Every body, not ferro-magnetic, places itself across such a line of magnetic force as we have conceived; and if the earth were made up of separate layers of ferro-magnetic and diamagnetic bodies, the result would be the formation of bands at right angles to each other. This is not the case, by reason of the intermingling of the two classes of substances. Out of the known chemical elements we find only about ten which are actively ferro-magnetic; the others combining with these give rise to either a weaker state, a neutral condition, or the balance of action is turned to the diamagnetic side. Sulphate of iron, for instance, is a magnetic salt; but in solution, water being diamagnetic, it loses its property. The yellow prussiate of potash dissolved in water is a diamagnetic body; but the red prussiate, which contains an atom less of potassium, is magnetic: but in the solid state they are both diamagnetic.[189] From this it would appear that the chemical composition of a body regulated its relation to magnetism. The following facts will show, however, that the molecular structure is more particularly concerned in determining the molecular condition of substances. M. Plücker, being desirous of finding the extent to which the _direction of the fibres_ in organic bodies might influence their magnetic or diamagnetic properties, was led to inquire whether in crystals the direction of the optic axes, which itself depends upon the arrangement of the particles, might not also exercise some influence. The first submitted to the action of the electro-magnet a thin plate of tourmaline, such as is employed in experiments upon polarization, having its optic axis parallel to its longest length. It was very quickly perceived that the plate was magnetic, by the effect of the iron that it contains; but it was suspended successively in three ways,--first, so that its longest side was vertical, then as that the shortest side was vertical, and finally so that the plate itself was horizontal. In the first case it is directed between the two points of the conical curvatures of the poles like a magnetic body; but, in the other two cases, on the contrary, it took the direction assumed by diamagnetic bodies--that is to say, a direction such that its longest length was perpendicular to the line joining the poles. This direction indicated that the optical axis was repelled by the two poles, and that this repulsion outweighed the magnetic properties of the crystal.[190] The relation of structure to physical phenomena of essentially different characters is remarkable. Savart, when making crystalline plates of quartz and carbonate of lime vibrate, succeeded in determining a relation between the acoustic figures that are produced in them, and the particular mode of the crystallization of the substance. He found that the direction of the optical axis is constantly connected with that of the principal forms of the acoustic figures.