On the night of Sunday, December 28, 1879, a violent storm (10 to 11 on the Beaufort Scale) was blowing at virtually right angles to the bridge. Witnesses stated that the storm was as strong as the worst they had seen in the 20-30 years they had lived in the area, and one described it as a 'hurricane', as bad as a typhoon he had seen in the area. China Sea.[10] Wind speed was measured in Glasgow - 71 mph (114 km/h; 32 m/s) (one-hour average) - and Aberdeen, but not in Dundee.

Higher wind speeds were recorded at shorter intervals, but at the inquest an expert witness warned of their unreliability and refused to estimate conditions at Dundee from readings taken elsewhere.[11] A modern interpretation of the available information suggests that the winds blew at 80 mph (129 km/h; 36 m/s).[12]

Use of the bridge was restricted to one train at a time by a "Block (Railway)") blocking system that used a baton as a token, which was picked up at one collateral signal cabin and delivered at the other. At 7:13 p.m. m. A train to and from Burntisland[13]​ (consisting of a type 4-4-0 locomotive, its tender, five passenger cars,[note 1]​ and a luggage van[14]​) slowed to collect the baton from the signal cabin at the south end of the bridge, and then headed towards the viaduct, gaining speed.

The signalman turned to note the passing of the train and then attended to the fire in the cabin, but a friend of his observed the train, and when he was about 200 yards (182.9 m) from the cabin, he saw sparks coming from the wheels on the east side, which he had also seen on the previous train.[15] During the investigation, John Black testified that the wind pushed the cars so that the flanges of the wheels made contact with the rails. Black explained that the internal protective barriers against derailment were slightly higher than the railings.[15]​[note 2]​ This arrangement would hold the good wheel when the derailment was caused by the disintegration of one of the two wheels on an axle, a real risk before steel wheels became widespread, a situation that had occurred in the Shipton-on-Cherwell railway accident) on Christmas Eve 1874.

The sparks continued for no more than three minutes, at which time the train was traveling along the high beams.[16]​ At that point "there was a sudden flash of bright light, and in an instant the darkness was total, the rear lights of the train, the sparks and the flash of light... disappeared at the same instant."[17]​ The signalman saw none of this and did not believe it when told.[note 3]​ When the train did not appear on the alignment leaving the bridge towards Dundee, he attempted to speak to the signal box at the north end of the bridge, but discovered that all communication had been lost.[18]​.

Not only was the train in the river, but so were the tall girders and much of the iron in its supporting pillars.[19] Divers who explored the wreckage later found the train still inside the girders, with the engine inside the fifth span of the 5-span southern division.[20] There were no survivors; Only 46 bodies were recovered[21]​ but there were 59 known victims. Fifty-six tickets to Dundee had been collected from the train passengers before crossing the bridge. Taking into account season ticket holders, cross-ticket tickets and railway employees, it was thought that 74 or 75 people were on the train.[17] It has been suggested that there were no unknown victims and that the higher figure of 75 arises from a double count in an initial newspaper report,[22] but the inquiry did not take its casualty figures from the "Dundee Courier", and took sworn testimony and made its own counts.

• - The locomotive fell again during the rescue, but was eventually recovered and running again.

• - The left front of the recovered locomotive tender.

• - Remains of two rescued passenger cars.

• - Fallen beams, with remains of a wooden passenger car.

Evidence

A judicial court of inquiry was immediately established (under Section 7 of the Railway Regulation Act 1871") "into the causes and circumstances occasioning" the accident): Henry Cadogan Rothery"), Commissioner of Shipwrecks, President, supported by Colonel Yolland") (Inspector of Railways) and William Henry Barlow, President of the Institution of Civil Engineers"). On January 3, 1880, they were gathering evidence in Dundee; They then appointed Henry Law (a qualified civil engineer) to carry out detailed investigations. While awaiting their report, they held further hearings in Dundee (26 February to 3 March). Once the collection of evidence was complete, they settled in Westminster (19 April to 8 May) to consider the engineering aspects of the collapse. redacted)"),[24]​ both engineers with extensive experience in major cast iron structures). The terms of reference did not specify the underlying purpose of the inquiry: to avoid a repetition, to assign blame, to clarify responsibility or culpability, or to establish precisely what had happened. This led to difficulties (culminating in confrontations) during the Westminster sessions. When the court reported its findings in late June, there was both an Inquiry Report signed by Barlow and Yolland, and a report drawn up by Rothery.

Two witnesses, who were looking at the high beams from the north almost straight on, had seen the lights of the train as far as the third and fourth high beams, when they disappeared; this was followed by three flashes in the high beams to the north of the train. One witness said that the flashes advanced towards the north end of the high beams with approximately 15 seconds between the first and the last;[25]​[note 4]​ while the other stated that they all occurred at the north end, with less time in between.[26]​ A third witness had seen "a mass of fire fall from the bridge" at the north end of the high beams.[27]​ A fourth said he had seen a beam fall into the river at the north end of the beams. high, then a light appeared briefly in the tall southern beams, disappearing as another beam fell; He did not mention the fire or the flashes.[28]​[note 5]​

'Ex-Provost' Robertson[note 6]​ had a good view of most of the bridge from his home in Newport-on-Tay,[31]​ but other buildings blocked his view of the tall southern girders. He had seen the train moving toward the bridge; then, on the high northern girders, before the train could reach them, he saw "two columns of steam illuminated with the light, first one flash and then another" and could no longer see the lights on the bridge;

Former Mayor Robertson had purchased a season ticket between Dundee and Newport in early November and became concerned about the speed of northbound local trains across the high beams, which had been causing noticeable vibrations, both vertical and lateral. After complaining three times to the station master in Dundee, his complaints did not affect the speed of the trains. Starting in mid-December he had used his season ticket to travel only south, using the ferry for northbound crossings.

Robertson had timed the train with his pocket watch, and to give the railroad the benefit of the doubt he had rounded to the nearest five seconds. The time measured across the beams (3,149 feet (960 m)) was normally 65 or 60 seconds,[note 8] but twice it had been 50 seconds. Watching from the shore, he had measured 80 seconds for trains traveling across the bridge, but not any he had ridden on. Local trains heading north often stopped to avoid express delays, then made up time as they traveled over the bridge. The gradient towards the bridge at the north end prevented similar high speeds on southbound local trains. Robertson said the movement he observed was difficult to quantify, although the lateral movement, which was probably 1 to 2 inches (25.4 to 50.8 mm), was definitely due to the bridge, not the train, and the effect was most marked at high speed.

Four other passengers on the train confirmed Robertson's stated times, but only one had noticed any movement of the bridge.[35][note 9] The Dundee station master had relayed Robertson's complaint about speed (he had been unaware of any concerns about sway) to the conductors, and then checked the times from cab to cab (at either end of the bridge the train was traveling slowly to pick up or deliver the witness). However, I had never verified the speed through the high beams.[37]​.

Painters who had worked on the bridge in mid-1879 said that it shook when a train passed.[38][note 10] When a train entered the high southern girders, the bridge shook at the north end, both from east to west and, more strongly, up and down.[41] The shaking was worse when the trains went faster, which they did: "when the Fife ship was very close and the train had only reached the south end of the bridge, "[42] A carpenter who had worked on the bridge from May to October 1879 also spoke of a lateral shaking, which was more alarming than the up and down movement, and greatest at the southern junction between the high beams and the low beams. He was unwilling to quantify the range of motion, but when pressed he gave a figure of about 2 to 3 inches (50.8 to 76.2 mm). When pressed again, he would only say that the movement was distinct, wide and noticeable.[43] One of the painters' foremen, however, said that the only movement he had seen had been from north to south, and that it had been less than 1/2 of an inch (12.7 mm).[44]

The British Northern Railway was responsible for maintaining the tracks, but relied on Bouch to oversee the maintenance of the bridge, and appointed Henry Noble as its bridge inspector.[45] Noble, who was a bricklayer, not an engineer, had worked for Bouch on the construction of the bridge.[46]

While checking the foundations of the piles to see if the riverbed was being cleared around them, Noble had noticed that some diagonal tie bars were 'chattering',[note 11]​ and in October 1878 he had begun to remedy this problem. Diagonal bracing was provided by flat bars running from a lug at the top of one column section to two sling plates bolted to a lug at the base of the equivalent section on an adjacent column. The sling bar and plates had a matching longitudinal groove. The tie bar was placed between the sling plates with the three slots aligned and overlapping, and then a swivel was inserted through the three slots and secured. Two "cotters" (metal wedges) were then placed[note 12]​ to fill the remainder of the slot overlap and driven in tightly to tighten the tie.

Noble had assumed that the cotter pins were too small and had not initially been inserted with enough force, but in the vibrating flanges the cotter pins were loose, and even if driven all the way in, they did not fill the slots to put the bars in tension. By placing an additional piece of casing between the loose cotter pins and inserting the cotter pins, Noble re-tightened the loose ties, which stopped rattling. There were more than 4,000 key joints on the bridge, but Noble said only about 100 had needed to be retensioned, most in October-November 1878. At his last inspection in December 1879, only two fastenings had needed attention, both on the piers north of the high beams. Noble had found cracks in four column sections, one below the tall beams and three north of them, which had been tied together with wrought iron hoops. Noble had consulted Bouch about the cracked columns, but not about the tableting of the diagonal pieces.[48]​.

Wormit foundry workers complained that the columns had been cast with 'Cleveland iron', which always contained slag; It was less easy to cast than 'good Scotch metal'[49][note 13] and was more likely to produce defective castings. The molds were moistened with salt water, the cores[50]​ were improperly clamped and moved, resulting in uneven wall thickness of the castings.[51]​ The foundry foreman explained that where the lugs had been cast imperfectly; the missing metal was added "by burning".[note 14]​ If a piece had cavities or other casting defects considered minor failures, they were filled with 'Beaumont egg',[note 15]​ a material of which the foreman had stocks for that purpose.[55]​.

The Gilkes production plant staff were inherited from the previous contractor. Under the resident engineer were seven subordinates, including a foundry manager. The original foundry manager left before most of the tall beam pillar sections were cast. His replacement also supervised the construction of the bridge and had no prior experience supervising casting work.[56] He was aware that they were repairing defects with hot casting,[57] but the foreman had concealed the use of the Beaumont egg from him.[58] When shown defects in the bridge castings, he said he would not have approved the filled pier parts for use, nor would he have passed profiles with wall thicknesses noticeably uneven.[56]​ According to his predecessor, the hot filling had only been carried out on the temporary 'lifting piers' pieces, which were used to lift the truss girders into place and were not part of the permanent structure of the bridge.[59]​ The instructions of the resident engineer were followed,[60]​ who was also not very experienced in casting techniques and trusted the foreman.[61]​.

While work practices were Gilkes' responsibility, his contract with the railroad stipulated that all work performed by the contractor was subject to Bouch's approval of the workmanship. Bouch would therefore share the blame for any resulting defective work on the finished bridge. The foreman of the original foundry, who had been fired for his drunkenness, testified that Gilkes personally approved of irregularities in the early foundries: "Mr. Gilkes, sometimes once a fortnight and sometimes once a month, would strike a profile with a hammer, first on one side and then on the other, and used to review most of the pieces by the way they sounded."[62] Bouch had spent over £9,000 on the inspection. (his total fee was £10,500)[63] but he did not present any witnesses who had inspected the castings on his behalf. Bouch himself had gotten up once a week while the design was being changed, but "later, when construction started, he didn't go as often."[64]

Bouch maintained his own "resident engineer", William Paterson, who looked after the construction of the bridge, its approaches, the line to Leuchars and the branch to Newport, who was also the engineer of Perth General Station.[64] Bouch told the court that Paterson's age was "very close to my own" but, in fact, Paterson was 12 years older than Bouch[note 16] and, at the time of the investigation, he was suffering from a paralysis and could not give evidence.[66] Another inspector appointed later[66]​ was then in South Australia and was also unable to give evidence. Gilkes' managers could not answer for any inspection of the castings by Bouch's inspectors.[67] The finished bridge had been inspected on Bouch's behalf to check the quality of assembly, but that was after the bridge had been painted (although still before the bridge was opened, and before the witness painters were on it in the summer of 1879), which hid any cracks or signs of hot repair (although the inspector said that, in In any case, he would not recognize those signs with the naked eye).[68] Throughout the construction, Noble had taken care of the foundations and masonry.[note 17]​.

Henry Law, who had examined the remains of the bridge, reported defects in workmanship and in the details of the design. Cochrane and Brunlees, who gave evidence later, largely agreed.

• - The pillars had not moved or settled, but the masonry of the bases showed little adhesion between the stone and the cement: the stone had been left too smooth and had not been wet before adding the cement. The anchor bolts, with which the bases of the pillars were held, were poorly designed and were inserted into the masonry with a design that was too simple, without sufficient grip.[70]​.

• - The "Flange (pipes)") connecting flanges between the column sections were not completely facing each other (that is, they had not been machined to obtain smooth, flat surfaces that fit perfectly together). The dowels that should have provided continuity from one section to the next were not always present,[note 18]​ and the bolts did not fill the corresponding holes. Consequently, the only thing that resisted the sliding of one flange over another was the tightening action of the bolts.[72] This tightening effect was reduced because the heads of the bolts and rivets were not thick enough; some had burrs") of up to 0.05 inches (1.3 mm) (he presented an example). This undermined any fastening effect, since using rivets in the joint at the base of a pillar and subsequently riveting them would leave more than 2 inches (50.8 mm) of free play at the top of the pillar. The riveted pieces used were abnormally short and thin.[73]​.

• - The pillar profiles had unequal wall thickness, as much as 1/2 inch (12.7 mm) difference; sometimes because the core had moved during casting, sometimes because the two halves of the mold were misaligned. Thin-walled metal parts were undesirable, both because of their lower intrinsic strength capacity and because (since they cooled more quickly) they would be more vulnerable to the generation of defects.

Here "(providing a sample)" is a nodule of cold metal that has formed. The metal, as expected on the thin part, is very imperfect. Here there is a defect that extends through the thickness of the metal. Here is another and here is another... The entire upper side of this pillar will be found to fit this description, completely filled with hollows and ashes. There are enough pieces here to demonstrate that these defects were widespread.[74]​

Bouch said that the uneven thickness was due to incorrect workmanship, and that if he had known this, he would have taken steps to mold the pieces vertically, but he still considered them safe.[75]

• - The horizontal iron braces of the Canal did not abut correctly with the body of the pillars; correct spacing depended on the bolts being properly tightened (previous comments about lack of coating applied here as well). Because the holes in the lugs were not pre-drilled, their position was only approximate and some horizontal braces had been field-installed, leaving burrs of up to 3/16 of an inch (4.8 mm).[74]

• - In diagonal bracing, the keys were roughly forged and left uncoated, and were too small to withstand the compressive force that the bracing bars could exert.[note 19]​.

• - In the southernmost fallen pillar, a piece of covering had been placed on each link bar to the base.[76]​.

• - The lug screw holes were cast with a conical shape; Consequently, contact between the bolt and the lug was made by the thread of the bolt bearing against the sharp edge at the outer end of the hole. The shoulder was easily crushed and allowed play to develop, and the off-center loading caused the lugs to fail at much lower loads than if the hole were cylindrical.[77] Cochrane added that the bolts would permanently bend (loosening the link of the tie bars to the point that they had to be reinforced by cover pieces) at a load even lower than that at which the cotter pins would deform; He had found some bent tie rod bolts as apparent confirmation.[78]

• - The bracing had failed by giving way to the lugs; In almost all cases, the fractures passed through the holes. Law had seen no evidence of hot-repaired lugs,[77] but some lug failures with fragments torn off around the rivets indicated the use of hot-repaired parts. Furthermore, the paint on the intact pillars would hide any evidence of hot repairs.[79]​.

• - On some pillars, the base sections were still standing; in others, sections of the base had fallen to the west.[80] Cochrane noted that some fallen girders lay on top of the eastern pillars, but the western pillars were on top of the beams; Therefore, the engineers[80]​[81][82]​ agreed that the bridge had broken before it fell, not as a result of its fall.

• - Marks on the south end of the southernmost tall beam indicated that it had moved eastward approximately 20 inches (508 mm) relative to the pillar before falling to the north.[83]​.

David Kirkaldy") tested samples of the bridge materials, both cast and wrought iron, as well as various bolts, stays and associated lugs. Both the wrought iron and cast iron had good strength, while the bolts "were of sufficient strength and suitable iron". Gussets[80]​ and lugs were weakened due to high local stresses where the bolts pierced them.[77]​ Four of the fourteen lugs tested were not solidly fixed, having failed at lower than expected loads. Some lugs on the top of the pillars lasted longer than the wrought iron ones, but the lower lugs were significantly weaker.[85]​.

Opinions and analysis

Bouch had designed the bridge, aided in his calculations by Allan Stewart&action=edit&redlink=1 "Allan Stewart (engineer) (not yet redacted)").[note 21]​ After the accident, Stewart had assisted William Pole")[note 22]​ in order to determine what the bridge should have resisted.[note 23]​ Under Stewart's authority, it was assumed that the bridge would be designed to resist a wind load of 20 pounds per square foot (1 kPa).

He was referring to advice given by Astronomer Royal, George Biddell Airy in 1873 when consulted on Bouch's design for a suspension bridge across the Firth of Forth; considering that wind pressures as high as 40 psf (1.9 kPa) could be found very locally, but averaging over a 1,600 ft (487.7 m) interval 10 psf (0.5 kPa) would be a reasonable tolerance.[91] This advice had been endorsed by several eminent engineers.[note 25] Bouch also mentioned the advice given by Yolland in 1869: that the Board of Trade did not require no special allowance for wind load for spans less than 200 feet (61.0 m), although he noted that this criterion was applicable for the design of beams, but not columns.[90][note 26]​.

Data were requested from scientists on the current state of knowledge on wind loading and from engineers on its application. Airy said the advice given was specific to suspension bridges and the Forth; stated that it was possible for a load of 40 psf (1.9 kPa) to act on an entire span of the Tay Bridge and would now recommend designing it for 120 psf (5.7 kPa) (i.e. 1.4 kPa with the usual safety margin).[91] The highest pressure measured at the Greenwich observatory "Greenwich (district)") had been 50 psf (2.4 kPa); it would probably rise more in Scotland.

Sir George Stokes agreed with Airy that 'cat's paws', ripples in water produced by gusts, could be several hundred yards wide. Standard wind pressure measurements were hydrostatic pressure, which had to be corrected by a factor of 1.4-2 to obtain the total wind load; with a wind of 60 miles per hour (96.6 km/h), this would be 12.5 to 18 psf (0.6 to 0.9 kPa).[93] Pole referred to the work of Smeaton, where it was said that strong winds could generate on the order of 10 psf (0.5 kPa), with higher values recommended for winds of 50 mph (80.5 km/h) or greater, with the except that these values were less secure.[94]​.

Brunlees had not taken into account the wind load on the Solway Viaduct" because the spans were short and low; if he had had to, he would probably have designed for 30 psf (1.4 kPa) with a safety margin of 4–5 (limiting the strength of the iron).[89] Both Pole and Law had used data from a book by Rankine.[note 27] Law agreed with Rankine that the highest wind pressure seen at Britain gave 55 psf (2.6 kPa) as the reason for designing at 200 psf (9.6 kPa) (i.e. 50 psf (2.4 kPa) with a safety factor of 4); It is not advisable to speculate as to whether this is a fair estimate or not."[95] Pole had ignored this because no reference was given; he did not believe that any engineers paid attention to it when designing bridges,[96] and he thought that 20 psf (1 kPa) was a reasonable allowance, considering that this value was what Robert Stephenson had assumed for the Britannia Bridge. Benjamin Baker "Benjamin Baker (engineer)") said he would design for 28 psf (1.3 kPa) with a safety margin, but in 15 years of observations he had not yet seen the wind knock down a structure that would resist 20 psf (1 kPa). readings.[97]​.

Findings

The three members of the tribunal did not reach an agreement to write a single report, although there were many points in common:[131]​.

• - Neither the foundation nor the beams had defects.

• - The quality of the wrought iron, although not the best, was not a decisive factor.

• - Cast iron was also pretty good, but had some molding problems.

• - The workmanship and assembly of the pillars were defective in many aspects.

• - The transverse reinforcement of the pillars and their fixings were too weak to withstand the strong gales. Rothery complained that the transverse reinforcement was not as strong or as well fitted as on the Belah Viaduct;[132] Yolland and Barlow stated that the weight/cost ratio of the transverse reinforcement was a disproportionately small fraction of the total weight/cost of the iron pieces.[133]

• - There was insufficiently strict supervision of the Wormit foundry (a large apparent reduction in strength in the cast iron was attributed to the fixings exerting pressure on the edges of the lugs, rather than acting orthogonally on them).[133]​.

• - Supervision of the bridge after its completion was unsatisfactory; Noble had no hardware experience nor any definite instruction to report on the condition of the hardware.

• - However, Noble should have reported the loose link pieces.[note 31]​ The use of covering pieces could have reinforced the pillars in a distorted way.

• - The 25 mile per hour (40.2 km/h) limit had not been enforced, and was frequently exceeded.

Rothery added that, given the importance to the bridge's design of test holes showing shallow bedrock, Bouch should have tried harder and examined the samples himself.[134]

According to Yolland and Barlow "the fall of the bridge was caused by the insufficiency of the crossbars and fixings to sustain the force of the gale on the night of December 28, 1879... the bridge had previously been stressed by other gales."[135]

Rothery agreed, asking "Can there be any doubt that what caused the bridge to collapse was wind pressure acting on a poorly constructed and poorly maintained structure?"[134]

Substantial differences between reports

Yolland and Barlow also noted the possibility that the failure was due to the fracture of a lee column. Rothery considered that the previous overloads affecting the bridge were caused "partly by previous gales, partly by the high speed at which northbound trains were allowed to pass over the high girders":[134] if the momentum of a train at 25 miles per hour (40.2 km/h) hitting the girders could cause the bridge trusses to fall, what must have been the cumulative effect of the trains' repeated braking at 40 miles per hour (64.4 km/h) at the north end of the bridge?[136] He therefore concluded, supported by circumstantial evidence, that the bridge could well have failed at the north end first;[137] and explicitly rejected the claim that the train had struck the girders before it fell. the bridge.[137]​.

Yolland and Barlow concluded that the bridge had failed first at the south end; and they did not express any explicit opinion as to whether the train had struck the girders or not.[135]​ Instead, they noted that, apart from Bouch himself, Bouch's own witnesses stated/admitted that the failure of the bridge was due to a shock load on the lugs highly stressed by wind loading.[138]​ Their report is therefore consistent with the view that the train had not struck the girder or with the view that a bridge with transverse reinforcement providing an adequate safety margin against wind loading it would have survived a train hitting the beam.

Yolland and Barlow noted that "there is no requirement issued by the Board of Trade with respect to wind pressure, and there appears to be no understood rule in the engineering profession with respect to wind pressure on railway structures; we therefore recommend that the Board of Trade should take such steps as may be necessary for the establishment of rules for that purpose."[139] Rothery dissented, opining that the engineers themselves should arrive at an 'understood rule', such as the French rule. of 55 psf (2.6 kPa)[note 32]​ or the 50 psf (2.4 kPa) of the United States.[141]​.

Rothery's minority report is more detailed in its analysis, more willing to place blame on named individuals, and more citable, but the court's official report is relatively brief and signed by Yolland and Barlow.[142] Rothery said that his colleagues had refused to join him in assigning blame, arguing that this was outside their terms of reference. However, previous Section 7 investigations had clearly felt free to blame (Thorpe railway accident")) or exonerate (Shippeton-on-Cherwell railway accident")) identifiable persons as they saw fit, and when Bouch's solicitor consulted with Yolland and Barlow, both refused to agree with Rothery in stating that "for these defects both in design, construction and maintenance, Sir Thomas Bouch is, in our opinion, the principal guilty".[143]​.

Section 7 Inquiries

No further judicial inquiries were carried out under Section 7 of the Railway Regulation Act 1871 until the Hixon railway accident in 1968 called into question both the Railway Inspectorate's policy regarding automated level crossings and the Ministry of Transport's (the Inspectorate's parent government department) management of the transfer of abnormal loads. A Section 7 judicial inquiry was deemed necessary to obtain the required degree of independence.[144]​ The structure and terms of reference were better defined than for the Tay Bridge inquiry. Brian Gibbens, QC, was supported by two expert advisers and made findings on fault/responsibility, but not liability/fault.[145]​.

Wind Pressure Commission (Railway Structures)

The Board of Trade established a commission of 5 people (Barlow, Yolland, Sir John Hawkshaw, Sir William George Armstrong and Stokes) to consider what wind load should be assumed when designing railway bridges.

Wind speeds were usually measured in 'miles traveled in an hour' (i.e. the wind speed averaged over an hour), so it was difficult to apply Smeaton's table[146]​ which related the pressure generated by the wind to its instantaneous speed.

where:.

By examining wind pressures and speeds recorded at Bidston Observatory), the commission found[147]​ that for strong winds the highest wind pressure could be represented very fairly,[note 33]​ by.

where:.

However, they recommended that the structures be designed to withstand a wind load of 56 psf (2.7 kPa), with a safety factor of 4 (2 when the bridge's resistance was based on gravity). They noted that higher wind pressures had been recorded at Bidston Observatory, but would still give loads within the recommended safety margins. The wind pressures reported at Bidston were probably abnormally high due to the peculiarities of the site (one of the highest points on the Wirral.[149][150]​): a wind pressure of 30 to 40 psf (1.4 to 1.9 kPa) would overturn railway carriages and such events were a rarity. To give a later, well-documented example, in 1903 a parked train was overturned on Levens Viaduct, but this was by a 'terrible gale' measured at Barrow-in-Furness at an average speed of 100 miles per hour (160.9 km/h), with estimated gusts of 120 miles per hour (193.1 km/h).[148]

Bridges

The Northern British Railway constructed a new double-track Tay railway bridge, designed by Barlow and built by William Arrol & Co.") of Glasgow, situated some 18 meters (19.7 yd) upstream and parallel to the original bridge. Work began on 6 July 1883 and the bridge opened on 13 July 1887. Sir John Fowler and Sir Benjamin Baker "Benjamin Baker (engineer)") designed the Forth Bridge, built (also by Arrols) between 1883 and 1890. Baker and his colleague Allan Stewart received most credit for the design and supervision of the construction work.[note 34]​ The Forth Bridge had a speed limit of 40 mph, which was not always properly observed.[152]​.

Bouch had also been an engineer for the British North, Arbroath and Montrose Railway, which included an iron viaduct over the South Esk. Closely examined after the collapse of the Tay Bridge, the constructed viaduct was inconsistent with the project, and many of the piers were noticeably outside design perpendicularity. It was suspected that the construction had not been adequately supervised: the foundation piles had not been driven sufficiently deep or firmly. Tests carried out in 1880 over a period of 36 hours using both deadweight and rolling stock loads led to the conclusion that the structure was seriously shifting and eight of the piers were declared unsafe. Condemning the structure, Colonel Yolland also expressed his opinion that "pillars constructed with cast iron columns of the dimensions used in this viaduct should not be approved in the future by the Board of Trade".[155] It had to be dismantled and rebuilt by Sir William Arrol, adopting a design by W. R. Galbraith before the line could be opened to traffic in 1881. wind loads of 19 psf (0.9 kPa).[158]​.

Memories of the accident

The locomotive, the NBR no. 224"), a 4-4-0 designed by Thomas Wheatley&action=edit&redlink=1 "Thomas Wheatley (engineer) (not yet drafted)") and built at Cowlairs Works") in 1871, was salvaged and repaired, remaining in service until 1919, with the nickname "The Diver." Many superstitious train drivers were reluctant to cross the new bridge.[159]​[160]​[161]​[162]​ The stumps of the pillars of the original bridge are still visible on the surface of the Tay. Monuments have been placed at each end of the bridge in Dundee and Wormit.[163]​.

A column from the bridge is on display at the Dundee Transport Museum.

On 28 December 2019, Dundee Walterfronts Walks organized a walk of remembrance to commemorate the 140th anniversary of the Tay Bridge disaster.[164]​.

Various additional evidence has been presented over the past 40 years, leading to "forensic engineering" reinterpretations of what really happened.[165][166]​.

• - Wikisource contains original works from or about Tay Bridge Disaster.

The disaster inspired several songs and poems, most famously written by William McGonagall. The poem, titled The Tay Bridge Disaster"), is widely considered to be of such low quality that it is even comical.[167] German poet Theodor Fontane, shocked by the news, wrote his poem Die Brück' am Tay.[168][169] It was published just ten days after the tragedy occurred. In turn, C. Horne's ballad "In Memory of the Tay Bridge Disaster" was published published as a broadside&action=edit&redlink=1 "Broadside (music) (not yet written)") in May 1880. He describes the time of the disaster as:[170]​.

• - David Kirkaldy").

• - Harry Watts").

• - Annex: Structural failures and collapses").

• - Annex: Disasters on bridges").

• - Annex: Railway accidents related to wind").

• - Lewis, Peter R. Beautiful Railway Bridge of the Silvery Tay: Reinvestigating the Tay Bridge Disaster of 1879, Tempus, 2004, ISBN 0-7524-3160-9.

• - Lewis, Peter R. Disaster on the Dee: Robert Stephenson's Nemesis of 1847, Tempus Publishing (2007) ISBN 978-0-7524-4266-2.

• - McKean, Charles, Battle for the North: The Tay and Forth Bridges and the 19th-Century Railway Wars: The Building of the Tay and Forth Bridges and the 19th Century Railway Wars Granta 2007.

• - Prebble, John"), The High Girders: The Story of the Tay Bridge Disaster, 1956 (published by Penguin Books in 1975) ISBN 0-14-004590-2.

• - Rapley, John Thomas Bouch: the Builder of the Tay Bridge, Stroud: Tempus, 2006, ISBN 0-7524-3695-3.

• - Rothery, Henry Tay Bridge Disaster: Report Of The Court of Inquiry, and Report Of Mr. Rothery, Upon the Circumstances Attending the Fall of a Portion of the Tay Bridge on the 28th December 1879. London: Her Majesty's Stationery Office"), 1880 OCLC 30875567.

• - Swinfen, David The Fall of the Tay Bridge, Mercat Press, 1998, ISBN 1-873644-34-5.

• - Thomas, John The Tay Bridge Disaster: New Light on the 1879 Tragedy, David & Charles, 1972, ISBN 0-7153-5198-2.

• - Wikimedia Commons hosts a multimedia category on Tay Bridge Disaster.

• - 91 black-and-white photographs of the wrecked piers of the Tay Bridge showing destroyed piers and girders, wreckage of train and steam engine from National Library of Scotland.

• - Tom Martin's engineering analysis of the bridge disaster.

• - Reappraisal of the Tay Bridge disaster The Open University.

• - The Tay Bridge Disaster at Failure Magazine").

• - Dundee local history center page on the disaster.

• - Tay Victims listing {reference only}.

• - Find a grave memorial of Tay River victims.

• - Firth of Tay Bridge Disaster 1879: Worst Structural Disaster in British History at Suburban Emergency Management Project.

• - Tay Bridge Disaster: Appendix to the Report Of The Court of Inquiry. Includes a large number of drawings of the bridge, and calculations of the result of wind pressure on the structure.

• - Report from the Select Committee on the North British Railway (Tay Bridge) Bill; together with the Proceedings of the Committee and Mins of Ev. All the oral evidence given, reproduced verbatim – a very large file but sometimes a useful corrective to reinterpretation by secondary sources.

• - The Tay Bridge Collection=%27MS%2030%27)) at.

• - Was Disaster Built into the First Tay Bridge? Article relating to the University of Dundee's holdings on the disaster.

On the night of Sunday, December 28, 1879, a violent storm (10 to 11 on the Beaufort Scale) was blowing at virtually right angles to the bridge. Witnesses stated that the storm was as strong as the worst they had seen in the 20-30 years they had lived in the area, and one described it as a 'hurricane', as bad as a typhoon he had seen in the area. China Sea.[10] Wind speed was measured in Glasgow - 71 mph (114 km/h; 32 m/s) (one-hour average) - and Aberdeen, but not in Dundee.

Higher wind speeds were recorded at shorter intervals, but at the inquest an expert witness warned of their unreliability and refused to estimate conditions at Dundee from readings taken elsewhere.[11] A modern interpretation of the available information suggests that the winds blew at 80 mph (129 km/h; 36 m/s).[12]

Use of the bridge was restricted to one train at a time by a "Block (Railway)") blocking system that used a baton as a token, which was picked up at one collateral signal cabin and delivered at the other. At 7:13 p.m. m. A train to and from Burntisland[13]​ (consisting of a type 4-4-0 locomotive, its tender, five passenger cars,[note 1]​ and a luggage van[14]​) slowed to collect the baton from the signal cabin at the south end of the bridge, and then headed towards the viaduct, gaining speed.

The signalman turned to note the passing of the train and then attended to the fire in the cabin, but a friend of his observed the train, and when he was about 200 yards (182.9 m) from the cabin, he saw sparks coming from the wheels on the east side, which he had also seen on the previous train.[15] During the investigation, John Black testified that the wind pushed the cars so that the flanges of the wheels made contact with the rails. Black explained that the internal protective barriers against derailment were slightly higher than the railings.[15]​[note 2]​ This arrangement would hold the good wheel when the derailment was caused by the disintegration of one of the two wheels on an axle, a real risk before steel wheels became widespread, a situation that had occurred in the Shipton-on-Cherwell railway accident) on Christmas Eve 1874.

The sparks continued for no more than three minutes, at which time the train was traveling along the high beams.[16]​ At that point "there was a sudden flash of bright light, and in an instant the darkness was total, the rear lights of the train, the sparks and the flash of light... disappeared at the same instant."[17]​ The signalman saw none of this and did not believe it when told.[note 3]​ When the train did not appear on the alignment leaving the bridge towards Dundee, he attempted to speak to the signal box at the north end of the bridge, but discovered that all communication had been lost.[18]​.

Not only was the train in the river, but so were the tall girders and much of the iron in its supporting pillars.[19] Divers who explored the wreckage later found the train still inside the girders, with the engine inside the fifth span of the 5-span southern division.[20] There were no survivors; Only 46 bodies were recovered[21]​ but there were 59 known victims. Fifty-six tickets to Dundee had been collected from the train passengers before crossing the bridge. Taking into account season ticket holders, cross-ticket tickets and railway employees, it was thought that 74 or 75 people were on the train.[17] It has been suggested that there were no unknown victims and that the higher figure of 75 arises from a double count in an initial newspaper report,[22] but the inquiry did not take its casualty figures from the "Dundee Courier", and took sworn testimony and made its own counts.

• - The locomotive fell again during the rescue, but was eventually recovered and running again.

• - The left front of the recovered locomotive tender.

• - Remains of two rescued passenger cars.

• - Fallen beams, with remains of a wooden passenger car.

Evidence

A judicial court of inquiry was immediately established (under Section 7 of the Railway Regulation Act 1871") "into the causes and circumstances occasioning" the accident): Henry Cadogan Rothery"), Commissioner of Shipwrecks, President, supported by Colonel Yolland") (Inspector of Railways) and William Henry Barlow, President of the Institution of Civil Engineers"). On January 3, 1880, they were gathering evidence in Dundee; They then appointed Henry Law (a qualified civil engineer) to carry out detailed investigations. While awaiting their report, they held further hearings in Dundee (26 February to 3 March). Once the collection of evidence was complete, they settled in Westminster (19 April to 8 May) to consider the engineering aspects of the collapse. redacted)"),[24]​ both engineers with extensive experience in major cast iron structures). The terms of reference did not specify the underlying purpose of the inquiry: to avoid a repetition, to assign blame, to clarify responsibility or culpability, or to establish precisely what had happened. This led to difficulties (culminating in confrontations) during the Westminster sessions. When the court reported its findings in late June, there was both an Inquiry Report signed by Barlow and Yolland, and a report drawn up by Rothery.

Two witnesses, who were looking at the high beams from the north almost straight on, had seen the lights of the train as far as the third and fourth high beams, when they disappeared; this was followed by three flashes in the high beams to the north of the train. One witness said that the flashes advanced towards the north end of the high beams with approximately 15 seconds between the first and the last;[25]​[note 4]​ while the other stated that they all occurred at the north end, with less time in between.[26]​ A third witness had seen "a mass of fire fall from the bridge" at the north end of the high beams.[27]​ A fourth said he had seen a beam fall into the river at the north end of the beams. high, then a light appeared briefly in the tall southern beams, disappearing as another beam fell; He did not mention the fire or the flashes.[28]​[note 5]​

'Ex-Provost' Robertson[note 6]​ had a good view of most of the bridge from his home in Newport-on-Tay,[31]​ but other buildings blocked his view of the tall southern girders. He had seen the train moving toward the bridge; then, on the high northern girders, before the train could reach them, he saw "two columns of steam illuminated with the light, first one flash and then another" and could no longer see the lights on the bridge;

Former Mayor Robertson had purchased a season ticket between Dundee and Newport in early November and became concerned about the speed of northbound local trains across the high beams, which had been causing noticeable vibrations, both vertical and lateral. After complaining three times to the station master in Dundee, his complaints did not affect the speed of the trains. Starting in mid-December he had used his season ticket to travel only south, using the ferry for northbound crossings.

Robertson had timed the train with his pocket watch, and to give the railroad the benefit of the doubt he had rounded to the nearest five seconds. The time measured across the beams (3,149 feet (960 m)) was normally 65 or 60 seconds,[note 8] but twice it had been 50 seconds. Watching from the shore, he had measured 80 seconds for trains traveling across the bridge, but not any he had ridden on. Local trains heading north often stopped to avoid express delays, then made up time as they traveled over the bridge. The gradient towards the bridge at the north end prevented similar high speeds on southbound local trains. Robertson said the movement he observed was difficult to quantify, although the lateral movement, which was probably 1 to 2 inches (25.4 to 50.8 mm), was definitely due to the bridge, not the train, and the effect was most marked at high speed.

Four other passengers on the train confirmed Robertson's stated times, but only one had noticed any movement of the bridge.[35][note 9] The Dundee station master had relayed Robertson's complaint about speed (he had been unaware of any concerns about sway) to the conductors, and then checked the times from cab to cab (at either end of the bridge the train was traveling slowly to pick up or deliver the witness). However, I had never verified the speed through the high beams.[37]​.

Painters who had worked on the bridge in mid-1879 said that it shook when a train passed.[38][note 10] When a train entered the high southern girders, the bridge shook at the north end, both from east to west and, more strongly, up and down.[41] The shaking was worse when the trains went faster, which they did: "when the Fife ship was very close and the train had only reached the south end of the bridge, "[42] A carpenter who had worked on the bridge from May to October 1879 also spoke of a lateral shaking, which was more alarming than the up and down movement, and greatest at the southern junction between the high beams and the low beams. He was unwilling to quantify the range of motion, but when pressed he gave a figure of about 2 to 3 inches (50.8 to 76.2 mm). When pressed again, he would only say that the movement was distinct, wide and noticeable.[43] One of the painters' foremen, however, said that the only movement he had seen had been from north to south, and that it had been less than 1/2 of an inch (12.7 mm).[44]

The British Northern Railway was responsible for maintaining the tracks, but relied on Bouch to oversee the maintenance of the bridge, and appointed Henry Noble as its bridge inspector.[45] Noble, who was a bricklayer, not an engineer, had worked for Bouch on the construction of the bridge.[46]

While checking the foundations of the piles to see if the riverbed was being cleared around them, Noble had noticed that some diagonal tie bars were 'chattering',[note 11]​ and in October 1878 he had begun to remedy this problem. Diagonal bracing was provided by flat bars running from a lug at the top of one column section to two sling plates bolted to a lug at the base of the equivalent section on an adjacent column. The sling bar and plates had a matching longitudinal groove. The tie bar was placed between the sling plates with the three slots aligned and overlapping, and then a swivel was inserted through the three slots and secured. Two "cotters" (metal wedges) were then placed[note 12]​ to fill the remainder of the slot overlap and driven in tightly to tighten the tie.

Noble had assumed that the cotter pins were too small and had not initially been inserted with enough force, but in the vibrating flanges the cotter pins were loose, and even if driven all the way in, they did not fill the slots to put the bars in tension. By placing an additional piece of casing between the loose cotter pins and inserting the cotter pins, Noble re-tightened the loose ties, which stopped rattling. There were more than 4,000 key joints on the bridge, but Noble said only about 100 had needed to be retensioned, most in October-November 1878. At his last inspection in December 1879, only two fastenings had needed attention, both on the piers north of the high beams. Noble had found cracks in four column sections, one below the tall beams and three north of them, which had been tied together with wrought iron hoops. Noble had consulted Bouch about the cracked columns, but not about the tableting of the diagonal pieces.[48]​.

Wormit foundry workers complained that the columns had been cast with 'Cleveland iron', which always contained slag; It was less easy to cast than 'good Scotch metal'[49][note 13] and was more likely to produce defective castings. The molds were moistened with salt water, the cores[50]​ were improperly clamped and moved, resulting in uneven wall thickness of the castings.[51]​ The foundry foreman explained that where the lugs had been cast imperfectly; the missing metal was added "by burning".[note 14]​ If a piece had cavities or other casting defects considered minor failures, they were filled with 'Beaumont egg',[note 15]​ a material of which the foreman had stocks for that purpose.[55]​.

The Gilkes production plant staff were inherited from the previous contractor. Under the resident engineer were seven subordinates, including a foundry manager. The original foundry manager left before most of the tall beam pillar sections were cast. His replacement also supervised the construction of the bridge and had no prior experience supervising casting work.[56] He was aware that they were repairing defects with hot casting,[57] but the foreman had concealed the use of the Beaumont egg from him.[58] When shown defects in the bridge castings, he said he would not have approved the filled pier parts for use, nor would he have passed profiles with wall thicknesses noticeably uneven.[56]​ According to his predecessor, the hot filling had only been carried out on the temporary 'lifting piers' pieces, which were used to lift the truss girders into place and were not part of the permanent structure of the bridge.[59]​ The instructions of the resident engineer were followed,[60]​ who was also not very experienced in casting techniques and trusted the foreman.[61]​.

While work practices were Gilkes' responsibility, his contract with the railroad stipulated that all work performed by the contractor was subject to Bouch's approval of the workmanship. Bouch would therefore share the blame for any resulting defective work on the finished bridge. The foreman of the original foundry, who had been fired for his drunkenness, testified that Gilkes personally approved of irregularities in the early foundries: "Mr. Gilkes, sometimes once a fortnight and sometimes once a month, would strike a profile with a hammer, first on one side and then on the other, and used to review most of the pieces by the way they sounded."[62] Bouch had spent over £9,000 on the inspection. (his total fee was £10,500)[63] but he did not present any witnesses who had inspected the castings on his behalf. Bouch himself had gotten up once a week while the design was being changed, but "later, when construction started, he didn't go as often."[64]

Bouch maintained his own "resident engineer", William Paterson, who looked after the construction of the bridge, its approaches, the line to Leuchars and the branch to Newport, who was also the engineer of Perth General Station.[64] Bouch told the court that Paterson's age was "very close to my own" but, in fact, Paterson was 12 years older than Bouch[note 16] and, at the time of the investigation, he was suffering from a paralysis and could not give evidence.[66] Another inspector appointed later[66]​ was then in South Australia and was also unable to give evidence. Gilkes' managers could not answer for any inspection of the castings by Bouch's inspectors.[67] The finished bridge had been inspected on Bouch's behalf to check the quality of assembly, but that was after the bridge had been painted (although still before the bridge was opened, and before the witness painters were on it in the summer of 1879), which hid any cracks or signs of hot repair (although the inspector said that, in In any case, he would not recognize those signs with the naked eye).[68] Throughout the construction, Noble had taken care of the foundations and masonry.[note 17]​.

Henry Law, who had examined the remains of the bridge, reported defects in workmanship and in the details of the design. Cochrane and Brunlees, who gave evidence later, largely agreed.

• - The pillars had not moved or settled, but the masonry of the bases showed little adhesion between the stone and the cement: the stone had been left too smooth and had not been wet before adding the cement. The anchor bolts, with which the bases of the pillars were held, were poorly designed and were inserted into the masonry with a design that was too simple, without sufficient grip.[70]​.

• - The "Flange (pipes)") connecting flanges between the column sections were not completely facing each other (that is, they had not been machined to obtain smooth, flat surfaces that fit perfectly together). The dowels that should have provided continuity from one section to the next were not always present,[note 18]​ and the bolts did not fill the corresponding holes. Consequently, the only thing that resisted the sliding of one flange over another was the tightening action of the bolts.[72] This tightening effect was reduced because the heads of the bolts and rivets were not thick enough; some had burrs") of up to 0.05 inches (1.3 mm) (he presented an example). This undermined any fastening effect, since using rivets in the joint at the base of a pillar and subsequently riveting them would leave more than 2 inches (50.8 mm) of free play at the top of the pillar. The riveted pieces used were abnormally short and thin.[73]​.

• - The pillar profiles had unequal wall thickness, as much as 1/2 inch (12.7 mm) difference; sometimes because the core had moved during casting, sometimes because the two halves of the mold were misaligned. Thin-walled metal parts were undesirable, both because of their lower intrinsic strength capacity and because (since they cooled more quickly) they would be more vulnerable to the generation of defects.

Here "(providing a sample)" is a nodule of cold metal that has formed. The metal, as expected on the thin part, is very imperfect. Here there is a defect that extends through the thickness of the metal. Here is another and here is another... The entire upper side of this pillar will be found to fit this description, completely filled with hollows and ashes. There are enough pieces here to demonstrate that these defects were widespread.[74]​

Bouch said that the uneven thickness was due to incorrect workmanship, and that if he had known this, he would have taken steps to mold the pieces vertically, but he still considered them safe.[75]

• - The horizontal iron braces of the Canal did not abut correctly with the body of the pillars; correct spacing depended on the bolts being properly tightened (previous comments about lack of coating applied here as well). Because the holes in the lugs were not pre-drilled, their position was only approximate and some horizontal braces had been field-installed, leaving burrs of up to 3/16 of an inch (4.8 mm).[74]

• - In diagonal bracing, the keys were roughly forged and left uncoated, and were too small to withstand the compressive force that the bracing bars could exert.[note 19]​.

• - In the southernmost fallen pillar, a piece of covering had been placed on each link bar to the base.[76]​.

• - The lug screw holes were cast with a conical shape; Consequently, contact between the bolt and the lug was made by the thread of the bolt bearing against the sharp edge at the outer end of the hole. The shoulder was easily crushed and allowed play to develop, and the off-center loading caused the lugs to fail at much lower loads than if the hole were cylindrical.[77] Cochrane added that the bolts would permanently bend (loosening the link of the tie bars to the point that they had to be reinforced by cover pieces) at a load even lower than that at which the cotter pins would deform; He had found some bent tie rod bolts as apparent confirmation.[78]

• - The bracing had failed by giving way to the lugs; In almost all cases, the fractures passed through the holes. Law had seen no evidence of hot-repaired lugs,[77] but some lug failures with fragments torn off around the rivets indicated the use of hot-repaired parts. Furthermore, the paint on the intact pillars would hide any evidence of hot repairs.[79]​.

• - On some pillars, the base sections were still standing; in others, sections of the base had fallen to the west.[80] Cochrane noted that some fallen girders lay on top of the eastern pillars, but the western pillars were on top of the beams; Therefore, the engineers[80]​[81][82]​ agreed that the bridge had broken before it fell, not as a result of its fall.

• - Marks on the south end of the southernmost tall beam indicated that it had moved eastward approximately 20 inches (508 mm) relative to the pillar before falling to the north.[83]​.

David Kirkaldy") tested samples of the bridge materials, both cast and wrought iron, as well as various bolts, stays and associated lugs. Both the wrought iron and cast iron had good strength, while the bolts "were of sufficient strength and suitable iron". Gussets[80]​ and lugs were weakened due to high local stresses where the bolts pierced them.[77]​ Four of the fourteen lugs tested were not solidly fixed, having failed at lower than expected loads. Some lugs on the top of the pillars lasted longer than the wrought iron ones, but the lower lugs were significantly weaker.[85]​.

Opinions and analysis

Bouch had designed the bridge, aided in his calculations by Allan Stewart&action=edit&redlink=1 "Allan Stewart (engineer) (not yet redacted)").[note 21]​ After the accident, Stewart had assisted William Pole")[note 22]​ in order to determine what the bridge should have resisted.[note 23]​ Under Stewart's authority, it was assumed that the bridge would be designed to resist a wind load of 20 pounds per square foot (1 kPa).

He was referring to advice given by Astronomer Royal, George Biddell Airy in 1873 when consulted on Bouch's design for a suspension bridge across the Firth of Forth; considering that wind pressures as high as 40 psf (1.9 kPa) could be found very locally, but averaging over a 1,600 ft (487.7 m) interval 10 psf (0.5 kPa) would be a reasonable tolerance.[91] This advice had been endorsed by several eminent engineers.[note 25] Bouch also mentioned the advice given by Yolland in 1869: that the Board of Trade did not require no special allowance for wind load for spans less than 200 feet (61.0 m), although he noted that this criterion was applicable for the design of beams, but not columns.[90][note 26]​.

Data were requested from scientists on the current state of knowledge on wind loading and from engineers on its application. Airy said the advice given was specific to suspension bridges and the Forth; stated that it was possible for a load of 40 psf (1.9 kPa) to act on an entire span of the Tay Bridge and would now recommend designing it for 120 psf (5.7 kPa) (i.e. 1.4 kPa with the usual safety margin).[91] The highest pressure measured at the Greenwich observatory "Greenwich (district)") had been 50 psf (2.4 kPa); it would probably rise more in Scotland.

Sir George Stokes agreed with Airy that 'cat's paws', ripples in water produced by gusts, could be several hundred yards wide. Standard wind pressure measurements were hydrostatic pressure, which had to be corrected by a factor of 1.4-2 to obtain the total wind load; with a wind of 60 miles per hour (96.6 km/h), this would be 12.5 to 18 psf (0.6 to 0.9 kPa).[93] Pole referred to the work of Smeaton, where it was said that strong winds could generate on the order of 10 psf (0.5 kPa), with higher values recommended for winds of 50 mph (80.5 km/h) or greater, with the except that these values were less secure.[94]​.

Brunlees had not taken into account the wind load on the Solway Viaduct" because the spans were short and low; if he had had to, he would probably have designed for 30 psf (1.4 kPa) with a safety margin of 4–5 (limiting the strength of the iron).[89] Both Pole and Law had used data from a book by Rankine.[note 27] Law agreed with Rankine that the highest wind pressure seen at Britain gave 55 psf (2.6 kPa) as the reason for designing at 200 psf (9.6 kPa) (i.e. 50 psf (2.4 kPa) with a safety factor of 4); It is not advisable to speculate as to whether this is a fair estimate or not."[95] Pole had ignored this because no reference was given; he did not believe that any engineers paid attention to it when designing bridges,[96] and he thought that 20 psf (1 kPa) was a reasonable allowance, considering that this value was what Robert Stephenson had assumed for the Britannia Bridge. Benjamin Baker "Benjamin Baker (engineer)") said he would design for 28 psf (1.3 kPa) with a safety margin, but in 15 years of observations he had not yet seen the wind knock down a structure that would resist 20 psf (1 kPa). readings.[97]​.

Findings

The three members of the tribunal did not reach an agreement to write a single report, although there were many points in common:[131]​.

• - Neither the foundation nor the beams had defects.

• - The quality of the wrought iron, although not the best, was not a decisive factor.

• - Cast iron was also pretty good, but had some molding problems.

• - The workmanship and assembly of the pillars were defective in many aspects.

• - The transverse reinforcement of the pillars and their fixings were too weak to withstand the strong gales. Rothery complained that the transverse reinforcement was not as strong or as well fitted as on the Belah Viaduct;[132] Yolland and Barlow stated that the weight/cost ratio of the transverse reinforcement was a disproportionately small fraction of the total weight/cost of the iron pieces.[133]

• - There was insufficiently strict supervision of the Wormit foundry (a large apparent reduction in strength in the cast iron was attributed to the fixings exerting pressure on the edges of the lugs, rather than acting orthogonally on them).[133]​.

• - Supervision of the bridge after its completion was unsatisfactory; Noble had no hardware experience nor any definite instruction to report on the condition of the hardware.

• - However, Noble should have reported the loose link pieces.[note 31]​ The use of covering pieces could have reinforced the pillars in a distorted way.

• - The 25 mile per hour (40.2 km/h) limit had not been enforced, and was frequently exceeded.

Rothery added that, given the importance to the bridge's design of test holes showing shallow bedrock, Bouch should have tried harder and examined the samples himself.[134]

According to Yolland and Barlow "the fall of the bridge was caused by the insufficiency of the crossbars and fixings to sustain the force of the gale on the night of December 28, 1879... the bridge had previously been stressed by other gales."[135]

Rothery agreed, asking "Can there be any doubt that what caused the bridge to collapse was wind pressure acting on a poorly constructed and poorly maintained structure?"[134]

Substantial differences between reports

Yolland and Barlow also noted the possibility that the failure was due to the fracture of a lee column. Rothery considered that the previous overloads affecting the bridge were caused "partly by previous gales, partly by the high speed at which northbound trains were allowed to pass over the high girders":[134] if the momentum of a train at 25 miles per hour (40.2 km/h) hitting the girders could cause the bridge trusses to fall, what must have been the cumulative effect of the trains' repeated braking at 40 miles per hour (64.4 km/h) at the north end of the bridge?[136] He therefore concluded, supported by circumstantial evidence, that the bridge could well have failed at the north end first;[137] and explicitly rejected the claim that the train had struck the girders before it fell. the bridge.[137]​.

Yolland and Barlow concluded that the bridge had failed first at the south end; and they did not express any explicit opinion as to whether the train had struck the girders or not.[135]​ Instead, they noted that, apart from Bouch himself, Bouch's own witnesses stated/admitted that the failure of the bridge was due to a shock load on the lugs highly stressed by wind loading.[138]​ Their report is therefore consistent with the view that the train had not struck the girder or with the view that a bridge with transverse reinforcement providing an adequate safety margin against wind loading it would have survived a train hitting the beam.

Yolland and Barlow noted that "there is no requirement issued by the Board of Trade with respect to wind pressure, and there appears to be no understood rule in the engineering profession with respect to wind pressure on railway structures; we therefore recommend that the Board of Trade should take such steps as may be necessary for the establishment of rules for that purpose."[139] Rothery dissented, opining that the engineers themselves should arrive at an 'understood rule', such as the French rule. of 55 psf (2.6 kPa)[note 32]​ or the 50 psf (2.4 kPa) of the United States.[141]​.

Rothery's minority report is more detailed in its analysis, more willing to place blame on named individuals, and more citable, but the court's official report is relatively brief and signed by Yolland and Barlow.[142] Rothery said that his colleagues had refused to join him in assigning blame, arguing that this was outside their terms of reference. However, previous Section 7 investigations had clearly felt free to blame (Thorpe railway accident")) or exonerate (Shippeton-on-Cherwell railway accident")) identifiable persons as they saw fit, and when Bouch's solicitor consulted with Yolland and Barlow, both refused to agree with Rothery in stating that "for these defects both in design, construction and maintenance, Sir Thomas Bouch is, in our opinion, the principal guilty".[143]​.

Section 7 Inquiries

No further judicial inquiries were carried out under Section 7 of the Railway Regulation Act 1871 until the Hixon railway accident in 1968 called into question both the Railway Inspectorate's policy regarding automated level crossings and the Ministry of Transport's (the Inspectorate's parent government department) management of the transfer of abnormal loads. A Section 7 judicial inquiry was deemed necessary to obtain the required degree of independence.[144]​ The structure and terms of reference were better defined than for the Tay Bridge inquiry. Brian Gibbens, QC, was supported by two expert advisers and made findings on fault/responsibility, but not liability/fault.[145]​.

Wind Pressure Commission (Railway Structures)

The Board of Trade established a commission of 5 people (Barlow, Yolland, Sir John Hawkshaw, Sir William George Armstrong and Stokes) to consider what wind load should be assumed when designing railway bridges.

Wind speeds were usually measured in 'miles traveled in an hour' (i.e. the wind speed averaged over an hour), so it was difficult to apply Smeaton's table[146]​ which related the pressure generated by the wind to its instantaneous speed.

where:.

By examining wind pressures and speeds recorded at Bidston Observatory), the commission found[147]​ that for strong winds the highest wind pressure could be represented very fairly,[note 33]​ by.

where:.

However, they recommended that the structures be designed to withstand a wind load of 56 psf (2.7 kPa), with a safety factor of 4 (2 when the bridge's resistance was based on gravity). They noted that higher wind pressures had been recorded at Bidston Observatory, but would still give loads within the recommended safety margins. The wind pressures reported at Bidston were probably abnormally high due to the peculiarities of the site (one of the highest points on the Wirral.[149][150]​): a wind pressure of 30 to 40 psf (1.4 to 1.9 kPa) would overturn railway carriages and such events were a rarity. To give a later, well-documented example, in 1903 a parked train was overturned on Levens Viaduct, but this was by a 'terrible gale' measured at Barrow-in-Furness at an average speed of 100 miles per hour (160.9 km/h), with estimated gusts of 120 miles per hour (193.1 km/h).[148]

Bridges

The Northern British Railway constructed a new double-track Tay railway bridge, designed by Barlow and built by William Arrol & Co.") of Glasgow, situated some 18 meters (19.7 yd) upstream and parallel to the original bridge. Work began on 6 July 1883 and the bridge opened on 13 July 1887. Sir John Fowler and Sir Benjamin Baker "Benjamin Baker (engineer)") designed the Forth Bridge, built (also by Arrols) between 1883 and 1890. Baker and his colleague Allan Stewart received most credit for the design and supervision of the construction work.[note 34]​ The Forth Bridge had a speed limit of 40 mph, which was not always properly observed.[152]​.

Bouch had also been an engineer for the British North, Arbroath and Montrose Railway, which included an iron viaduct over the South Esk. Closely examined after the collapse of the Tay Bridge, the constructed viaduct was inconsistent with the project, and many of the piers were noticeably outside design perpendicularity. It was suspected that the construction had not been adequately supervised: the foundation piles had not been driven sufficiently deep or firmly. Tests carried out in 1880 over a period of 36 hours using both deadweight and rolling stock loads led to the conclusion that the structure was seriously shifting and eight of the piers were declared unsafe. Condemning the structure, Colonel Yolland also expressed his opinion that "pillars constructed with cast iron columns of the dimensions used in this viaduct should not be approved in the future by the Board of Trade".[155] It had to be dismantled and rebuilt by Sir William Arrol, adopting a design by W. R. Galbraith before the line could be opened to traffic in 1881. wind loads of 19 psf (0.9 kPa).[158]​.

Memories of the accident

The locomotive, the NBR no. 224"), a 4-4-0 designed by Thomas Wheatley&action=edit&redlink=1 "Thomas Wheatley (engineer) (not yet drafted)") and built at Cowlairs Works") in 1871, was salvaged and repaired, remaining in service until 1919, with the nickname "The Diver." Many superstitious train drivers were reluctant to cross the new bridge.[159]​[160]​[161]​[162]​ The stumps of the pillars of the original bridge are still visible on the surface of the Tay. Monuments have been placed at each end of the bridge in Dundee and Wormit.[163]​.

A column from the bridge is on display at the Dundee Transport Museum.

On 28 December 2019, Dundee Walterfronts Walks organized a walk of remembrance to commemorate the 140th anniversary of the Tay Bridge disaster.[164]​.

Various additional evidence has been presented over the past 40 years, leading to "forensic engineering" reinterpretations of what really happened.[165][166]​.

• - Wikisource contains original works from or about Tay Bridge Disaster.

The disaster inspired several songs and poems, most famously written by William McGonagall. The poem, titled The Tay Bridge Disaster"), is widely considered to be of such low quality that it is even comical.[167] German poet Theodor Fontane, shocked by the news, wrote his poem Die Brück' am Tay.[168][169] It was published just ten days after the tragedy occurred. In turn, C. Horne's ballad "In Memory of the Tay Bridge Disaster" was published published as a broadside&action=edit&redlink=1 "Broadside (music) (not yet written)") in May 1880. He describes the time of the disaster as:[170]​.

• - David Kirkaldy").

• - Harry Watts").

• - Annex: Structural failures and collapses").

• - Annex: Disasters on bridges").

• - Annex: Railway accidents related to wind").

• - Lewis, Peter R. Beautiful Railway Bridge of the Silvery Tay: Reinvestigating the Tay Bridge Disaster of 1879, Tempus, 2004, ISBN 0-7524-3160-9.

• - Lewis, Peter R. Disaster on the Dee: Robert Stephenson's Nemesis of 1847, Tempus Publishing (2007) ISBN 978-0-7524-4266-2.

• - McKean, Charles, Battle for the North: The Tay and Forth Bridges and the 19th-Century Railway Wars: The Building of the Tay and Forth Bridges and the 19th Century Railway Wars Granta 2007.

• - Prebble, John"), The High Girders: The Story of the Tay Bridge Disaster, 1956 (published by Penguin Books in 1975) ISBN 0-14-004590-2.

• - Rapley, John Thomas Bouch: the Builder of the Tay Bridge, Stroud: Tempus, 2006, ISBN 0-7524-3695-3.

• - Rothery, Henry Tay Bridge Disaster: Report Of The Court of Inquiry, and Report Of Mr. Rothery, Upon the Circumstances Attending the Fall of a Portion of the Tay Bridge on the 28th December 1879. London: Her Majesty's Stationery Office"), 1880 OCLC 30875567.

• - Swinfen, David The Fall of the Tay Bridge, Mercat Press, 1998, ISBN 1-873644-34-5.

• - Thomas, John The Tay Bridge Disaster: New Light on the 1879 Tragedy, David & Charles, 1972, ISBN 0-7153-5198-2.

• - Wikimedia Commons hosts a multimedia category on Tay Bridge Disaster.

• - 91 black-and-white photographs of the wrecked piers of the Tay Bridge showing destroyed piers and girders, wreckage of train and steam engine from National Library of Scotland.

• - Tom Martin's engineering analysis of the bridge disaster.

• - Reappraisal of the Tay Bridge disaster The Open University.

• - The Tay Bridge Disaster at Failure Magazine").

• - Dundee local history center page on the disaster.

• - Tay Victims listing {reference only}.

• - Find a grave memorial of Tay River victims.

• - Firth of Tay Bridge Disaster 1879: Worst Structural Disaster in British History at Suburban Emergency Management Project.

• - Tay Bridge Disaster: Appendix to the Report Of The Court of Inquiry. Includes a large number of drawings of the bridge, and calculations of the result of wind pressure on the structure.

• - Report from the Select Committee on the North British Railway (Tay Bridge) Bill; together with the Proceedings of the Committee and Mins of Ev. All the oral evidence given, reproduced verbatim – a very large file but sometimes a useful corrective to reinterpretation by secondary sources.

• - The Tay Bridge Collection=%27MS%2030%27)) at.

• - Was Disaster Built into the First Tay Bridge? Article relating to the University of Dundee's holdings on the disaster.