66 
Proceedings of the Royal Society of Edinburgh. [Sess. 
carefully prepared, and was kept in a thermostat at 30°±002. From it, 
20 c.c. samples were withdrawn by means of warm pipettes provided each 
with a glass-wool filter-cap, and were quickly transferred to a series of 
“ Pasteur ” tubes in a rack in the bath. The rack was then put into a bath at 
40° to 50°, and to each tube 05 c.c. of acid was added, a different concentration 
of acid being used for each tube. One sample was not mixed with acid, and 
served as a control. The ground caps of the tubes, which fitted perfectly 
when lubricated, were replaced, and the whole rack was completely 
immersed in the thermostat, where after a while it was rotated for some 
hours at a time, if necessary during two days. If crystals were formed, 
they were very easily seen on inverting the tube under the water of the 
bath, when they rolled down to the narrowed end of the cap ; it was 
possible in this way to detect quite minute deposits. 
This procedure alone might lead to erroneous conclusions owing to 
supersaturation ; therefore, after the above examination was completed, the 
narrowed ends of the tubes, containing some of the solution, were allowed 
to project above the surface of the thermostat- water, and were cooled till 
local crystallisation set in ; the tubes were then submerged again and 
rotated for a considerable time, and subsequently examined for undissolved 
crystals. 
By two or more repetitions of these experiments, the value of A, the 
critical concentration of added acid, could be determined to within about 
0*2 gram-equivalents per litre. The three test cases, shown in Table I., are 
those of sodium chloride, barium chloride, and barium nitrate ; and it is 
seen that the agreement is excellent in the first two cases. With the 
nitrate, the amounts of precipitate were small and therefore difficult to 
observe, and the discrepancy between found and calculated result may be 
due to this cause. It appears, however, that for the chlorides interpolation 
is valid. 
Table I. 
Salt. 
NaCl. 
iBaCl 2 . 2H 2 0. 
iBa(N0 3 ) 2 . 
NH 4 C1. 
KC1. 
(B 
5-40 
349 
0-85 
Previously found < 
u 
0-95 
0-85 
1-06 
-A-ealC.) = &yB 
5T 
3*0 
0-9 
A found . max. 
5*2 
3-0 
1-4 
6-2 
5*4 
min. 
5*0 
2*7 
12 
6-0 
5-0 
