ON RAILWAY BAR CORROSION. 109 
is greater or less with the same breadth of gauge, dependent upon the po- 
sition of the meeting-points of the ends of the rails, which are sometimes 
nearly opposite across the line, but often not so, i.e. the pole of one length 
of rail is opposite nearly to the neutral point of the rail at the other side of 
the railway. 
It is manifest from what precedes, that the polar intensity of any given 
rail ina line of railway depends, not only upon the rolling traffic it is exposed 
to, but also upon the direction of the line of railway itself; that the bars in 
railways running north and south are in a higher state of magnetism, other 
things being the same, than in those running east and west, by the effects of 
terrestrial induction. 
In long lengths of railway, running east and west especially, but in some 
degree in every direction, there are constant thermo-electriec currents tra- 
_yersing the rails from end to end, due to changes in atmospheric temperature, 
between day and night, &c.: such currents have been already noticed by Mr. 
William Barlow as perpetually traversing the wires of the electric telegraph, 
and such currents may occasionally be due to other causes of disturbance of 
the equilibrium of terrestrial electricity than those due to temperature. 
But the passage of locomotive engines over railways is a cause of electric 
currents traversing the bars of a much more important and formidable cha- 
racter. Ifa galvanometer be placed in connexion with a rail forming part 
of a line of way, and also with the earth at some distance, or with the rails 
at a distant point in advance, powerful deflections are produced by the ap- 
proachof a locomotive engine, or even by therolling along of a heavy train with- 
out an engine; in the latter case the effects are comparatively feeble, and appear 
to be due to the repeated compressions and rendings of crystallized surfaces, 
to which the surfaces rolling and rolled on are subjected ; an action shown 
by Becquerel and others, to be an efficient producer of electric disturbance ; 
but in the case of the locomotive engine, a torrent of electric force is let 
loose and finds its way into the earth along the rails, which, from the im- 
perfect contact of their junctions, permit it to pass along the line with diffi- © 
culty, and thus the equilibrium is gradually restored in great part through 
the chairs, fastenings and ballast, the resistance being greatest where the line 
is laid on Jongitudinal sleepers, and these are quite dry. 
There can be little doubt, that if a portion of railway were insulated to- 
lerably from beneath on wood, and a wood block inserted at given points, so 
as to cut off a segment of rails from the line, powerful sparks would flash 
from the rails and train as it passed over this portion of the railway. Indeed 
all this may be inferred from the well-known facts of the hydro-electric 
machine. i 
These electrical effects of locomotive engines in motion are somewhat 
uncertain, owing to the saline contents of the water usually employed fre- 
quently interfering; but after an engine has run a considerable distance 
without stopping, and the passages of efflux steam have become cleaned by 
its continued passage, and the engine does not prime, the evolution of elec- 
tricity_is always considerable. 
_ As each railway bar may be viewed in the light of a conductor through 
which currents of electricity of variable intensity and quantity are per- 
petually flowing, it is obvious that magnetic currents are also in constant 
circulation round each bar and at right angles to its length: thus if upon 
the bar a, 6 an engine run over from south to north, the north end of a mag- 
Netic needle placed above the rail will be deflected to the left-hand, and vice 
