The Tay Bridge. 27 



over and stop traffic whenever a wind blowing across the 

 bridge exerted a pressure of more than 20 lbs. per square 

 foot. The contrivance would be inexpensive, and might 

 prevent an appalling catastrophe. 



2. The second theory is that the girders themselves were 

 overturned by the pressure of the wind on their upper 

 members, turning about the lower leeward edge of each lower 

 member as axis. This theory is favoured by the circumstance 

 that, owing to the space between them being utilised for 

 traffic, it was impossible to adopt the usual transverse tracing 

 in a vertical plane. This sort of overturning would be 

 antagonised by the combined moments of resistance of all 

 the cross-girders, distance-pieces, &c. The exact amount 

 of this aggregate moment we have not sufficient data to 

 compute, and therefore cannot follow up the theory further. 

 It is, however, completely negatived by the appearance of 

 the wreck, which shows the main girders and their cross- 

 girders still united at right angles. The suggestion, though 

 thus not applicable in the present case, is worth keeping in 

 view, for it is quite conceivable that large plate-girder 

 bridges might be very heavily strained in this way, and 

 should be correspondingly braced. 



3. The third theory is that the bridge was overturned 

 as a whole by the wind, the skeleton piers turning about 

 the base of the leeward column as a hinge. In order to 

 test this view by computation we must ascertain the weight 

 and dimensions of the structure. There appears to have 

 been, according to the best information we have, about 380 

 tons of material in each iron pier with 245 feet length of 

 girder. The width of the base, taken to the centres of the 

 columns, being 22 feet, the moment of stability becomes 

 380 x 11 = 4180 ft.-tons. 



It is difficult to determine the exact area of the structure 

 which was effectively exposed to the wind. Each girder 

 presents a surface of 1960 square feet, but the question 

 arises to what extent the leeward girder will be 

 sheltered. Some have assumed that the windward girder 

 alone was affected, but this is manifestly a mistake, 

 for the bars of which the girders were composed being 

 from 6 to 20 inches in width only, and the spaces between 

 them triangles and squares of from 70 to 400 square feet 

 each, the amount of shelter given would be but small at 

 a distance of 15 feet — about as much, as has been humor- 

 ously remarked, as a man would experience on a windy day, 



