57 6 Mr. Whitehurst’s Experiments 
My experiments are as follows: i ft. One penny* 
weight of gold, made red-hot, became apparently lighter; 
but, when reftored again to the temperature of the atmo- 
fphere, its former weight was perfectly reftored* 
2 d, One pennyweight of iron, heated as above, was 
alfo apparently lighter; but, when it became cool again, 
its weight was viftbly augmented. 
It is now feveral years fince I made thefe experi- 
ments; but I well remember to have repeated them 
feveral times, and that the refults were always the fame. 
It may be neceflary to remark, that the beam ufed 
in thefe experiments was lenfibly affected by the y^th 
part of a grain; and likewife, that each of the metals 
was heated upon charcoal, by means of a candle and a 
blow-pipe, and both were brought nearly to a ft ate of 
-fufton. 
It feems neediefs to obferve, that the apparent levity 
-of the gold and of the iron, when hot, was owing to the 
nfcent of the rarified air above the fcale, and to the ten- 
•dency of that underneath to reftore the equilibrium of 
its preflure. The increafe of weight in the iron might 
probably arife from its having, in fome degree, acquired 
•the property of fteel, by means of the flame and char- 
coal. 
I am at a lofs to account for the fallacv which feems 
to have attended M. buffon’s experiment; but it feems 
probable, that the heat of the mats of iron employed by 
him, had a greater effect on that arm of the beam from 
which 
