A harmonic correlation technique has been used effectively to isolate a primary signal from two independent data sets each of which is contaminated by the presence of interfering signals. Each of the two data sets is decomposed by a harmonic analysis, and then the correlation coefficients and amplitude ratios between the two decompositions are computed for each harmonic. Only harmonics which satisfy predetermined selection criteria are retained. Those criteria are in the form of requiring the correlation coefficient to exceed some magnitude, say, &rgr;0, and the amplitude ratio not to exceed some value, say, R0. The resulting pair of retained signals are then typically highly correlated. It is suggested that appropriate values of &rgr;0 lie between 0.25 and 0.6 and of R0 between 3.0 and 6.0, depending on the relative amplitudes of the primary and interfering signals. Critical examination of the technique shows that under many conditions it can be an effective tool for eliminating signals which are not a good approximation to the primary signal. This conclusion is confirmed by simulated examples of satellite magnetic anomaly data with interfering fields from ionospheric currents.