a body may be cold at one time and hot at another it is simply a state or condition of a body. Temperature is sometimes defined as being the 'state of a body with respect to sensible heat;' but we had better from the outset avoid the use of any terms which would seem to indicate that there are different kinds of heat.

When two water cisterns at different levels are connected by a pipe, the difference of level produces a pressure which causes the water to flow from the higher cistern to the lower one. So when a hot body is placed in contact with a colder body heat flows from the hot body to the colder one. Here the difference in temperature corresponds to the difference in level of the water cisterns, and the heat flows from the body at a higher temperature to that at a lower temperature, just as water tends to flow from a higher to a lower level. We are thus led to the following definition :—

Temperature is a condition of bodies that determines which of two bodies when placed in contact will part with heat to the other.

The body which parts with the heat is said to have the higher temperature; if there is no transference of heat the two bodies are said to be at the same temperature.

4. Effects of Heat-Expansion. The principal effects produced by applying heat to a body are (1) change of size, (2) change of temperature, and (3) change of state. We shall

begin by observing how heat causes solids, liquids, and gases to expand, and then go on to see how this expansion can be used as a means of measuring temperature.

EXPT. 2.-Expansion of Solids. -This can be observed by taking a brass ball and a ring (Fig. 1) which fits it loosely, so that the ball can just pass through the ring when both are cold. After the ball has been heated over a spirit-lamp or in the flame of a Bunsen burner it will no longer pass through the ring but will rest upon it, thus showing that it has expanded. As the

Fig. 1.

ball cools it contracts to its original size, and the ring, being warmed by contact, also expands a little so that after a few minutes the ball drops through.

EXPT. 3.-Expansion of Liquids.—Fit a small1 flask with a Fill the flask with water

cork and long narrow glass tube.

coloured by a little litmus tincture or solution of indigo; push the cork in so as to force the coloured water half-way up the tube and see that no air is enclosed below the cork. Make a movable index out of a piece of card or stiff paper by cutting two slits near the ends and slipping it over the tube, as in Fig. 2. On immersing the flask in lukewarm water its contents will expand and the column of coloured water in the tube will gradually ascend.

Liquids expand with heat more rapidly than solids, and air and other gases more rapidly than either.

Fig. 2 [1/10].

Fig. 3 [1/10].

EXPT. 4. Expansion of Gases. The expansion of air can be observed in a simple way with a flask (empty) fitted as above. Hold the flask upside down with the open end of the tube just dipping under water; the heat of the hand applied to the flask is enough to make the air expand, as can be seen by the bubbles which escape through the water. Το fit up the instrument for permanent use take a bottle with a fairly broad bottom, pour a little coloured water into it, and by means of a good thick cork fix the tube in the neck of the bottle with the end dipping under the water (Fig. 3). In the side of the cork cut a slit so as to allow the air to pass

1 In fitting up apparatus for an experiment the exact size will depend largely upon whether it is to be used for private observation or class demonstration. In order to avoid unnecessary details in the text, many of the figures are drawn to scale. The scale is either marked on the engraving or beneath it: thus Fig. 2 is one-tenth [1/10] of the natural size. The sizes thus indicated are generally those suitable for class work.

1

freely in and out of the lower bottle.

Place your hand on the flask, or heat it gently by holding a lighted match near it, until about a dozen bubbles of air have been expelled; the air will contract on cooling, and the column of liquid will ascend about half-way up the tube, which should be provided with a movable index as before. The instrument now forms a fairly sensitive thermoscope; by watching the movements of the column of water you can find out when heat is imparted to, or taken away from the air in the flask. Thus you can tell whether a given sample of water is hotter or colder than the air; for you can pour a few drops of the water upon the inverted bottom of the flask, and if the temperature of the water is higher than that of the air, expansion will occur and the column will descend and vice versa. Or again, by moving the thermoscope from one room to another you can compare the temperatures of a series of rooms.

CHAPTER II

THERMOMETERS

5. What we have called an air-thermoscope (Art. 4) is sometimes called an air-thermometer, but it does not deserve to be dignified with this name, for a thermometer (as its name implies) ought not only to indicate changes of temperature but also to measure them. Our air-thermoscope does not always give the same indication when exposed to the same temperature; its indication, i.e. the height at which the column of water stands, depends upon the pressure of the atmosphere, which varies from time to time. For this and other reasons it would not be easy to use it (in the form described) for measuring temperature. Liquids are not affected by changes of atmospheric pressure as gases are. A rough kind of thermometer can be made by blowing a bulb at one end of a narrow glass tube and filling it with coloured water or some other liquid the liquid should then be warmed and the open end of the tube sealed. But how is the thermometer to be graduated? You might affix a paper or cardboard scale to it and divide this into a convenient number of equal parts, say twenty or a hundred; but you would find if you made another thermometer in the same way that the two would not give the same indications at the same temperature. Your graduation being an arbitrary one, the indications of one instrument cannot be compared with those of another. A more scientific way would

be to begin by choosing two standard temperatures as 'fixed points' to which all thermometers could be referred. The fixed points always adopted are the temperature at which ice melts and that at which water boils. If you immerse a thermometer in melting ice, or in a mixture of ice and water, you

will find that it always registers the same temperature; this is called the melting-point of ice, and is taken as the 'lower fixed point' of the thermometer. The temperature at which water boils varies slightly with the atmospheric pressure, but for a given pressure it is constant, so that if we fix upon a certain pressure as the standard atmospheric pressure, then the boiling-point of water will give us a second or 'higher fixed point' for our thermometer. These two fixed points are very convenient standards of reference, for both water and ice can always be obtained in a state of purity.

Thermometers are never filled with water: a water-thermometer could only be used through a very limited range of temperature; it would be impossible to mark the fixed points, and even between these the expansion of water is very irregular. Mercury (quicksilver) can be used through a much wider range of temperature; it expands more regularly than other liquids, and for these and other reasons it is almost always preferred as a thermometric fluid.

6. Construction of a Mercury Thermometer.—A bulb is blown at one end of a capillary glass tube, i.e. a tube with a very fine bore. On account of the contained air and the fineness of the bore the bulb cannot be filled by pouring mercury down, but if you gently warm it so as to expel some of the air,

Fig. 4. THERMOMETERS.

and then dip the open end of the tube under mercury, a little of this will be sucked up as the air cools and contracts. The process of heating and alternate cooling must be repeated several times until the bulb and tube are filled with mercury; finally the whole is carefully heated, so as to drive off any contained air and moisture. While the mercury is still warm the end of the tube is closed by cautiously directing a blow-pipe