190 
RECIPROCATING SPRINGS. 
cation underground, with one of these spiracles of the globe. 
Be this as it may, the notion is too absurd to be mentioned in 
the present improved state of Natural Philosophy, in any other 
light than as a curious document of the puerile conceits with 
which the philosophers of ancient times amused their hearts. 
In the foregoing attempt to complete the theory, I have had 
recourse to a well known phenomenon j water is beaten into 
foam by being agitated ; which was the case by Mr. Swainston’s 
vessel, because a strong current fell into it from the pump. 
There is, however, one objection still remaining, which de- 
serves to be considered : the levity of foam, compared with 
the superior weight of water, may lead some persons to suspect, 
that this light substance will not mix with water, but will float 
on the surface of the reservoir, in which it is formed. Sup- 
posing this suspicion to be well-founded for the sake of argu- 
ment, we must allow the foregoing theory of reciprocating 
vessels to be defective in a very essential point $ because if 
foam cannot sink, the air that proceeds from it, cannoi find 
its way into the tubes or siphons, which convey the water 
from such vessels. Being unwilling to leave this objection un- 
Experiments answered, I resolved to put the truth of this principle to the 
to shew that test of direct experiment $ which was done in the following, 
foam, &c. simple manner: A small bell glass, being first filled with 
will descend, water, was inverted in six quarts of the same fluid, contained 
in a small tub. Things being thus prepaied, the contents of the 
open vessel were agitated briskly } and the air which entered 
the water, found its way into the inverted glass, the upper 
part of which it occupied. The water of the tub was agitated 
by the motion of a whisk, or a bundle of slender twigs ; it 
was some times taken up in a pitcher, and returned into the 
vessel quickly, from the height of a foot or more $ both me- 
thods proved successful, but the former appeared to introduce 
air into the glass with more expedition than the latter did $ 
the difference here mentioned, may however depend entirely 
upon management and accidental circumstances, The experi- 
ment which I have now related, shews the foregoing objection 
to be of no moment $ consequently the present theory of 
irregular leciprocation may be pronounced to stand upon a safe 
foundation,, and unexceptionable principles. 
The 
