An operational study has been made of algorithms employed in short base line underwater position location systems. Some important sources of systematic error have been uncovered. The issues studied include isospeed vs. isogradient ray tracing, effect of the depth velocity profile and water layer thickness, approximate vs. exact array tilt corrections, and ray tracing initialization methodology. It is shown that the practice of constant speed extrapolation of depth-velocity information can cause considerable mischief. The best remedy is to measure speed all the way to the bottom. It is further shown that the systematic errors are periodic functions of the azimuth direction of the sound ray from the receiver array. The amplitudes of these functions are greater for the more severely tilted arrays. An alternative algorithm is proposed that reduces these errors by at least an order of magnitude. Keywords: Ray tracing; Calibration; Underwater tracking; Short baseline systems; Systematic errors. (JHD)