250 WRITINGS OF JOSEPH HENRY. [1846 



room so as to form a parallelogram of about sixty feet by- 

 thirty on the sides; and in the cellar of the same building, 

 immediately below, another parallelogram of the same dimen- 

 sions was placed. When a spark from an electrical machine 

 was transmitted through the upper parallelogram an induced 

 current was developed in the lower one, sufficiently powerful 

 to magnetize needles, although two floors intervened, and 

 the conductors were separated to the distance of thirty feet. 

 In this experiment no electricity passed through the floors 

 from one conductor to the other; the effect was entirely due 

 to the repulsive action of the electricity in motion in the 

 upper wire on the natural electricity of the lower. In another 

 experiment two wires, about 400 feet long, were stretched 

 parallel to each other between two buildings; a spark of 

 electricity sent through one produced a current in the other, 

 though the two were separated to the distance of 300 feet; 

 and from all the experiments it was concluded that the dis- 

 tance might be indefinitely increased, provided the wires 

 were lengthened in a corresponding ratio. 



That the same effect is produced by the repulsive action 

 of the electrical discharge in the heavens is shown by the 

 following modification of the foregoing arrangement. One 

 of the wires was removed and the other so lengthened at one 

 end as to pass into my study and thence through a cellar 

 window into an adjacent well. With every flash of lightning, 

 which took place in the heavens within at least a circle of 

 twenty miles around Princeton, needles were magnetized in 

 the study by the induced current developed in the wire. 

 The same effect was produced by soldering a wire to the 

 metallic roof of the house, and passing it down into the well; 

 at every flash of lightning a series of currents, in alternate 

 directions, was produced in the wire. 



I was also led, from these results, to infer that induced 

 currents must traverse the line of a railroad, and this I found 

 to be the case. Sparks were seen at the breaks in the con- 

 tinuity of the rail with every flash of a distant thunder 

 cloud. 



Similar effects, but in a greater degree, must be produced 

 on the wire of the telegraph, by every discharge in the 



