A tumbler which had been toughened in Monsieur de la Bastie's works, was, in my presence, thrown upon the ground, yet it did not break. It was a large soda water glass. I kept it for some time, and after considering the matter carefully, I felt, that if it were thrown down in such a way that the whole of its side, from base to rim, came in contact with the ground at once, and it then stood this test, it would prove that the whole of the glass was in the condition of the Rupert's Drops, and would therefore bear the concussion without fracture. I held the glass and let it fall, so that it actually reached the hard floor on its side. It immediately broke all to pieces. Now on the first occasion when this glass was thrown down, it was tossed somewhat upwards into the air, and the bottom being heavier reached the ground first, and it did not break. I have also seen in glass-houses, where the tempering process is carried on, tumblers thrown down in a similar manner, and I noticed, that whenever they fell upon their bottoms, they were uninjured, as also in cases where they fell upon their rims in such a manner, that the curve of the rim acted as an arch, as in the old trick of turning a wine-glass off the table so as not to break; but in other cases where the tumblers fell flat upon their sides, fracture followed. I carefully gathered together the pieces of the large tumbler which I broke myself in this manner, and examined them, and found that the solid bottom was broken in the same manner as the Prince Rupert's drops break, viz., into a large number of small pieces, having in all respects similar properties. The glass for an inch or two above the bottom broke into small pieces, but larger than those into which the bottom itself broke, and the upper portion of the tumbler was fractured just as an ordinary tumbler would be. On careful examination, microscopic and otherwise, the small pieces were found to have the character of Prince Rupert's, whereas the larger from the upper part of the glass had none of these characteristics in the slightest degree. These observations led me to perform an experiment. A toughened tumbler was filled with plaster of Paris, which was allowed to set. Its outside was then encased in plaster of Paris, and when the whole was hardened, a pair of pincers were applied to a portion of the tumbler's rim, and with a violent wrench the tumbler was broken. A rather smart shock was communicated to the arm of the operator, very much resembling, as he said, the shock of an electrifying machine. On removing the plaster of Paris, it was found that the whole of the tumbler was fractured, and, as will be seen by the accompanying illustration, in a manner similar to that which has already been described. From this and other similar experiments, I was led to the conclusion that none of the toughened articles which have cavities in them, have thoroughly undergone the toughening process.