408 PROFESSOR J. W. MALLET OR THE ATOMIC WEIGHT OF GOLD. 
steel rolls quite free from grease, and the strip finally cleaned for use with, hot hydro¬ 
chloric and then nitric acid. 
In a letter of later date he wrote, “ In preparing the ‘ proof ’ I seldom make over 
10 oz, in one lot; from 8 to 10 oz. is the usual amount. There is comparatively little 
trouble in making 999‘9 fine, hut beyond that it is rather troublesome; and it seems 
that, with all the care, the final result is sometimes a little in doubt. ” 
The fine gold received from the Philadelphia Mint is designated as B in this paper, 
in connection with the experiments in which it was used. 
8 . Purification of “ Proof Gold ” obtained from the United States Assay Office at 
Neiv York. — Br. H. G. Torrey, Chief Assayer in this office, was obliging enough to let 
me have several samples of his finest proof gold, used in checking the regular assays 
in his department. He informed me that this proof gold was independently prepared 
at New York, but was occasionally compared with that of the Philadelphia Mint. He 
furnished the following bi’ief statement as to its preparation—The process used in 
preparing the gold is to dissolve ‘ cornets ’ (or gold from assays) in nitrodrydrochloric 
acid, and after filtration precipitating by oxalic acid, and after thorough washing- 
melting under borax. The operation is conducted with the utmost care throughout.” 
The gold from this source is designated as C in this paper. 
4. Gold from the “ Trial Plate ” of Fine Gold of the English Mint .—Professor 
PiOBERTS-Austen, Chemist to the Royal Mint, was so kind as to let me have a specimen 
ol a few grammes of gold cut from the trial plate of the pure metal prepared by him in 
1873, In its preparation use was made of potassium chloride and alcohol to separate 
any platinum present in the original material, a long 2 :)eriod of subsidence was allowed for 
the deposit of any silver chloride from the solution, and the whole process was applied 
on a large scale, resulting in the purification of some 70 ounces of fine gold, of which 
Professor Roberts-Auster himself has said : “ I have not been able to prepare, or to 
obtain from any source, gold of greater purity, even in small quantities,” It seems, 
however, that the apparent standard of this gold was slightly reduced in rolling, 
the finished plate being counted as 999'95 fine in comparison with the same gold 
before rolling. A memorandum given me by Professor Roberts-Austen states that 
this trial plate gold is 999'98 fine as compared with the purest gold obtained by 
Btas for the Belo-ian Mint. 
This specimen of English trial plate gold is designated as D in the present papei-. 
All the samples of gold received from others—B, C, and D—were, before using 
them, carelully boile.d in nitric acid to remove any })ossible traces of silver or other- 
metal derived Irom tlie shears used in cutting- the plates. They were also previously 
well washed with ether, to remove any grease, and afterwards with pure water, and 
were finally heated to redness in the Sprengel vacuum. 
It may be remarked, in advance, that I have not been able to trace any probable 
connection between the history of the several samples of gold used and the values 
obtained for tlie atomic weight of the metal. Within the limits of accuracv attained. 
