DURING METALLIC SUBSTITUTIONS, 
93 
To remove all uncertainty as to the strength of the solutions, a considerable quan- 
tity of each salt was dissolved in water, and a portion of the solution carefully 
analysed by precipitating the oxide of copper. The solutions were all employed in a 
perfectly neutral state. 
In the second series of experiments, more dilute solutions were taken, and the 
inerement of temperature observed directly in the solution in which the precipitation 
occurred. The zinc, cooled to the same degree as the liquid, was introduced after 
the temperature of the former had been observed, and the whole rotated for a period 
of one minute and a half. After the final temperature was taken, a few drops of the 
liquid were quickly withdrawn for future examination, and the apparatus was again 
rotated for a period of one minute and a half. On again introducing the thermo- 
meter, the temperature of the liquid was always found to be a few hundredths of a 
degree higher than at the preceding observation, although the whole of the copper 
had been previously precipitated, and on repeating the same operation several times, 
nearly the same development of heat occurred on each occasion. This secondary 
evolution of heat arose from two distinct causes, the oxidation of the precipitated 
metal by the air contained in the upper part of the glass vessel, and the voltaic circle 
formed by the precipitated copper with the zinc in excess. The influence of the 
former circumstance was clearly proved by repeating the experiment with the vessel 
as nearly filled as possible with the solution, which considerably diminished the 
amount of the secondary development of heat. But without entering into a minute 
discussion of the efficient causes of this rise of temperature, it is sufficient for the 
present object to observe that the same causes must have been in operation, even in 
a more intense degree, during the greater part of the first period of agitation, and 
would render the increment then observed too high. The application of the required 
correction is very difficult, and the uncertainty on this point prevents absolute accu- 
racy being attained in the following numerical results. As the most probable 
estimate, I assumed the correction to be equal to the increment observed during the 
second period of rotation, without applying any correction to this increment for the 
cooling influence of the air. The amount of this correction was usually about 0°-l C, 
It should be carefully remembered that the pi’ecipitation was in every experiment 
proved to be complete at the end of the first agitation, by removing a few drops of 
the solution and afterwards carefully testing it. In the first series of experiments 
with the salts of copper, no correction was applied for this secondary development 
of heat, because it was impossible to ascertain its amount, which however was pro- 
bably less than in the experiments of the second series. 
First Series. — Sulphate of Copper and Zinc. 
The solution of sulphate of copper weighed 43’3 grms. and contained 1*100 grm. 
oxide of copper. The specific heat of the solution of sulphate of zinc which was 
formed, was found by direct experiment to be 0*935, and consequently its thermal 
