Besides the voltaic cell, in which the current is produced by chemical action, there is the thermo-electric battery, or thermopile, which produces current directly from heat energy. About 1822 Seebeck was experimenting with voltaic pairs of metals, and he found that a current could be produced in a complete metallic circuit consisting of different metals joined together, by keeping these joinings at different temperatures. Fig. 10 shows a simple arrangement for demonstrating this effect, which is known as the “Seebeck effect.” A slab of bismuth, BB, has placed upon it a bent strip of copper, C. If one of the junctions of the two metals is heated as shown, a current flows; and the same effect is produced by cooling one of the junctions. This current continues to flow as long as the two junctions are kept at different temperatures. In 1834 another scientist, Peltier, discovered that if a current was passed across a junction of two different metals, this junction was either heated or cooled, according to the direction in which the current flowed. In Fig. 10 the current across the heated junction tends to cool the junction, while the Bunsen burner opposes this cooling, and keeps up the temperature. A certain amount of the heat energy is thus transformed into electrical energy. At the other junction the current produces a heating effect, so that some of the electrical energy is retransformed into heat. [Illustration: FIG. 10.--Diagram to illustrate the Seebeck effect.] [Illustration: FIG. 11.--Diagram to show arrangement of two different metals in Thermopile.] A thermopile consists of a number of alternate bars or strips of two unlike metals, joined together as shown diagrammatically in Fig. 11. The arrangement is such that the odd junctions are at one side, and the even ones at the other. The odd junctions are heated, and the even ones cooled, and a current flows when the circuit is completed. By using a larger number of junctions, and by increasing the difference of temperature between them, the voltage of the current may be increased. Thermopiles are nothing like so efficient as voltaic cells, and they are more costly. They are used to a limited extent for purposes requiring a very small and constant current, but for generating considerable quantities of current at high pressure they are quite useless. The only really important practical use of the thermopile is in the detection and measurement of very minute differences of temperature, which are beyond the capabilities of the ordinary thermometer. Within certain limits, the electro-motive force of a thermopile is exactly proportionate to the difference of temperature. The very slightest difference of temperature produces a current, and by connecting the wires from a specially constructed thermopile to a delicate instrument for measuring the strength of the current, temperature differences of less than one-millionth of a degree can be detected. CHAPTER V THE ACCUMULATOR