126 Scientific Proceedings, Royal Dublin Society. 
additional advantage that their use greatly facilitates the with- 
drawal of the charge of absorbent from the receivers, and also of 
the water employed subsequently to wash them 
out, in cases where that method was employed. 
Two different forms of these vessels were used 
at different times. The construction of one of 
them is shown in fig. 3. It was the one first 
used, and was intended not only for performing 
the titrations im vacuo with an aliquot part of 
the absorbent, but also for filtering the latter 
from barium carbonate. In using it, the space 
between H and C was first rendered vacuous by 
attachment of EH to an air-pump and the two 
taps closed. The upper end A was next attached 
to the rubber junction IJ. of the receiver (the 
latter being clamped in an inverted position to 
a retort stand), and the stopcock C slowly 
opened, when the absorbent was drawn out of 
the receiver through the asbestos plug at B, serv- 
ing as a filter, and thence into the measuring 
bulb D. When exactly 35 c.c. of clear fluid had 
thus been obtained, O was closed, the vessel in- 
verted, andthe burette attached to # by an india- 
rubber junction, both the end of # and the 
junction being first filled with the standard acid. 
In performing the operation both the vessel and 
burette were clamped to a retort-stand. 
burette was then filled up to the zero-mark with acid, and the 
titration performed by opening L. 
Fic. 3. 
No.1 Titrating vessel. 
The 
is probably now well known, but it does not seem to have been taken into account by 
most of our predecessors in the subject under consideration. 
We have found the following amounts in different samples of freshly-prepared 
distilled water used in the chemical laboratory of the Queen’s College, Belfast :— 
(1) contained 2°475 c.c. COz at N. T. P. per litre. 
(2) 99 3°090 35 
(3) 0 3°706 a 
(4) 59 5°703 
The magnitude of the error which might thus be introduced into the determination 
