THE BRITANNIA AND CONWAY TUBULAR IRON BRIDGES. 
337 
duced simply by each of the normal forces (longitudinal, transversal, vertical) ; and we 
shall exhibit the proportion of the disturbance produced to the force which produces it. 
Values of the fraction 
Disturbance of magnetic force 
Normal force which produces it* 
Description of Stations. 
Traction for 
longitudinal 
force. 
Fraction for 
transversal 
force. 
Fraction for 
vertical 
force. 
+ 1-6745 
— 01199 
-0-4729 
-1-0212 
-0-9421 
— 0-9391 
-0-2348 
-0-9838 
-0-9556 
-0-1278 
-1 1570 
-1 0044 
— 1-0038 
-0-3987 
-0 9593 
+0 0059 
-0-2332 
-0-8494 
-1-0075 
-1-0442 
-0-9885 
-0-2797 
-0-8335 
-0-9240 
+ 1-1911 
— 0 7355 
-0-6160 
Chester Tower 
Middle of Conway Bridge 
Holyhead Tower 
-5-8326 
+ 1-9801 
-8-5674 
-0-7299 
-0-8174 
-0-8026 
-0-8335 
—0-9400 
-0-7468 
The numbers in the third and fourth columns of this Table, including the Conway as 
well as the Britannia Bridge, present a good agreement. Their general result is this : 
that in the axis of a rectangular tube, at all parts except very near the ends, the action 
of external magnetic forces in planes normal to the axis is absolutely destroyed. In the 
second column for the Britannia Bridge there is one anomaly at Middle of Anglesey Water 
Tube of which I can give no explanation. It is a certain fact of observation ; it may 
arise from some peculiar steely character of the iron in its neighbourhood. Putting this 
aside, the other numbers for the Britannia Bridge are exactly what we should have 
expected from a structure so closely resembling a bar magnet ; the longitudinal force is 
greatly increased at the ends, but is diminished in all other points; not, however, as in 
the transversal forces, to its complete annihilation, but diminished by about one fourth 
part. For all these cases the circumstances of position of the bridge with respect to the 
magnetic meridian are favourable. 
For the second column, as applying to the Conway Bridge, the position is unfavourable, 
the axis of the bridge being very nearly transversal to the magnetic meridian, and the 
denominator of the fraction consequently very small. Still, referring to the last Table 
but two, there can be no doubt that the character of the forces is such as we should have 
expected in a tube whose length is directed in the N.W. quadrant, instead of the S.W. 
I can give no accurate explanation of this phenomenon. It would almost seem that, even 
in a structure so simple and so rigorously rectangular as the bridge-tube, we cannot treat 
the effect of one of the rectangular forces as being strictly confined to that rectangular 
direction ; yet I do not see how the transversal actions can explain longitudinal force 
in one direction rather than in the other. 
It may be worth adding that, in the course of this discussion, I have conceived that 
possibly some difference of effects may have arisen from the difference in the direction 
of the iron planks of which different parts of the bridges are built ; and to test this, I 
