Seismic first arrival times from crosshole, VSP, and reversed VSP (RVSP) experiments are collectively inverted by least‐squares for the velocity distribution between two boreholes. The tomographic reconstruction exhibits a large lateral velocity contrast that is not supported by the surface reflection data from the same location. After examining the traveltime residuals from the three tomographic datasets separately, we conclude that the velocity contrast is due primarily to static delays in the RVSP first arrival times. When a first‐order correction is made for the statics, tomographic inversion results in a velocity reconstruction that is more consistent with the surface reflection data. To isolate the velocity errors produced by the RVSP statics, we compute a residual tomogram by subtracting the statics adjusted tomogram from the original. The residual tomogram shows that the statics introduce errors not only in the region sampled by the RVSP rays, but they indirectly contaminate other regions of the tomogram as well. We reproduce this velocity error distribution as part of a model study designed to simulate the effects of statics on tomographic velocity reconstructions. Results indicate that traveltime errors on the order of 2 percent can result in tomographic velocity errors of up to 7 percent.