beheld this first Change, then you must presently drop in about four or
five drops of Oyl of Vitriol, and continuing to shake the Glass pretty
strongly, that it may the Nimbler diffuse it self, the whole Colour, if you
have gone Skilfully to work, will immediately disappear, and all the Liquor
in the Glass will be Clear and Colourless as before, without so much as a
Sediment at the Bottom. But for the more gracefull Trial of this
Experiment, 'twill not be amiss to observe, First, That there should not be
taken too much of the Solution of Sublimate, nor too much of the Oyl of
Tartar drop'd in, to avoid the necessity of putting in so much Oyl of
Vitriol as may make an Ebullition, and perhaps run over the Glass.
Secondly, That 'tis convenient to keep the Glass always a little shaking,
both for the better mixing of the Liquors, and to keep the Yellow Substance
from Subsiding, which else it would in a short time do, though when 'tis
subsided it will retain its Colour, and also be capable of being depriv'd
of it by the Oyl newly mention'd. Thirdly, That if any Yellow matter stick
at the sides of the Glass, 'tis but inclining the Glass, till the clarify'd
Liquor can wash alongst it, and the Liquor will presently imbibe it, and
deprive it of its Colour.

Many have somewhat wondred, how I came to light upon this Experiment, but
the Notions or Conjectures I have about the differing Natures of the
Several Tribes of Salts, having led me to devise the Experiment, it will
not be difficult for me to give you the Chymical Reason, if I may so speak,
of the _Phænomenon_. Having then observ'd, that _Mercury_ being dissolv'd
in Some _Menstruums_, would yield a dark Yellow Precipitate, and supposing
that, as to this, common Water, and the Salts that stick to the _Mercury_
would be equivalent to those Acid _Menstruums_, which work upon the
_Quick-silver_, upon the account of their Saline particles, I substituted a
Solution of Sublimate in fair Water, instead of a Solution of _Mercury_ in
_Aqua-fortis_, or Spirit of _Nitre_, that simple Solution being both