Page:Popular Science Monthly Volume 38.djvu/663

This page has been proofread, but needs to be validated.
NON-CONDUCTORS OF HEAT.
645

finally the outer surface heats whatever is in contact with it. Such conveyance of heat from particle to particle, without sensible motion, is termed conduction.

Strictly speaking, according to modern theory, radiant heat is a peculiar kind of undulation communicated to a supposed exceedingly subtile, all-pervading ether; and conduction is an oscillation of the molecules of the conductor itself. But, though we no longer consider heat to be a substance, it is convenient to use the old terms figuratively in describing the phenomena, just as we still say the sun rises and sets, though it is the earth that moves.

When we sit near an open fire, we are warmed by radiation through the intervening air, while the air itself is heated by contact with the fire and passes up the chimney. So radiation and convection, or radiation and conduction, may go on at the same time, and when cooling takes place it is not always easy to tell how much of the effect is due to each of the causes respectively. Hence, substances that are put around hot bodies to retard the change of temperature are often called indiscriminately non-conductors, though in fact they may act partly by preventing convection or by intercepting radiation. Practically, indeed, it is of little consequence to decide exactly how the loss of heat is prevented, but, in the full study of retentive coverings, we must not altogether lose sight of the distinction between mere conduction and general transmission.

It is a matter of much interest as well as of economical importance to find out what substances are most suitable to keep hot bodies warm and cold bodies cool; and several methods have been devised for making either absolute or comparative trials. After due consideration of the plans used by different experimenters, the writer has adopted, for the many determinations which he has had occasion to make, an apparatus which may be arranged in three different ways: First, a short, cylindrical metallic vessel, with the flat ends vertical, is kept at a constant high temperature by a continual current of steam or hot water passing in at the bottom and out at the top. The non-conductor, of a regular thickness, say one inch, is applied to one of the flat faces of the heater. The other surface of the covering is in contact with a thin brass box, or calorimeter, filled with a known quantity of water to receive the transmitted heat. The number of degrees which the water is raised in an hour gives a definite measure of the amount of heat that the covering allows to pass through.

Secondly, in trying liquids or air for their conducting power it is desirable to get rid of convection by heating from above, so that the hottest part of the fluid shall be and remain at top.