"4. The specific weight of this piece was determined by a good hydrostatical balance, and being, for the greater certainty, compared to coined and to other very pure gold, used by M. Buffon in his experiments, their density was found, with water, in which they were plunged, Pure gold - 19-87/1 Coin gold - 17-1/2 Platina - 14-2/5 "5. This piece of platina was put upon steel to try its ductability; it supported the hammer very well for a few strokes; its surface became flat and even, a little smooth in the parts which were struck, but it split soon after, and nearly a sixth part separated. The fracture presented many cavities, some of which had the whiteness and brilliancy of silver, and in others we remarked several points like chrystalization; the tops of these points examined with the lens, was a globule absolutely similar to that of the third experiment. All the other parts of this piece of platina were compact, the grain finer and closer than the best brass, which it resembled in colour. We offered several of these pieces to the loadstone, but not one was attracted thereby. We powdered them again in an agate mortar, and then remarked that the magnetical bar raised up some of the smallest every time they are placed under it. "This new appearance of magnetism was so much the more surprising, as the grains were detached from the agglutinated mass of the second experiment, which seemed to have lost all sensibility at the approach and contact of the loadstone. In consequence we again took some of these grains, which were alike powdered, and soon perceived the smallest parts sensibly attach themselves to the magnetic bar. It is impossible to attribute this effect to the smoothness of the bar, or to any other cause foreign to magnetism. A piece of smooth iron, applied in the same manner on the parts of this platina did not raise up a single grain. "By these experiments, and the observations which have arisen therefrom, we may judge of the difficulty of determining the nature of platina. It is very certain that it contains some parts which are verifiable even without the addition of a fierce fire; it is also certain that all platina contains iron and attractable parts; but if the Prussian alkali never affords blue but with the grains which the loadstone attracts, we should conclude, that those which resist it are pure platina, which of itself has no magnetical virtue, and of which iron does not make an essential part. We must suppose that a sufficient fusion, or perfect cupellation, might decide the question; at least, these operations appear to have, in fact, deprived it of every magnetic virtue, by separating it from all foreign bodies; but the last observation proves, in an incontrovertible manner, that this magnetic property was, in reality, only weakened, and perhaps masked or buried, since it reappeared when it was ground."