If a heavy substance, like gold, be greatly extended by hammering and beating into thin leaves, and then dropped from the hand, the resistance of the air becomes very apparent; and a gold coin and a piece of gold-leaf would not reach the earth at the same time if allowed to fall from any given height. This fact is easily displayed by the assistance of a long glass cylindrical vessel placed on the air-pump, with suitable apparatus arranged with little stages to carry the different substances; upon two of them may be placed a feather and a gold coin, and on the third, another gold coin and a piece of gold-leaf. In arranging the experiment, great care ought to be taken that the little stages are all nicely cleaned, and free from any oil, grease, or other matter which might cause the feathers or the gold-leaf to cling to the stages when they are disengaged, by moving the brass stop round that works in the collar of leathers. Sometimes these leathers are oiled, [Page 15] and in that case, when the vacuum is made, the oil, by the pressure, is squeezed out, and, passing down, may reach the stages and spoil the experiment, by causing the feathers and gold-leaf to stick to the brass, producing great disappointment, as the illustration, usually called the "_guinea and feather glass experiment_" takes some time to prepare. The air-pump being in good order, the long glass is first greased on the lower welt or edge, and then placed firmly on the air-pump plate. The top edge, or welt, may now be greased, and the gold coins, feathers, and gold-leaf arranged in the drop-apparatus; this is carefully placed on the top of the glass, and firmly squeezed down. The author has always found a tallow candle, rolled in a sheet of paper (so as to leave about half the candle exposed), the best grease to smear the glass with for air-pump experiments; if the weather is cold, the candle may be placed for a few minutes before an ordinary fire to soften the tallow. Pomatum answers perfectly well when the surfaces of glass and brass are all nicely ground; but as air-pumps and glasses by use get scratched and rubbed, the tallow seems to fill up better all ordinary channels by which air may enter to spoil a vacuum. [Illustration: Fig. 17.] [Illustration: Fig. 18.]