[ 12 5 1

III. Of estimating the fineness of gold from its gravity.

The great excess of the weight of gold, above that ofthe metals used for its alloy, affords another method ofjudging of the quantity of alloy or debasement, in any givenmixture where the species of alloy is known.

It may here be proper to caution the reader against anerror which has sometimes been fallen into, in computingthe specific gravities of mixts from those of their ingredi-ents. If the gravity of one metal was nine, and that ofanother eighteen, it has been inadvertently reckoned thatthe gravity of a mixture of equal parts of the two wouldbe the medium between nine and eighteen, or thirteenand a half. If by equal parts were meant equal bulks, thiswould indeed be the cafe but when the parts are taken-by weight, as they are always understood to be in mix-tures of this kind, it is otherwise.. For eighteen weightsof the one metal,, on being immersed in water, will losetwo, and eighteen of the other will lose one ; so thatthirty-six.of the mixt will lose three : whence the specificgravity (which is found by dividing the weight in air bythe loss in water) instead of being thirteen and a half, turnsout but twelve.

Fine gold, as we have seen before, loses in water onegrain in every nineteen and three tenths nearly; whereasfine silver loses one grain in about eleven : from whenceit is easy to find the loss of any number of grains ofeach, and consequently of any assignable mixture of thetwo metals. Thus fifty grains of gold will lose abovetwo and a half, and fifty grains of silver somewhat morethan four and a half; so that a mixture of equal partsof the two will lose above seven in a hundred, or one infourteen. In like manner, a mixture of gold with halfits weight of silver will be found to lose one part in

fifteen