Observations at regular intervals of the location of newly hatched grape root borer, Vitacea polistiformis (Harris), larvae moving freely within circular petri dish bioassays were used to measure and compare their response to dry filter paper discs treated with ethanol- or hexane-based extracts of roots from known and potential Vitaceae hosts and a nonhost. Larvae responded most strongly to discs treated with ethanol extracts, suggesting the presence of behaviorally active, polar compounds associated with roots. In single extract bioassays comparing extract versus solvent treated discs, larvae responded positively to ethanol extracts from all Vitis species and rootstocks and Virginia creeper [Parthenocissus quinquefolia (L.) Planch.], but not to apple (Malus domestica Borkh). Paired extract bioassays, in which an extract from the commercially important 3309 rootstock was used as the standard and presented simultaneously with extracts from other root sources, revealed examples of equal, significantly weaker and significantly stronger responses to the 3309 extract. Extracts of the 420 A and V. riparia ‘Gloire’ rootstocks appeared to possess qualities that elicited a consistently greater response than to 3309 extract in these pair-wise comparisons. The active compounds were eluted in ethanol during a 30-min extraction; larvae responded equally to 30- and 60-min 3309 root extracts in paired extract bioassays. Larvae responded equally to extracts of 3309 roots from three spatially separate vineyards in northern Virginia. These results are discussed in relation to the subterranean, plant-insect interactions of grape root borer neonates with the numerous native and non-native Vitis species that may serve as hosts in the eastern United States.