The first technique estimates a set of values for the magnetic intensity at each flight-line/tie-line intersection point. These are obtained after an intersection-moving procedure to reduce loop closures and are subject to the constraint that the datums calculated from the differences between the raw data and the corrected data have minimum along-line variation. This least-squares method is equivalent to the older manual loop closure methods used for magnetic, gravimetric, and geodetic surveying. The second method involves the fitting of polynomials to the observed flight-line/tie-line intersection errors along each line in the survey. These are then subtracted from the raw data and intersection locations are adjusted to give minimum intersection errors. The least-squares method was found to be superior as a leveling procedure. It produced datums which were similar in amplitude and had a high correlation with the base station magnetometer recordings. The polynomial method produced datums which had greater amplitudes and showed a tendency to fluctuate at the ends of lines.--Modified journal abstract.