All minerals are supposed to be either magnetic or diamagnetic, and in a large number, exact measurements have been made of the relative amount of magnetic three in each mineral. As was to be expected, the compounds of Iron and Nickel are, in the great majority of cases, the cause of the magnetism in mineral ; but there exists a large border land of slightly magnetic mineral substances, the cause of whose magnetism has never been very clearly shown. I do not intend, in this paper, to deal with those which are only faintly magnetic, but with those whose magnetism is quite distinct, such as Obsidian, Basalt, &c. The method used for testing the relative amount of magnetism in any specimen was as follows.