= 
208 Conduction of Galvanic Eléctricity through Moist Air. 
been able to notice any irregularity in these changes except as 
to the time. The irregular disturbances are very interesting, and 
may be identical with the magnetic storms of Gauss. Upon cer- 
tain days they are hardly perceptible, though they never cease 
altogether. On some days they are violent, if I may be allowed 
the expression. 'The needle does not take a sudden start and re- 
turn as when influenced by lightning, but moves gradually with- 
out oscillation to some fixed point, from: which it will return 
sometimes in two minutes and sometimes in ten or fifteen min- 
utes. An extended series of observations will be necessary be- 
fore any deductions can be safely made. If the wires should be 
separated by a slight interval during a thunder storm, doubtless 
electrical sparks would be visible. During heavy storms, a flash 
of lightning twenty miles distant from the wires of Morse’s tele- 
graph will induce electricity in the wire sufficient to operate the 
magnets, and work the telegraph, sometimes recording several 
signals. A flash of lightning in Baltimore, forty miles distant from 
this place, ‘will 7 the ee: at this end of the line. 
Washington, D. C. 
Appendiz. —Mr. Lane remarks, in his communication on the 
electric conduction in metals,* that in his experiments he found 
no confirmation of the statement made in this Journal t a few 
years since, ‘that the conducting power of a wire is greatly i im- 
paired by bending or twisting it.” Mr. Lane has somewhat mag- 
nified my meaning by introducing the word greatly, which I did 
not use, and it is probable that he has really confirmed my state- 
ment, as he remarks, that after winding and unwinding a thick 
Wire hava times over a cylinder less than an inch in diameter, 
the conducting power appeared scarcely affected, implying that 
some effect was produced ; and if in Mr. Lane’s ingenious and 
well arranged experiments, which were made with only a few 
feet of wire, he found even the slightest loss of conducting pow- 
er, it is reasonable to suppose that in many hundred feet of wire 
the loss would be a very notable quantity. A statement which 
I also made in addition to the above, should have been explained. 
The Statement was that ‘‘a wire which has once heen wound 
pee a magnet is not fit for the same purpose again.” There are 
ms Le _ * Bee this Journal, Second Series, 1, 230. - 
CER SE Bey ite this Journal, Fi irst Series, xxxv, 109. 
ei alent: 
