Metals in the Formation of Alloys. 407 



watch the scouring effect in the globe due to the chemical 

 action ; the filings were almost instantly expelled from it 

 through the apertures by reason of the rapid evolution of 

 gas. But in the cases of a few of the alloys and mixtures 

 containing large percentages of copper this effect was only 

 partial, and a considerable proportion of the powder remained 

 in the bulb with the result tbat solution was much delayed. 

 In these instances the bulb was replaced by a tiny glass 

 basket suspended from a spiral hook formed at the bottom of 

 the glass tube, so that when the basket containing the filings 

 was driven to the bottom of the acid the basket was over- 

 turned and the filings scattered over the bottom of the flask. 



When solution was started the flask was quickly inserted 

 into a large cylindrical metal can through an aperture in the 

 cover. The can had a water-jacket round the sides and 

 bottom, and the contained air could be kept at a practically 

 constant temperature, which was a fraction of a degree 

 centigrade below the maximum which would be attained by 

 the solution, by heating the water in the jacket by means of a 

 bunsen-burner placed below the can. The temperature of the 

 contained air was read by a sensitive thermometer suspended 

 in the can, the bulb of the thermometer and the bottom of the 

 flask being close together near the centre of the can. 



Solution was generally effected in 55 to 60 seconds, and 

 while it was going on the flask was shaken so as to give a 

 rotatory motion to the acid. Complete solution and maximum 

 temperature were nearly coincident in time. 



Each experiment was repeated from three to six times until 

 consistent results for the heat of solution in each case were 

 obtained, and the mean of these was taken. In each case of 

 solution the nitrous products remained in the liquid, the flask 

 in which solution took place being kept gas-tight by the 

 cork. 



The total weight of the whole apparatus — flask, cork, ther- 

 mometer, and sealed glass tube with attached globe or basket 

 — was 42 grammes, and the water equivalent was found to be 

 5' 7 grammes. The specific heat of the nitric acid used was 

 repeatedly determined, and the mean of several values was '658. 

 The specific heat of the metal used is negligible and was not 

 taken into account. 



with the violent action of nitric acid on zinc and copper, and the abun- 

 dant evolution of gas which accompanies it. But the facility with which 

 the gases may be condensed by the acid solution is probably not so gene- 

 rally known, and when the experiment is made for the first time it cannot 

 fail to excite surprise," 



