534 
5IE. T. J. BAKER ON TEIE THERMO-CHEMISTRY 
No difference in the thermal Ijeliaviour of the specimens of copper obtained by the 
different processes was observed. 
The powdered copper was sifted through the finest linen obtainable, and was used 
at once. 
It was found necessary to moisten the powder before immersing it in the liquid in 
the calorimeter, because entangled air prevented actual contact between them, thus 
preventing or seriously delaying dissolution of the metal. 
For this purpose the copper was weighed on a paraffined tissue-paper tray, 
measuring about 3 centims. square, and a few drops of water were then thoroughly 
incorporated with it by means of a stirrer made of the same paper. 
Since not more than '3 gramme of water was used, its introduction produced no 
appreciable change in the heat capacity of the calorimeter. 
[The foregoing method of preparing the copper for dissolution was apjDlied to the 
zinc and all the alloys employed in this work, and was absolutely essential to 
success. ] 
Blank experiments were made to ascertain the temperature corrections to be 
applied for radiation, and for cooling consequent on the escape of chlorine from the 
free surface of the solution. 
Notwithstanding every care and precaution, it was only found possible to deal 
with the limited number of alloys w'hich could be easily powdered by blows in a 
mortar, and many of these wmre so difficult to dissolve that the corrections for radia¬ 
tion and cooling by loss of chlorine became too significant and masked any relations 
existing between the differences of observed and calculated heats of dissolution. 
It was, however, discovered that the observed heat of dissolution of an alloy was 
ahvays less than that of a mere mixture of the same composition as calculated from 
the ascertained heats of dissolution of copper and zinc, and that these differences 
were too great to be attributed to error of experiment. 
This evidence of the existence of heat of combination in copper-zinc alloys induced 
the author to seek a more suitable solvent. 
[Note on the Heat o f Solution of Chlorine in Water. 
It was easy to measure the heat of dissolution of copper powder in chlorine-water 
with considerable accuracy, as the process did not occupy more than one or one and 
a half minutes. 
Four experiments, none of wdfich differed more than ‘25 per cent, from the mean, 
gave 
Cu. CLAq = 58,315 
Thomsen gives Cu . CL . Aq = 62,710 
Therefore heat expended in withdrawing CL 
from aqueous solution by copper 
