We investigated some important factors which affect the transdermal flux of ketoprofen, a nonsteroidal anti-inflammatory agent, as a first step to provide some basic knowledge for the development of a iontophoretic transdermal patch system. Factors such as current density, polarity, buffer (HEPES) and electrolyte concentration and pH were studied using hairless mouse skin. The effect of poly(L-lysin), which is known to affect the electro-osmotic flow through skin, on flux was also studied. Passive flux was about $20;{mu}g/cm^2hr$ at pH 4.0, but was negligible at pH 7.4 where all ketoprofen molecules dissolved are ionized (ketoprofen pKa=5.94). At pH 4.0, application of anodal current increased the flux further above the passive level, however anodal flux at pH 7.4 was much smaller than passive flux at pH 4.0. The application of cathodal current at pH 4.0 increased the average flux to $30-40;{mu}g/cm^2hr$, depending on the current density applied. At pH 7.4, cathodal flux was only about $5;{mu}g/cm^2hr$. Decrease in buffer and electrolyte concentration increased this cathodal flux about 10 fold. However decrease in HEPES buffer concentration 100 fold did not affect the flux. Anodal flux of acetaminophen was much larger than cathodal flux, indicating that electroosmotic flow can be playing an important role in the flux. Poly(L-lysin) increased the cathodal flux at pH 7.4. These results provide some important insights into the mechanism of transdermal flux of ketoprofen and the role of electroosmotic flow.