Oct. 19, 1882] 



NATURE 



599 



upper bar). With this eaormous load at this stupendous 

 height, would the citizens of London in the Churchyard 

 below feel themselves in perfect security ? I think not ; 

 and I claim the same privilege of entertaining the sense 

 of insecurity for the proposed Forth Bridge. 



The danger arising from the endwise action of 50 large 

 a force on so long a bar or frame, is produced by the 

 curvature technically called "buckling," and there 

 appears to be fear of its occurrence in various parts of 

 the bracket, and in some parts sequentially, that is to say. 

 that a buckling of a minor order might lead to a buckling 

 of a more important order. Thus, proceeding from the 

 pier, the first support of the great lower bar is by a 

 suspension-rod from the great upper bar ; to which, as 

 regards merely the suspension-rod, there can be no 

 objection. But the upper attachment of this suspension- 

 rod is supported by a thrust-rod about 340 feet long. Can 

 this rod be considered safe against buckling ? In the 

 total absence of experiment or explanation, I may be 

 permitted to express a doubt of safety. And if that rod 

 fail, the corresponding part of the great lower bar will 

 sink, it will buckle under its enormous end-thrust, and 

 the bridge will be ruined. The second support of the 

 great lower bar depends, in like manner, on a thrust-rod 

 whose length is 2J.0 feet ; considerations of the same kind 

 apply to it, though probably in a minor degree. 



Experienced engineers must have known instances in 

 which buildings have failed from want of consideration 

 of buckling. The following occurred within my know- 

 ledge. When the Brunswick Theatre was built, the con- 

 struction of its trussed iron roof was greatly extolled, 

 and Mr. Whewell and myself, then residing at Cam- 

 bridge, and proposing to visit London about the same 

 time, had arranged to inspect the truss. But before we 

 reached London it was ruined. There was no adequate 

 bracing of the principal rafters in the plane of the roof ; 

 the suspension of a very slight weight on the great tie 

 caused the rafters to buckle sideways, and the roof fell, 

 destroying the building. 



I am not aware whether a theory of bucMing finds 

 place in any of the books which treat of engineering in a 

 somewhat mathematical form. But there ought to be 

 such. It can be formed with no difficulty and little 

 trouble, giving such a form of result, that all that will be 

 required in any case, to determine the end-pressure which 

 can safely be applied to the end of a bar, will be ex- 

 pressed in terms of the length of the bar, and the curva- 

 ture caused by a tranversal strain (determined by simple 

 experiment). This theorem ought to be applied in every 

 instance. 



I need scarcely to remark that every construction is 

 liable to chance-errors of unforeseen character, and I 

 think that the proposed construction, which depends for 

 its safety entirely on the maintenance of the thrust-prin- 

 ciple in perfection, is more liable than any other to 

 danger from these causes. A rivet-head may slip, or a 

 screw may strip, and all may be imperilled. Robert 

 Stephenson, when building the Menai Bridge, used every 

 caution that an active mind could invent : in particular 

 he provided that the masonry fcr final support of the 

 tubes should be raised as quickly as possible to take the 

 bearing of the tubes at every moment. Yet an accident, 

 though a small one, did happen. The ends of the tubes 

 were raised by the power of hydraulic presses ; the 

 cylinder of one of theie presses burst, and the end of the 

 tube fell three or four inches. This minute fall, in 

 the judgment of the attendant engineers, gave a strain 

 to the tube such as it never sustained before or since. 

 (This accident came first to my knowledge in a singular 

 way. With the assistance of my friends, Capt. Tupman, 

 R.M.A., and James Carpenter, Esq., and before having 

 heard of the accident, I made experiments on the state 

 of permanent magnetism of the great iron tubes. One 

 of these showed an ar.omalv, somewhat similar to that of 



iron (heavily struck. On my mentioning this to Mr. 

 Edwin Clark and others, the phenomenon was at once 

 referred to the accidental shock which I have described.) 

 Much has been said on the action of the wind, and on 

 the difference of that action upon a suspended bridge, 

 and upon a girder bridge. In regard (first) to the 

 amount of pressure, 1 refer to a former letter of mine, 

 correctly cited in the evidence before the Committee on 

 the fall of the Tay Bridge, in which I state that the 

 maximum pressure may be more than 40 lbs. on 

 the square foot (I should say more than 50 lbs. for Scot- 

 land), but that this action is so limited, both in time and 

 in local extent [<nd is, I add, so continually vaning in 

 direction], that the average of direct pressure probably 

 would not exceed 10 lbs. on the square foot. In regard 

 (secondly) to the difference of wind-action in the two 

 systems of construction ; — the immediate effect of the 

 wind appears to me to be a shock, of limited extent, 

 which is much less likely to be injurious on a com- 

 paratively flexible frame suspended from above, than 

 on a jointed frame where every joint must be tight, and 

 where ruin will follow disturbance. In the proposed 

 Forth Bridge, however, there is risk of danger of the 

 most serious kind, which may perhaps surpass all the 

 other dangers. It arises from the horizontal action of 

 the wind on the great projecting brackets, and its ten- 

 dency to wrench them laterally from their attachments. 

 The ruinous force depends, not simply on the magnitude 

 of the wind's pressure, but also on its leverage; as mea- 

 sured by the proportion cf the height of the Tay Bridge 

 or the length of the bra., et of the Forth Bridge, to the 

 separation (in each case) of their horizontal attachments to 

 the solid piers. This leverage is considerably greater in 

 the instance of the proposed Forth Bridge than it was in 

 that ot the unfortunate Tay Bridge ; and we may reason- 

 ably expect the destruction of the Forth Bridge in a 

 lighter gale than that which destroyed the Tay Bridge. 



I may now collect the heads of my remarks on the 

 proposed Forth Bridge : — 



I. The proposed construction is, as applied to railway- 

 bridges, entirely novel. 



II. The magnitude of its parts is enormous. 



III. These has been no succession of instances of the 

 construction, with rising degrees of magnitude, which 

 might furnish experimental knowledge of tome of the 

 risks of construction. 



1Y. The safety of the bridge depends entirely on a 

 system of end-thrusts upon very long rods ; a system 

 which appears generally objectionable, but particularly so 

 when the length of the rods is very great. 



Y. No reference is made to theory applied to the 

 buckling of rods under end-thrusts. 



VI. The liability to ruinous disturbance by the lateral 

 power of the wind acting with the leverage of the long 

 brackets appears to be alarmingly great. 



My own impression is, that the proposed construction 

 is not a safe one, and I should be happy to hear that it is 

 withdrawn. 



I refer unhesitatingly to " the Suspension Bridge " as 

 the construction which I should recommend. On this 

 system generally I remark : (1) that I am incredulous as 

 to the oscillation of 8 feet in extent, or any sensible part 

 of it; (2) that if the railway is slightly arched upwards to 

 the degree corresponding to depression caused by an 

 average train, such a train will run on a horizontal plane ; 

 (3) that a stiffening lattice may be used with very good 

 effect against vertical oscillations from all causes. 



The considerable htight of the piers, and the great 

 length of the suspension-chains, are matttrs to be viewed 

 carefully. 



To reduce them as far as possible, I would suggest for 

 examination the following proposals ; — 



