48 JPresidenfs Address. [Feb. 



some 3 or 4 diameters from sun-sefc that the Solar Spectrum exhibits the 

 atmospheric lines or absorption-bands, though there are some exceptions to 

 this rule ; the conditions under which the Spectrum is thus seen are indica- 

 ted by " Sun-set" in contradistinction to observations made "Sun-high" 

 between 10 a. m. and 2 p. M. the latter Spectrum is already established in 

 KirchofF's well known Maps, on which the Sun-set Spectrum is now being 

 drawn. Proceeding in this manner, Mr. Hennessey has mapped the atmos- 

 pheric lines from the extreme red to F, and the results may be expected to 

 appear in due course in the papers of the Royal Society of London. 



In my own Department, Minting, there is some reason to hope that 

 Spectroscopic analysis will ere long prove a valuable aid in the operations of 

 refining and assaying, Mr. Chandler Roberts, Chemist of the London Mint, 

 in remarking on the use of the Spectroscope in determining the amount of 

 Gold present in the Gold-Copper Alloy, i. e, the Sovereign, writes : — 



" In former reports I have shown that the existing method known as 

 the gold parting assay, affords results which are trustworthy to the 

 ToVoo P^^^ ^^ ^^® original weight of the assay piece, and without being un- 

 mindful of the really wonderful accuracy of the method now used, I am 

 satisfied that in the examination of a series of gold-copper alloys by means 

 of the spectroscope differences of composition more minute than the 

 Tolo-o ^^^ ^® readily distinguished- The advantage of the proposed 

 method consists in the fact that the value of a single assay piece, can be de- 

 termined in a few minutes while an assay b}/ the ordinary method can 

 hardly be completed in less than two hours." 



Now, — This comparison of the time occupied, does not give any adequate 

 idea of the practical benefit that would be derived from the perfecting of this 

 system of assay, if it be found to be sufficiently trustworthy ; for though the 

 period necessary for the assay will not exceed a few minutes, two hours, 

 which is given as the time necessary for any single assay under favourable 

 circumstances, can never in practice represent the time actually occupied in 

 large operations and which we have never been able to reduce below two 

 days for Silver and one for Gold. 



In a paper lately read before the Royal Society, some further informa- 

 tion was given regarding the progress of the experiments that are being 

 carried on. 



The Qualitative Spectrum analysis is known to depend on the position 

 of the lines. The Quantitative Spectrum analysis, on the other hand, depends 

 on the length, brightness and thickness of these lines. 



The position, length and, to a certain extent, thickness are definite 

 exhibitions and are therefore susceptible of accurate measurement, but 

 brightness is not so definite and can only be estimated. For the more accu- 

 rate reading and recording Spectroscopic analysis photography is now used. 



