hot body with a damp cloth. Count Rumford, who long ago did much valuable work in the experimental study of heat, concluded that fluids have no conducting power at all, but transmit heat solely by convection; and, accordingly, water is still sometimes spoken of as an exceedingly poor conductor. But later investigators have disproved the correctness of that idea. Our own trials show that, when convection is obviated, water transmits in a given time six times as much heat as hair-felt of the same thickness, and nearly eight times as much as still air. Others have found that bisulphide of carbon and ether transmit heat even better than water; but most liquid substances are slower conductors. Thus it takes more than twice as long for a given amount of heat to pass through cotton-seed oil or lard oil as through water.
As to the gases, some physicists seem to have proved that heat passes through air more readily than through a vacuum, while hydrogen has six times as much transmissive power, and carbonic acid half as much as air; but none of them used apparatus that would give absolutely certain results.
To show more clearly the retentive power of various substances, we subjoin the following table, in which the first column of figures shows the net percentage of solid matter in a given space. The second column of figures gives the number of English units of heat transmitted in one hour through one square foot of the covering one inch thick, the average difference of temperature between the heater and the water in the calorimeter being 100° Fahr. By the English unit of heat is meant as much heat as will raise the temperature of one pound of water 1° Fahr. Of course, the smaller the number in the last column the better is the substance for keeping a body warm or cold.
In some of the experiments the source of heat was steam at 310° Fahr. In the others a stream of water at about 176° Fahr. was kept running through the heater.
It is plain that in choosing non-conductors for practical service we should take into account something more than their heat-retaining power. They should be of materials that are abundant and cheap; clean and inodorous; light and easy of application; not liable to become compacted by jarring, or to change by long keeping; not attractive to insects or mice; not likely to scorch, char, or ignite at the long-continued highest temperature to which they may be exposed; not liable to spontaneous combustion when partly soaked with oil; not prone to attract moisture from the air; and not capable of exerting any chemical action on surfaces with which they are placed in contact. There is no one thing which combines all the desirable good qualities, but there is a considerable range of substances which fulfill most of the require-
