This report compares predictions of groundwave propagation time with experimental values. The prediction methods included an integral equation technique that accounts for both terrain and impedance variations, a method developed by J.R. Wait that accounts for impedance variations, and Millington's or Pressey's technique that accounts for inhomogeneous impedance. Experimental incremental time-of-arrival data were acquired between eight sites along a propagation path from Search-light, NV. to Ft. Cronkhite (near San Francisco Bay). The major conclusions reached from the analysis and comparison of results are: terrain variations produce significant local and cumulative phase variations, in this case, a cumulative large increase in total secondary phase; impedance values need to be known to within less than a factor of two to be sufficient for preparing charts for navigation in restricted waterways; a priori definition of terrain variation can be performed with required accuracy - a priori definition of surface impedance values with sufficient accuracy is unlikely without supporting measurements; and a combination of predictions and measurements is required for accurate chart preparation. Predictions can be used to define appropriate measurement locations and to interpolate between measurement values. (Author).