1897 - 98 .] Dr A. Galt on Heat of Combination of Metals. 145 
For the mixture 826*0 (gramme-water) heat units Centigrade. 
„ alloy 791*6 „ „ „ 
Difference, 34*4 „ „ „ 
Thus the heat of combination of *7 gramme of copper with *3 
gramme of zinc is 34*4 (gramme-water) heat units Centigrade, or 
less than half the quantity (76*6) obtained as the heat of combina- 
tion of *52 gramme of copper with *48 gramme of zinc, the total 
weight of metal in each case being one gramme. 
Having obtained these interesting results with copper and zinc, 
Lord Kelvin advised that the experiments should be repeated with 
other metals. Copper and silver were fixed upon, and I am in- 
debted to Messrs Thomas Smith & Son, silversmiths, Queen Street, 
Glasgow, for the care and trouble they have taken in making, 
from pure silver and copper, two alloys, one of which contained 
216 parts by weight of silver to 63 parts by weight of copper, or 
77*4 per cent, of silver to 22*4 per cent, of copper, these being the 
relative combining proportions of the metals usually observed in 
chemical reactions in which they take part. The other alloy con- 
tained about equal weights of these metals. Separate specimens 
of silver and copper of the same quality as those in the alloys were 
also supplied. 
In the preliminary experiments on these four specimens, it was 
found that while all dissolved slowly in nitric acid, and therefore 
required to be in the state of fine filings so as to have solution 
completed in a short time, the silver was particularly slow in this 
respect ; and it was absolutely necessary to use it in the form of 
exceedingly fine filings. A special set of very fine files was used 
in the making of the filings, — a very slow and laborious process. 
It was also found that the silver dissolved more readily in fairly 
dilute nitric acid than in stronger acid. Kepeated tests showed 
that acid of density 1*265 at 15° C. was, perhaps, the best. 
A quantitative analysis of the first mentioned alloy showed that its 
composition was 76*73 per cent, silver and 23*27 per cent, copper. 
Comparative tests were made with one gramme (*7673 gramme of 
silver mixed with *2327 gramme of copper) of the mixed metals 
and one gramme of the alloy. Owing to the heat of solution being 
much less than that observed in the copper-zinc experiments only 
