Magnetotelluric measurements were performed simultaneously at two sites 4.8 km apart near Hollister, California. The data obtained at each site were analyzed using the two magnetic field measurements at the other site as a remote reference. Even for data for which the E-E predicted coherencies were as low as 0.1, the apparent resistivities obtained using this technique were consistent with resistivities calculated from high coherency data at adjacent periods. Apparent resistivities calculated by a conventional analysis of the same data were biased downward by magnetic noise, sometimes by as much as two orders of magnitude.