Page:Proceedings of the Royal Society of London Vol 2.djvu/33





N this continuation of the author's remarks on Rupert's drops, he observes, that when they are made of colourless glass, and are held before a sheet of white paper, even the unbroken drops exhibit some appearances which he considers as marks of a crystalline texture; and when they are broken in such a manner that the parts are retained in their relative situation, the structure shows a cleavage in the direction of lines diverging from the apex, or rather in the form of conical layers, having a common axis, with their apex towards the blunt end of the drop.

Since the surface of these drops is cooled suddenly by plunging into water, and fixed before the central parts are contracted, the author conceived that their specific gravity might be less than that of annealed drops; and indeed a difference was found upon trial, amounting to th part in the aggregate bulk of its external dimensions; but it appeared, on examination, that the contraction of the glass had taken place from within outwards, so as to leave cavities in the interior sufficient to account for the difference of weight, without supposing any intrinsic difference of density in the glass itself.

Dr. Brewster further observed, that by heating a drop to redness, and allowing it to cool gradually, these cavities disappeared entirely, so as to prove that they did not arise from included air, but merely from the internal contraction of the glass.

Since considerable difficulty occurs in preserving unannealed drops of flint glass, the author has paid attention to the circumstances most favourable to their preservation, and recommends removing them

Rh