The bathymetry and magnetic anomaly of a seamount in the Gulf of Guinea have been surveyed. The seamount is regular in shape and rises about 2000 m to a depth of 2650 m at the peak. The magnetic anomaly is well developed and consists of a low to the north and a high to the south of the peak, the field changing by 700 gammas between these points. In an attempt to duplicate the observed anomaly by model studies, it was found that despite the regularity of the observed anomaly, only a poor fit could be obtained when the seamount was modeled with a body equivalent to the topographic shape. The best fitting calculated anomaly was of shorter wavelength than the observed anomaly, suggesting that the average depth to the magnetic material is deeper than the topography indicates. Further model studies have shown that the top portion of the seamount is relatively much less magnetic than the rest and that there is magnetic material below the level of the surrounding ocean floor. The nonmagnetic top of the seamount is explained by the hypothesis of Bonatti and Nayudu, calling for intense alteration of basalt to palagonite and smectites, which have a very low magnetization. This alteration would be more intense at the top of a seamount and would thus cause a relatively nonmagnetic top. Other published seamount surveys contain many examples where the same thing appears to occur. The changes in direction of magnetization produced by the various models for the seamounts are very small; hence, the original paleomagnetic pole positions should not be in error by more than a few degrees. The paleomagnetic significance of results from seamounts is not affected.