The paper reports results of experiments with oil-in-water emulsion, which were carried out in a laboratory crossflow microfiltration unit equipped with backflushed zirconia ceramic membranes. The results demonstrate that the membrane backflushing can maintain the permeate flux at a level which is nearly three-fold over the long-term flux in the absence of membrane backflushing. It was observed that the effect of backflushing was the more pronounced when the backpulse duration was shorter, the transmembrane pressure difference was higher, and the retentate velocity was lower in forward filtration. An optimum backflushing frequency which maximized the average permeate flux was found to be in the range of 1 to 50 s depending on the operating conditions. The magnitude of the transmembrane pressure difference in the reverse flow had a relatively small effect. Attempt has also been made to explain the results in terms of a simple semiempirical model of the process. Parameters evaluated from dynamic and steady state experiments without membrane backflushing were shown to be useful in estimating the performance of the process with membrane backflushing. The influence of backflushing duration and frequency, transmembrane pressure difference, and retentate velocity on average permeate flux were well predicted using this model.