[ 221 ] 
The liquor, decanted after boiling, neither changed' 
colour with galls, nor fliewed any precipitation with 
lime-water.. 
Experiment III,. 
A common quart-bottle was half filled with dif- 
tilled water, to which were added an hundred grains, 
of fteel-filings. To thefe was introduced, by means 
of the bent tube, as much fixed air, obtained from a 
folution of alcaline fait in the vitriolic acid, as was 
judged fufficient to fill the bottle. The whole be- 
ing then fhaken, with my hand over its mouth, the' 
bottle fluck like a cupping-glafs. About the fame 
quantity of air being again added, the bottle, after 
fhaking, had lefs adhefion than before. On repeat- 
ing this experiment,, a third time, with frefli air, the: 
adhefion was fcarcely perceptible. And after the' 
fourth trial, a fmall portion of air was obferved to if- 
fue from the bottle.. The water now gave a deep> 
colour with tindure of galls. 
This experiment was repeated with fixed air 
from different combinations. As alfo by palling this, 
air through a veflel of pearl-afh, to arreft any acid' 
which might efcape from the effervefcing mixtures.. 
But the folutions of iron, in all the trials, appeared to. 
be exadly fimilar,, except fome trifling difference, im 
tafle and fmell. 
Experiment IV., 
A bottle, with the like quantity of fleel-filings and* 
didiiled water as in Experiment, the Firff, remained. 
