220 



REPORT — 1873. 



These small variations assume serious proportions -when they affect the 

 value of large quantities of bullion ; for instance, the value of gold coined at 

 the Mint during the past year -was £15,200,000, and a persistent error in 

 the assay reports of only ^^^ part would have been attended -with a gain or 

 loss to the Department of no less than ^£1500. 



The Committee hojie that their labours -will ultimately result in a clear 

 definition of the conditions under -which errors arise. 



The method of gold-assaying, as practised in the Mint, is given in the 

 Appendix*; and this method, kno-wn as the parting assay, has been de- 

 liberately adopted by all assayers, -with slight variations of manipulation, 

 ■which have not as j'et been minutely examined, as the Committee considered 

 that when -widely divergent results are obtained the gold employed by one or 

 other of the assayers as " check pieces " is impure, and that either the amount 

 of impurity has not been ascertained -with accuracy, or it altogether escapes 

 detection. It foUo-ws, therefore, that the -weight of the check " cornets," -when 

 compared with the initial weight of the portion of metal operated upon, ap- 

 pears to indicate the presence of an amount of gold which is in excess of 

 the true amount of precious metal present in the alloy. 



The Committee obtained specimens of gold from different sources*, and 

 tested them side by side with gold prepared, in accordance with the direc- 

 tions of the Lords Commissioners of Her Majesty's Treasury, by the Chemist 

 of the Mint for use as trial-plate for testing the coinage. 



Great care was taken in the preparation of this gold, 80 ounces of which 

 were precipitated from 100 gallons of chloride of gold ; and as experiments 

 have already shown that it is very pure, the Committee propose to adopt it 

 as the basis for a new series of comparisons, and, further, to invite assayers 

 to submit samples of the gold used by them in order that they may be 

 tested side by side with this standard plate. 



APPENDIX. 

 No. I. 



Experiments to determine the effect produced on the weight of assay pieces 

 of fine gold (each weighing 1000) by submitting them to the process of 

 assay. 



Appendix III. 



