Background: Ultrasound-targeted drug delivery relies on the unique nature of ultrasound contrast agents – they are microbubbles that respond strongly to ultrasound. Intravenously injected microbubbles are smaller than a blood cell. By increasing the ultrasound power, the bubbles can be ruptured at the targeted endothelial wall, locally releasing any molecules in the bubble shell. Furthermore, ultrasound-activated microbubbles are known to cause sonoporation – the process by which ultrasound drives molecules through cellular membranes. However, techniques are required to selectively detect and rupture only those microbubbles on target walls. Method: Experiments are presented on the behaviour of microbubbles on walls. For accuracy, imaging measurements are made on model microbubbles larger than contrast agents. Bubble size was varied and the resonant frequency peak determined. Results: Microbubbles on walls have a shifted frequency in good agreement with theory: a 20–25% downshift from the frequency when far from walls. Effects other than the presence of the wall account for less than 5% of the shift. Discussion: Theory predicts the frequency downshift should be sustained for actual contrast-agent sized bubbles. The effect of real, compliant cell walls requires investigation. An appropriate downshift in the applied ultrasound frequency could selectively tune gene or drug delivery. To make this feasible, it may be necessary to manufacture monodispersed microbubbles.