1823.] M. Schweiggei-'s Electromagnetic Multiplier. 441 



Exper. 1. — We formed a rectangular circuit abed, fig. 4, 

 one-half of which was antimony, and the other of bismuth ; acd 

 and a b d soldered together, so that the two contiguous were of 

 antimony, and the other two of bismuth. One of the sides was 

 nearly four and a half inches long, and the other three inches ; 

 the circuit was placed horizontally upon stands, with two of its 

 sides in the direction of the magnetic needle, and the compass 

 was placed upon one of them. Having left the circuit for a 

 time sufficient to regain the equilibrium of temperature, which 

 might have been disturbed during the placing of it, ice was put 

 upon one of the two solderings, a or d, which unite the two hete- 

 rogeneous metals. The compass showed a deviation of 22 or 23 

 degrees ; the temperature of the air was 67° of Fahr. ; at a tem- 

 perature of 68°, the deviation was observed to be 30 degrees. 

 We neglected to note the temperature of the atmosphere at the 

 commencement of the experiment. We shall, therefore, only 

 compare the results of experiments made at the same period. 



Exper. 2. — Another circuit, fig. 5, was formed of the same 

 length as the former; but having the opposite sides of the 

 same metal, a b and c d being bismuth, and a c and h d of anti- 

 mony so that the circuit was composed of two thermoelectric ele- 

 ments rendered active by ice placed upon two opposite angles. 

 This circuit produced a deviation of 30 to 31 degrees, under the 

 same circumstances in which the simple circuit produced a 

 deviation of 22 to 23 degrees. The temperature in this circuit 

 has its equilibrium soon restored, so that the thermoelectric 

 effect appears weaker than it would do without this circumstance. 



Exper. 3. — A circuit ABCD, fig. 6, the circumference of 

 which was double that used in the first experiment, was put in 

 action by ice placed upon one of the solderings. The deviation 

 was only from 13° to 15°, under the same circumstances which, 

 with the circuit, fig. 4, gave 22 or 23 degrees. 



Exper. 4. — Another circuit, fig. 7, was formed, of the same 

 length as the preceding, but it had four alternations, or four 

 thermoelectric elements a b {a being the antimony, and b the 

 bismuth). This circuit was put in action by placing ice upon 

 every other soldering. The deviation of the needle amounted to 

 3l£ degrees, under the same circumstances in which the simple 

 circuit of equal length produced a deviation of only 13 to 15 

 degrees, but the circuit used in Exper. 2, fig. 5, which had only 

 half its circumference, and half the number of elements, pro- 

 duced nearly the same effect as that obtained in this experiment. 

 Thus it appears, which will be confirmed in the sequel, that the 

 deviations of the needle produced by the thermoelectric circuit 

 increase with the number of the elements when the length of the 

 circuit remains the same, but that they become weaker in pro- 

 portion as the length is increased. It is also evident, and it 

 will be rendered still more so in the sequel, that these two 

 ■effects balance each other; so that the effect of a circuit is not 



