From studies of about 200 samples of metamorphic and igneous rocks from the Adirondacks it has been possible to relate the magnetic oxide assemblage to the lithology of the rocks and to their magnetic properties so that the magnetic anomaly produced by a particular rock type can be estimated. The intermediate members of the Fe 2 O 3 -FeO.TiO 2 -(TiO 2 ) system are found almost uniformly to have reverse remanent magnetism and as much as several percent of disseminated magnetic oxides of this system may give rise to intense negative aeromagnetic anomalies. Where both magnetite and magnetic oxides of the Fe 2 O 3 -FeO.TiO 2 -(TiO 2 ) system occur together in the same rock the resultant magnetic properties are such as to indicate that the magnetite grains have normal or positive remanent magnetization and the grains of the magnetic oxides of the Fe 2 O 3 -FeO.TiO 2 -(TiO 2 ) system have reverse or negative remanent magnetization. The value of the reverse magnetism of titanhematite, ilmenohematite, and rutiloilmenohematite is such that it can neutralize the total positive magnetism (remanent magnetism plus induced magnetism) of an equivalent or greater amount of magnetite. In the metamorphism of a pyroxene gabbro, giving a positive magnetic anomaly, to a hornblende-plagioclase amphibolite, the primary magnetite may be taken up by the hornblende, leaving ferrian ilmenite as the only oxide and the amphibolite may then give a neutral or negative anomaly.Various investigators have suggested that the reverse magnetization of rocks is produced by a reversal of the earth's magnetic field at the time they cooled through the Curie point. This hypothesis does not seem to fit the geological, chemical, and magnetic data for the Adirondack rocks, for the reverse magnetization here appears to be related to the content of the intermediate members of the Fe 2 O 3 -FeO.TiO 2 -(TiO 2 )system. These mineral mixtures have the property of "self-reversal," that is, the ability to become magnetized in a direction opposite to that of the existing magnetic field. Until the existence of "self-reversal" can be disproved, one cannot state categorically that the reversed magnetization of a rock has been produced by a reversed field of the earth.