Page:Handbook of Precious Stones.djvu/34

 There are several corrections which are needed before an exact result can be reached. They are these : Firstly, the stone and the water must be compared at the same temperature, usually that of 60 F. or 15.6 C. This is the most important correction and the only one usually applied ; it is well to avoid the necessity of introducing it, by conducting the experiment at the standard temperature. The second correction originates in the fact that the stone is weighed in air, and consequently is buoyed up to some extent by that fluid, appearing lighter than it would be if weighed in vacuo. The third correction depends upon the material of the weights. These, if of brass, displace from one-half to one-third of the amount of air displaced by the stone in the other pan of the balance, and consequently involve another error. The several corrections we have named may be learned with sufficient accuracy by the following methods : The correction for temperature may be applied by multiplying the difference between the weight in air and the weight in water, not by unity, but by the actual specific gravity of water at the observed temperature, then proceed with the calculation as before. The correction on account of the air and the brass weights is given by the formula :

$$y = w.0012 \left ( \frac{1}{d} - 0.12 \right ) $$

where w is the observed weight in air of a given substance ; d its approximate specific gravity; .0012 the mean density of atmospheric air; .12 the reciprocal of the specific gravity of brass, and