348 J. P. Cooke, Jr., on determining amount of 
by the apparatus represented in the accompanying figure which, 
although it may appear complicated at first sight, is really a 
simple combination of parts, which are familiar in every labora 
tory. 
To the sides of a common copper water bath are attached 
three tubes as is shown in the figure. The tube on the left con- 
nects with a Mariotte’s flask by which the water is maintained 
at a constant level in the bath. The upper tube on the right 
. connects with a carbonic acid gas generator, while the lower 
tube (connected with the sink by a rubber hose), serves simpl 
as an overflow for the water. On the cover of the water-ba 
close to the rim is a depressed circular groove, which receives 
the lip of 'an inverted glass tunnel. When the apparatus is in 
ni | 
SI | i 
i ie , 
‘ = (i " i / 
i 
: Hit h, ! Af 
ir 
tu) Mf, 
this groove is kept full of water by the spray from the boil- 
ge 
ing liq 
assay will be readily understood. A small amount of the finely 
