Splitter plates are commonly used to minimize the influence of the wind tunnel floor boundary layer when testing a semispan wing. Little information about the flow over a splitter plate exists in the literature. This paper discusses two separate experiments designed to study the effects of several variables on the flow over a splitter plate to obtain guidelines for the design of splitter plates. The first experiment investigated the effects of the streamwise length and profile of the splitter plate leading edge. Significant separated flow was observed on the leading edge of the splitter plate, and the size of this separated region was seen to grow as the length of the leading edge section of the splitter plate was reduced. The separation effects were reduced through the use of a streamlined leading edge profile. The second experiment investigated the effects of the tunnel floor boundary layer thickness on the flow over the splitter plate. A boundary layer suction/blowing system was developed that could significantly alter the empty tunnel floor boundary layer. Boundary layer measurements on the splitter plate showed the flow on the splitter was relatively insensitive to the tunnel floor boundary layer justifying the use of such a device. This paper also discusses the influence of the geometry of the shroud beneath the plate and the angle of attack of the shroud. Several guidelines are given regarding the design of splitter plates.