%) at doses that produced low bee mortality (< 30%) were considered mite selective. The six most selective of the 22 treatments tested (clove oil, benzyl acetate, thymol, carvacrol, methyl salicylate, and Magic3) were further evaluated to estimate LD50 values and selectivity ratios (A. mellifera LD50/V. jacobsoni LD50) at 24, 43, and 67 h after exposure. Tau-fluvalinate was the most selective treatment, but thymol, clove oil, Magic3, and methyl salicylate demonstrated selectivity equal to or greater than formic acid. The effect of mode of application (complete exposure versus vapor only) on bee and mite mortality was assessed for thymol, clove oil, and Magic3 by using a 2-chambered dish design. Estimated V. jacobsoni LD50 values were significantly lower for complete exposure applications of thymol and Magic3, suggesting that both vapor and topical exposure influenced mite mortality, whereas estimated values for clove oil suggested that topical exposure had little or no influence on mite mortality. These results indicate that essential oil constituents alone may not be selective enough to control Varroa under all conditions, but could be a useful component of an integrated pest management approach to parasitic mite management in honey bee colonies.