Page:Memoirs of the Geological Survey of Great Britain, Volume 1.djvu/455

This page needs to be proofread.
442
INFLUENCE OF MAGNETISM ON CRYSTALLIZATION

442

INFLUENCE OF MAGNETISM ON CRYSTALLIZATION

seen that this influence is capable of being bent out of the line of force by the proximity of a body of the magnetic class, although not a magnet in the usual acceptation of the term. We shall have, however, to con- sider eventually the manner in which the chemical, crystalline and molecular forces react upon magnetic power ; and it is highly probable that we may, by a more extensive examination of the subject, arrive at the knowledge of some disposing forces, or peculiar properties of matter different firom those with which we are at present acquainted.

16. In my paper on this subject already mentioned,* I have described an experiment, in which silver was precipitated from its solution by the proto-sulphate of iron, and I have there mentioned and given a drawing of two curiously formed curved spaces, crossing the curves produced by finely divided silver. In my first experiments, the protosulphate of iron was placed on the glass plate in the centre of the poles, and the solution of silver poured over it, and then the above-mentioned curves commenced from die dissolving salt, and proceeded in the line of the magnetic curves towards the opposite pole, being checked at a short dis- tance from it I have since found that t^e experiment is much more satisfactorily tried, by filling two tubes closed at one end with a satu- rated solution of the iron salt, and then inverting these over the poles, and steadily supporting them there during the experiment. By this means the heavier fluid (sulphate of iron) is gradually difiused through the lighter one (nitrate of silver) for a considerable distance around the magnetic poles. When the experiment is -made in this way the result is as follows : —

Fig. 8.

The first action, which takes place very slowly, is shown by a cloudiness around the poles, which gradually arranges itself in curves radiating

» * Philosophicml Mmgnrine, Jmnuary, 1846, toI. jlxtHI. p. 4.

Digitized by

Google