The balls attached to the copper bar fall off, by the melting of the wax, much sooner than those hanging from the iron

one, thus proving conclusively that copper is a better conductor of heat than iron, although their capacities for heat are about the

same.

Thermal conductivity must be measured (other things being equal) by the quantity of heat which passes; therefore the rate at which conduction (whether internal or external) goes on, is proportional to the cross area of the section, or the surface through which it takes place. It may be expressed numerically in so many units of heat per square foot per minute, or per hour. For example, engineers speak of the evaporating power of a boiler, as so many pounds of water raised into steam at a certain pressure per square foot of grate surface per hour, or plus per square foot of the additional heating surfaces, although in reality it depends on many things besides the mere conduction of the plates.

To compare plates of different materials, we must take them all of the same thickness and superficial area, and subject them all on the one side to a certain temperature, and on the other side to the same number of degrees more or less.

It is important that the engineer should appreciate the relative conducting capabilities of the different metals that he has to deal with. For instance, the fire-box of a locomotive is made of copper in preference to iron, partly on account of its greater conductivity and partly on account of its withstanding the destructive action of the fire better. Again, the outsides of boilers and of cylinders are carefully lagged with some badly conducting substances, such as hair or felt, and wood, so that as little heat as possible may escape therefrom. The following table gives roughly the relative conductivities of a few of the more common metals :

-

Convection. When the application of heat to a fluid causes it to expand or to contract, it is thereby rendered rarer or denser than the neighbouring parts of the fluid; and if the fluid is at the same time acted on by gravity, it tends

to form an upward or downward current of the heated fluid; this is accompanied with a current from the more remote parts of the fluid in the opposite direction. This action is rendered very apparent by the following simple experiment:

Take a flask partially filled with water, mix a few grains of bran with it, and apply a lighted spirit-lamp to the bottom of the flask. In a few minutes the water will be seen to circulate in the direction shown by the arrows in the figure. The water nearest the flame is rendered lighter, and, therefore, rises upwards, while the denser water falls under the action of gravity, to be in turn heated and raised.

The actual transfer of heat throughout the water takes place by conduction, but the diffusion is much assisted by the motion of the fluid, or convection currents, as they are termed.

The following experiment is also very instructive :—Take a test tube filled with water (left hand, Fig. A), and apply a spirit

lamp near the surface of the water. You may hold it there for ten minutes or more, and the water at the bottom of the tube is scarcely perceptibly warmer than at first. Now apply the lamp to the bottom of the tube (right hand, Fig. B); in a few minutes the water begins to boil. Why this difference? The convection

currents set up have assisted the naturally bad conducting power of the water by bringing, in turn, every portion of it into close proximity with the source of heat.

It is for the reasons just mentioned, that the fire-place in a boiler is placed near the bottom instead of near the surface of the water, and it is of great moment not only to give a free and easy path for convection currents in boilers, but to stimulate them by such appliances as hydro-kineters. The better the circulation of the water in a boiler, the more rapidly will it be heated and the steam generated. In many boilers (such as those used on board steamers) the internal construction is so mixed up with tubes and stays, that the water has great difficulty in passing from out-ofthe-way corners to the more highly heated parts over the flues; and, if circulation is not assisted, the convection currents "short circuit," as it were (to use an electrical term), and thus leave the more remote portions in comparative chill. For a similar purpose, large boiler flues are provided with "baffling plates," to compel the hot gases to take a circuitous course, in order that eddies may be formed, and for the further object of promoting a better mixture of air with the inflammable gases.

The art of promoting a good draught in a furnace, or of properly ventilating a building or a ship, depends upon inducing and guiding the convection currents in the proper direction. This subject is, therefore, of considerable importance to engineers.

LECTURE VI. QUESTIONS.

1. Name the different ways by which heat is transferred from one part of a body to another part of the same body, and also from one body to another not in contact with it.

2. Explain, and illustrate by an example not mentioned in this lecture, how radiation takes place. Why are the covers of steam-engine cylinders polished or electro-plated?

3. Explain, and illustrate by an example of your own, how conduction of heat takes place in bodies. Name four of the best conductors in their order of conductivity. Name also a few of the worst conductors.

4. What is meant by "convection currents"? How does convection assist the engineer in raising steam in a boiler? Illustrate your answers.

5. Describe an experiment by which you would show that water is an extremely bad conductor of heat. For what reason should heat be applied from below when it is required to heat a large mass of water rapidly?

E

LECTURE VII.

CONTENTS.-Nature of Heat-Heat is not a Substance-Rumford, Davy, and Joule's Experiments-Conversion of Work into Heat-Joule's Mechanical Equivalent of Heat-First Law of Thermo-dynamics.

UNTIL the end of last century, two rival theories had been entertained regarding the nature of heat-one, that heat consisted of a subtle elastic fluid, termed caloric, penetrating through the pores or interstices of matter, like water in a sponge; the other, that it was an internal commotion among the particles or molecules of matter.

The former of these theories, or hypotheses, that heat is matter, called the "materialistic doctrine of heat," taught by Professor Black of Glasgow University and others, was most conclusively overthrown by the celebrated experiments of Count Rumford and Davy. It is very remarkable, however, that fifty years elapsed before scientific men generally became converted to the conclusions to be drawn from them. It was not until Joule, during the period extending from 1840 to 1849, had supplied several fresh proofs that heat is not a material substance, but one form of energy, which may be applied to or taken from bodies in various ways, and that the amount of energy, in whatever form applied or removed, may be estimated in mechanical units of work or footpounds, that what is now known as the Kinetic theory of heat, became generally accepted, and the science of thermo-dynamics placed on a firm basis.

Count Rumford's experiments on the production of heat by friction, were carried out in the following manner, and communicated to the Royal Society in 1798:

In casting guns it was usual to leave a projecting cylindrical "head" of metal at the muzzle, so as to insure sound metal in the gun. The guns were cast in a vertical position with the muzzle end upwards, very much in the same way as large water or gas pipes are now made, the effect of adding the "head" to the casting being to add pressure to the fluid metal in the lower parts, thus expelling air and gases towards the surface, and into the "head," which was cut off before boring out the gun.

Rumford obtained a casting for a six-pounder brass gun from