ON RAILWAY BAR CORROSION. 107 
When this bar and some others were turned at right angles to the mag- 
netic meridian, and examined with the suspended needle moved vertically 
up and down before their extremities, the existence of two poles at each end 
of the bar was made evident; thus the needle Fig. 8. 
was quiescent at a certain point in the depth 
of the bar when held thus opposite the end " | 
of it, but was attracted or repelled in op- Nec = 
posite directions when above or below this 
oint. 
This fact was subsequently ascertained to apply to all the railway bars; _ 
that is to say,.a railway bar when polar is a magnet of such thickness, that 
it presents poles at its solid angles, not only with reference to length, but to 
depth, these poles being always of unequal intensity, the bar being in fact in 
the predicament of a cube or large parallelopiped of iron when exposed to 
magnetic induction. Hence the bar last adverted to with three neutral points 
had in fact sixteen poles distributed along its length; eight on the top and 
eight on the bottom flange or edge, and alternately of greater and of less in- 
tensity along the same edge, thus :— 
Fig. 9. 
N 

meulr 
-|--meubral 
-|-aewbral 


These secondary poles, or those of depth, were not altogether due to ter- 
restrial induction, as they preserved their signs, though with diminished 
intensity, when the bar was turned upside down, 2. e. when the top edge 
became the lower, the direction and ends of the bar remaining unmoved, but 
they are probably due to induction within the bar itself. 
Similar phenomena to the above present themselves in the same bars when 
examined after removal from the Dalkey line, as was to have been expected, 
there being no change in their condition in either case, viz. not having been 
travelled over on either line. 
The set No. 2, not coated, and exposed both to oxidation and to abrasion of 
traffic, when examined in the same way, after experiment on the Kingstown 
line, all proved to be powerfully magnetic with polarity ; the idio-polarity 
being almost in every case sufficiently intense to neutralize or reverse the po- 
larity of terrestrial induction. In almost every instance there were two 
well-defined poles at the extremities of the bars, with one neutral point 
between. 
In every rail the S. pole of the bar was found in the direction in which 
the traffic came in upon it, and the N. pole at that at which the traffic rolled 
off from it.. Now the traffic passed over these bars in a direction from S.E. to 
N.W., and hence the direction of permanent polarity conferred upon the bar, 
coincides with the direction (in this instance at least) in which the traffic 
rolled over it. 
The same bars when examined after removal from the Dalkey line (where 
it will be remembered the traffic is in both directions from W.N.W. to E.S.E, 
and vice versd, and the polarity of the bars acquired on the Kingstown line 
having been afterwards destroyed by heating them all to redness), were 
found powerfully idio-polar; some of the bars much more so than in the 
former case, so as to be able, when brought into the direction of dip, and thus 
