This paper presents experimental results demonstrating the feasibility of high frequency ultrasonic sensing and sorting for screening single oleic acid (lipid or oil) droplets under continuous flow in a microfluidic channel. In these experiments, hydrodynamically focused lipid droplets of two different diameters (50 μm and 100 μm) are centered along the middle of the channel, which is filled with deionized (DI) water. A 30 MHz lithium niobate (LiNbO(3)) transducer, placed outside the channel, first transmits short sensing pulses to non-invasively determine the acoustic scattering properties of the individual droplets passing through the beam's focus. Integrated backscatter (IB) coefficients, utilized as a sorting criterion, are measured by analyzing the received echo signals from each droplet. When the IB values corresponding to 100 μm droplets are obtained, a custom-built LabVIEW panel commands the transducer to emit sinusoidal burst signals to commence the sorting operation. The number of droplets tested for the sorting is 139 for 50 μm droplets and 95 for 100 μm droplets. The sensing efficiencies are estimated to be 98.6% and 99.0%, respectively. The sorting is carried out by applying acoustic radiation forces to 100 μm droplets to direct them towards the upper sheath flow, thus separating them from the centered droplet flow. The sorting efficiencies are 99.3% for 50 μm droplets and 85.3% for 100 μm droplets. The results suggest that this proposed technique has the potential to be further developed into a cost-effective and efficient cell/microparticle sorting instrument.