228 Mr Chree, On the effect of an electric [May 21, 
have escaped detection. This affords strong independent evidence 
in favour of the view stated above. It is noteworthy that in this 
last experiment the quantity of salt that dissolved from first to 
last was less than two-fifths of that required to saturate the 
solution at its final temperature, yet 14 hours elapsed between 
taking the first and last samples and during the last 40 minutes 
the temperature was nearly constant. 
In the analysis of the calcium chloride, as already explained, 
undistilled water was sometimes used and in treating the samples 
taken on separate occasions different strengths of AgNO, solution 
were employed. Also on two occasions the current was allowed 
to produce its full heating effect. Thus it is only between samples 
taken on the same occasion that a strict comparison holds. In 
the following table are shown the average of the results obtained 
by analysis. 
Date of Before current Timecurrent During current Interval After current 
experl- 9 —<———*~——-{\, run,in ~————“~——, elapsed,in  —— —s> 
ment Temp. Vol.AgNO; minutes Temp. Vol.AgNO; minutes Temp. Vol.AgNO; 
Dec OF 1625 som ho nb eo omemaaT * 17-175 
SD Der ee 39 152° 1880 20 =152° 18:86 
joo L4G diske Luo s 60 244° 18:95 20° 21 S02 
... 16, 14° 24°30 42 204° 24r30 20 19° 24°45 
The solution was made some 4 or 5 days before the first 
observation, and yet by direct analysis it was found that between 
Dec. 9th and 12th the quantity of salt dissolved had increased at 
least $ per cent. Samples taken on the 14th showed almost 
certainly a slight increase in the strength of the solution since the 
12th. Samples however taken two mouths subsequently showed 
no increase in strength since the 16th. 
On two occasions, about the date when the samples last men- 
tioned were taken, the solution was heated up and three samples 
taken precisely as in the case of the KCl solution, distilled water 
being employed. On each occasion the temperature was raised 
8z°, the heating lasting in the first experiment for 55 minutes 
and in the second for 45. After taking the second samples cooling 
was allowed, on the first occasion for 45 minutes with a fall of 13° 
in temperature, and on the second occasion for 35 minutes with a 
fall of 4°. The AgNO, solutions employed on the two occasions 
were slightly different, but 17 c.c. was about the average quantity 
required for titration. The results obtained agreed very well, and 
indicated on an average that the second sample required ‘12 c.c. 
more AgNO, than the first, and the third -2c.c. more than the 
second. 
It will be seen that the heating effect on these occasions was 
about the mean of the heating effects in the two last electrical 
* Not observed. + Probably about 30, certainly under 60. 
