Lerrs & Buake—The Carbonic Anhydride of the Atmosphere. 145 
It was suggested to us that it might be possible to demonstrate 
the action of weak baryta water on glass objectively by separating 
and collecting the products. Acting on this suggestion we have 
found it easy to separate and even to determine one such product, 
namely, silica. The following plan was adopted :— 
Four Winchester quarts, similar to those we used in our pre- 
vious experiments, were carefully cleaned, rinsed with distilled 
water and drained. Hach was then charged with 50 ¢.c. of the 
baryta solution, stoppered, and placed on its side for a certain 
time, with occasional rolling and shaking. The contents of all 
four were then transferred to a platinum dish, and each bottle 
separately rinsed out with 60 c.c. of the dilute acid in three 
separate portions, these washings being added to the contents of 
the platinum dish. The latter was then heated on a water bath 
until :ts contents were quitedry; 1¢.c. of strong hydrochloric acid 
next added, and finally the insoluble residue either washed into a 
filter and weighed (after complete washing and ignition) or rinsed 
into tubes which were sealed up. 
The same bottles were used in all our experiments, conducted 
in this way. The blank experiments were made by evaporating 
200 ec. of the baryta water with 240 c.c. of the acid in the 
same platinum dish, the subsequent treatment being as above 
described. 
Appearance of residue in tube, or weight 
| 
| 
No. of Time of Exposure. 
Exper.ment. of silica. 
i 42 hours. Considerable flocculent deposit. 
2 Blank experiment. Minute deposit. 
| 3 72 hours. 0:0066 grm. 
| 4 Blank experiment. ; 0:0006_ =, 
5 72 hours. Heavy deposit. 
A similar experiment made with a glass flask also yielded 
silica, so that the action is not confined to the glass of which 
Winchester quarts are made. 
The action of alkalies upon glass has been repeatedly investi- 
gated, and, in connection with this matter, we may refer to the 
