We have analyzed glasses in eight howardites with the aim of distinguishing their origins as impact melts or pyroclasts. Although theoretical calculations predict that pyroclastic eruptions could have taken place on Vesta, the occurrence of pyroclastic glasses in HED meteorites has not been documented. This study involved petrographic examination of textures, electron microprobe analysis of major and minor elements, and LA-ICP-MS analysis for selected trace elements. Previously documented textural and compositional differences between lunar impact-melt and pyroclastic glasses partly guided this study. This work yielded no positive identification of pyroclastic glasses. The most likely explanation is that pyroclastic glasses never formed, either because Vesta contains insufficient volatiles to have powered explosive eruptions, or because eruptive conditions produced optically dense fire-fountains that allowed melt drops to collect as lava ponds. The impact-melt glasses were grouped (low-alkali, Ca-rich, and K-rich) based on compositions. We suggest that these glasses are the result of impacts onto known HED lithologies. The low-alkali glasses are impact melts of bulk HED lithologies. We hypothesize that the Ca-rich and K-rich glasses result from oversampling of plagioclase and of mesostasis that experienced liquid immiscibility, respectively, during micrometeorite impacts into eucrite targets.