PROFESSOR J. W. MALLET ON THE ATOMIC WEIGHT OF GOLD. 
40 U 
the results appear to have been sensibly the same by each method for all the gold used. 
Nor is there apparent in the results of Kruss, or those of Thorpe and Laurie, any 
evidence of a difference fairly traceable to the nature of the metal employed by them. 
A considerable part of the gold prepared by myself was, after having once served for 
a determination of the atomic weight, redissolved and reprecipitated, and was after¬ 
wards more than once used in subsequent determinations, and yet no sign was 
obtained of any resulting influence upon the later values of the atomic weight as 
obtained, evidence being thus furnished of the purity, not only of the gold itself, but 
of the reagents used to act upon it, so far as any contamination of the metal 
was concerned. It may, therefore, be concluded with reason that the gold used 
in these experiments was of uniform character, and uniformly free from any known 
impurities, to such an extent, at any rate, as to sensibly change the results obtained. 
It is to be noted that the only known elements havino- hio-her atomic weio-hts than 
that of gold are mercury, thallium, lead, bismuth, thorium, and uranium. The 
presence of any of these in the gold experimented on, even in traces too minute to 
weigh, is in a very high degree unlikely. The presence of any other element or 
elements than these would, for analogous compounds, tend to lower the value obtained 
for the atomic weight of gold; so that, in considering the chances of error due to the 
nature of the metal used as gold, we should be inclined to say that the risk was 
rather in the direction of too low than too hig^li a result beinof reached. But, if the 
possibility be observed of compounds not analogous being erroneously compared, the 
contrary error will be seen to be possible. 'Ihus, in case the composition of an auric 
haloid salt obtained from a given amount of metallic gold should be examined, if any 
unsuspected silver were present there would be required for the same amount of the 
halogen three atoms of silver insstead of one atom of gold, and, therefore, the apparent 
weight of gold as compared with that of the halogen would be increased instead of 
diminished, and a higher value obtained for the atomic weight sought. 
General Precautions Observed in the Exi?eriments for Determination of the 
Atomic Weight. 
All the reagents used were prepared or purifled by myself, and most carefully 
tested for any traces of such impurities as might reasonably be suspected, and as 
could affect their apjrlieation to the piu'pose in view. Particular care w'as bestowed 
upon the examination of the distilled water, acids, and other materials used in large 
quantity. To remove organic matter from the water required, it was distilled from a 
small amount of potassium permanganate and sulphuric acid. 
Scrupulous care to exclude atmospheric dust was observed. In the evaporation of 
some of the gold solutions the process was carried out in a glass bottle of considerably 
larger capacity than the volume of liquid to be treated, furnished with a well ground- 
glass stopper of special construction, as shown in figs. I and 2, the latter representing 
MDCCCLXXXTX.—A. 3 G 
