position of the slide will indicate the amount of contraction which the clay has undergone during the firing. If the goods 'slip' in the kiln, this method is rendered inaccurate.

Optical pyrometers ('pyroscopes') in various forms are occasionally used for determining the temperature of a kiln, but they are scarcely accurate enough for this purpose. They are based for the most part on the comparison of the colour of the kiln with that of a standard flame, or electric lamp filament, the comparison being facilitated by inserting coloured glasses, or prisms, between the brighter light and the eye. The accuracy

of which these instruments are capable is not, apparently, much greater than that of the practised eye of a skilled fireman, though they are of great assistance to a beginner. A modification of this plan consists in measuring the photometric value of the light from the kiln by allowing it to fall on a small disc of paper supported vertically in a large horizontal iron tube (the paper having a grease-spot at its centre) and allowing the light of a standard candle or lamp to illuminate the other side of this paper. The standard light is then moved nearer to or further from the paper until the grease-spot disappears. When this is the case, both sides of the paper are equally illuminated and the distance of the standard light from the paper then becomes an indirect measure of the temperature of the kiln. Except with very skilled observers this method is not sufficiently accurate for best work.

Various attempts have been made to dilute the hot air of the kiln with cold air, and to measure the temperature of the mixture with an ordinary thermometer, but without much success, though a similar instrument in which a current of water (instead of air) was allowed to flow through a coil of pipe inside the kiln has been used in the porcelain works at Sèvres and Limoges.

Siemens' pyrometer, in which a small piece of metal is placed in the kiln and withdrawn from thence into a measured quantity of cold water, and the rise in the temperature noted, may be used in connection with potters' kilns, but it does not possess any advantage over more easily manipulated pyrometers, such as the electrical ones or Seger cones. On the Continent a draught gauge is often used in order to control the temperature of the kilns. In this case the temperature itself is not measured, but, as it depends on the rate of combustion and this again on the air supply, an indirect reading is obtained. This method is exceedingly useful as an auxiliary, but, as the figures obtained in draught measurements are always so small, really accurate results cannot be obtained by this means.

The clay worker is, therefore, reduced in practice to three forms of pyrometer if he wishes to obtain constant results. These are :— (1) An accurate electric pyrometer; (2) Fusible masses; (3) Trials.

Of these the first is the most accurate in skilled hands, but is expensive in first cost and in upkeep, and requires considerable skill in use; the others may be used successfully by any intelligent workman who is used to firing.

Two main forms of electrical pyrometer are in use, viz., ' resistance pyrometers' and 'thermo-couples.' The former is in many ways the best and most accurate, except for the very highest temperatures, and, working chiefly under the patents of Professor Callendar and Dr. Griffiths, the Cambridge Scientific Instrument Co. have been able to supply these instruments at a moderate price and of remarkable accuracy and simplicity. The most suitable form of instrument is shown in fig. 28.

Briefly, it consists of a galvanometer, or electricity measurer, connected to an accumulator and to a Wheatstone bridge or current balancer, along which slides a self-acting arrangement,

whereby the recording pen is moved backwards and forwards along the face of the paper cylinder according as the temperature rises or falls. To this recording apparatus is connected, by means of suitable wires, a resistance coil of platinum or a platinum alloy enclosed in a porcelain tube; this forms the thermometer' and is placed inside the kiln, and the connecting wires carried, in a manner similar to telegraph wires, to the recorder, this latter being advantageously placed in the office. As the paper roll revolves at the same time that the pen moves across its surface, a temperature-time diagram is obtained similar to that shown in fig. 29-in which vertical distances represent the temperature, and horizontal distances (from left to right) represent the time.

By means of a suitable switch any number of kilns can be connected, and their indications recorded, though if more than one kiln is at the same temperature the diagram becomes somewhat complicated.

In the thermo-couple' the resistance coils and Wheatstone bridge are unnecessary, and the current produced when two metals joined together are heated is measured directly by the galvanometer. The late Sir. W. C. Roberts-Austen, working on an instrument designed by Le Chatelier, has done an excellent work in introducing this instrument into this country. He also made it self-recording by causing the light from the galvanometer mirror to fall on a rotating drum of photographic paper which is afterwards developed.

Quite recently a combination of a pyroscope (p. 303) and of a thermo-couple pyrometer has been placed on the English market. This instrument, which is an improved form of the original Féry pyrometer, is capable of being used continuously at temperatures above 1000° C., at which the Callendar instrument becomes impracticable for commercial temperature-recording on account of the great wear and tear of the porcelain tubes and the wire forming the resistance.

In the Féry pyrometer the heat radiated through a 'sight-hole' in the kiln is received on to the junction of a copper constantan couple, and the radiation effect is measured by means of a sensitive galvanometer.

As the amount of heat radiated from the inside of a kiln, under the conditions under which such a pyrometer is used, is proportional to the fourth power of the temperature of the kiln, the readings of the galvanometer may, by suitable calibration, be made directly in C.

By using the radiated heat, instead of inserting a part of the pyrometer in the kiln, the wear and tear of the instrument is reduced almost to zero; an ingenious focussing arrangement enables the distance of the hot part of the kiln from the instrument to be varied within wide limits, and the original cost of the instrument is much less than that of the other forms of electric pyrometer, whilst its readings, so far as the author has been able to test them, seem to be equally accurate.

The construction of the instrument is shown in fig. 30, which is a sectional view of the pyroscope. This consists of a metal cylinder about 7 inches long and 4 inches wide, having a small eye-piece (0) at one end. Inside this cylinder, just behind the eye-piece, is a concave mirror which focusses the heat radiated from the kiln on to the thermo-couple junction, the two wires of

which are marked R and D in the figure. In order that the accuracy of the focussing may be as great as possible, two small mirrors are placed immediately in front of the junction in such a manner that when the part of the kiln giving out the radiant heat it is desired to measure is in focus, a clear view of it (reflected from the large mirror on to the smaller ones) is obtained on looking through the eye-piece (O).

With this arrangement the distance of the heated object from the instrument is a matter of indifference so long as the image on the junction is more than sufficient to cover it. When this is the case the instrument gives identical readings at 3 feet and 30 feet from the object.

That the indications of the instrument, when used as directed,

FIG. 30.

are independent of distance can be proved from theory as well as verified in actual work.

The radiated heat develops an electric current whose pressure is proportional to the temperature of the junction receiving the radiation. This current is led by wires from the terminals (b b') to a sensitive galvanometer, which, for convenience, may be graduated to show temperature-degrees instead of milli-volts. This enables the indications of the instrument to be read and understood without any calculation.

As one of the chief objects of measuring the temperature of a kiln is to have a continuous record of the manner in which it has been heated, it is advisable to substitute a recording galvanometer for the direct reading instrument just mentioned.

The Scientific Instrument Co., Ltd., Cambridge (who are the chief makers of electric temperature recorders in this country),