BACKGROUND: Aphids, including the generalist herbivore Myzus persicae, transmit cucumber mosaic virus (CMV). CMV (strain Fny) infection affects M. persicae feeding behavior and performance on tobacco (Nicotiana tabacum), Arabidopsis thaliana and cucurbits in varying ways. In Arabidopsis and cucurbits, CMV decreases host quality and inhibits prolonged feeding by aphids, which may enhance virus transmission rates. CMV-infected cucurbits also emit deceptive, aphid-attracting volatiles, which may favor virus acquisition. In contrast, aphids on CMV-infected tobacco (cv. Xanthi) exhibit increased survival and reproduction. This may not increase transmission but might increase virus and vector persistence within plant communities. The CMV 2b counter-defense protein diminishes resistance to aphid infestation in CMV-infected tobacco plants. We hypothesised that in tobacco CMV and its 2b protein might also alter the emission of volatile organic compounds that would influence aphid behavior. RESULTS: Analysis of headspace volatiles emitted from tobacco plants showed that CMV infection both increased the total quantity and altered the blend produced. Furthermore, experiments with a CMV 2b gene deletion mutant (CMV∆2b) showed that the 2b counter-defense protein influences volatile emission. Free choice bioassays were conducted where wingless M. persicae could choose to settle on infected or mock-inoculated plants under a normal day/night regime or in continual darkness. Settling was recorded at 15 min, 1 h and 24 h post-release. Statistical analysis indicated that aphids showed no marked preference to settle on mock-inoculated versus infected plants, except for a marginally greater settlement of aphids on mock-inoculated over CMV-infected plants under normal illumination. CONCLUSIONS: CMV infection of tobacco plants induced quantitative and qualitative changes in host volatile emission and these changes depended in part on the activity of the 2b counter-defense protein. However, CMV-induced alterations in tobacco plant volatile emission did not have marked effects on the settling of aphids on infected versus mock-inoculated plants even though CMV-infected plants are higher quality hosts for M. persicae.