REPORT ON THE ACCIDENT TO THE GREY GORGE SUSPENSION BRIDGE.

Appendix to the Journals of the House of Representatives, 1876 Session I, E-02

E.—2

1876. NEW ZEALAND.

REPORT ON THE ACCIDENT TO THE GREY GORGE SUSPENSION BRIDGE.

Presented to both Houses of the General Assembly by Command of His Excellency.

No. 1. The Hon. the Minister for Public Works to the Commissioners of Audit. (No. 3130.) Public Works Office, Gentlemen, — Wellington, Bth August, 1876. The Government, being desirous to have a very strict and exhaustive inquiry into the cause of the accident which recently occurred to the Suspension Bridge over the Gorge of the River Grey, will be much obliged if you will undertake such an inquiry. The Engineer-in-Chief has been informed that this request has been made to you, and he will place such information as you may require from the Public Works Department at your disposal; and you will be good enough to take such other evidence as you may consider necessary to enable you to arrive at a thoroughly exhaustive conclusion. As the question has already been before the House of Representatives, and the Government have promised it further information, I have to request that you will hold the inquiry at as early a date as may be convenient. The services of a shorthand writer will, if possible to be procured, be placed at your disposal. I have, &c, J. E. FitzGerald, Esq., C.M.G., Edward Richardson. Dr. Knight, Commissioners of Audit, Wellington.

No. 2. To the Hon. the Minister for Public Works. Sir,— Audit Office, 9th September, 1876. In reference to the inquiry, which you did us the honor to request that we would make into the cause of the recent accident to the Suspension Bridge over the Grey River at Brunnerton, we beg to submit the following report: — The bridge which is the subject of this inquiry was erected over the Grey River at the Brunner Gorge, for the purpose of enabling coal trucks to be brought from the coal mine on the north side to the railway on the south side of the river. It was not designed to bear a locomotive engine; and it was not contemplated that more than 10 tons would ordinarily be placed on it at once, although calculated to bear a much heavier strain. I—E. 2.

2

E.—2

The general structure of the bridge may be described as follows : —Piers were erected at either side of the river, at a distance of 300 feet apart, over which two chains were stretched, and carried by tunnels into the ground at either side. These tunnels were to be 35 feet in length; but on the south side were extended to upwards of 70 feet, in order to reach the solid rock. Shafts were sunk from the surface, meeting the ends of the tunnels, in order to provide access to the apparatus for anchoring the chains. Each chain was to be composed of seven twisted wire ropes, 4f-inches in circumference, laid side by side, and above them six other ropes, each made of thirty telegraph wires spliced together but not twisted, placed side by side, the whole united every 10 feet by clips, forming a flat chain of nearly 12 inches in width by 2 inches in depth. Of the fourteen twisted wire ropes used, however, ten were 5-|-inch and four were 4^-inch. The chains were not parallel, but curved outwards at the ends, being 8 feet apart—the breadth of the roadway—in the middle; 30 feet apart at the piers ; and about 65 feet apart at the bottom of the tunnels, where fastened to the ground. The bridge itself consisted of two trusses of 300 feet in length and 10 feet in height, with a wooden roadway 8 feet wide between them, carrying a railway in middle. The whole suspended from the ropes by iron rods bolted through cross beams underlying the trusses. The chains were fixed to the ground as follows :—A solid mass of masonry, of about 7 feet by 6 feet, was built into each wall of the tunnel, the inner face being brought to a sloping plane perpendicular to the strain of the chain. Against the inner edge of each wall was placed a cast-iron girder the whole height of the masonry; and across these girders, and crossing the opening of the tunnel, was a cast-iron beam, the anchor, to which the ropes were fastened. Each anchor was 4 feet 6 inches in length, 12 inches in width, and 5 inches in depth, with a web at the back of 3 inches in width, rising by a curve from the ends to 5 inches in depth in the middle ; so that the middle section of the anchor showed a flange of 5 by 12, with a web sby 3 inches. The unsupported space between the bearings on the girders was 3 feet 6 inches. At either side of the central web the anchor was perforated by holes —on the lower side seven in number, on the upper side six, each 2 inches in diameter. The several ropes, of which the whole chain was formed, were fastened to screw bolts 2 inches in diameter, passing through the holes in the anchor, and screwed up by nuts at the back; the seven lower bolts taking the twisted wire ropes, and the six upper ones the telegraph wire ropes. The strain of each chain was thus distributed over a space of about 2 feet in the middle of the 3^ feet between the bearings. It is obvious that the perforations of the anchor by so many holes greatly reduced its strength; but for this allowance has been made in the calculations. It was intended to build up the tunnels, in front of the anchors and surrounding the chains, with solid concrete, at least to the depth of the masonry, 6 feet; and there is no question that, had this been done, the whole apparatus for securing the ropes would have been more than sufficiently strong to support any weight which could have been brought upon the bridge. The Engineer-in-Chief stated to us that his intention was that this should be done as soon as the ropes were in position; and that he gave verbal instructions to that effect. Such instructions, however, were not in writing, and we notice that neither in the specifications nor in the plans is there any allusion to this filling up with concrete. On the other hand, we ascertained from the District Engineer, that as the contractor was required to bring the chains and bridge into a specific form, it was desirable, if not necessary, to retain the power of regulating the strain on the chains, by means of the screws, after the roadway was completed, which could not have been done had the chains been buried in concrete. Mr. Carruthers, however, is of opinion that no adjustment was required after the chains were once brought to a proper strain; and thought that the District Engineer understood that the concrete was to be built in as soon as the chains were in position.

3

E.—2

The facts as to the breaking of the bridge are described in the accompanying Report, Appendix A, from the District Engineer. As the accident occurred at 4 o'clock in the morning, these facts are necessarily meagre. All that can be positively known must be derived from the subsequent condition of the materials. The anchors on the south side broke; which broke first cannot now be ascertained ; the other followed immediately. Both broke in the same relative places, about a foot from the outside bearing relatively to the line of the bridge; that is to say, the eastern anchor about a foot from the eastern end, and the western anchor the same distance from the western end of the anchor. This is not theoretically the weakest point of the beam; but it is apparent that in both, the fracture occurred at the point at which two holes are most nearly opposite one another. This is clearly shown in the plan attached to Mr. O'Connor's report. The only question remains to assign a cause for the accident. It appears from the calculation of the Engineer-in-Chief (Appendix B) that the anchor ought to have borne a strain of 396 tons, if equally distributed, equivalent to 198 tons at the middle. Prom the reports of the superintendents of the two principal iron-foundries in Wellington, who have examined the fractured anchors, and whose reports are appended (Appendix C), each anchor is estimated to bear a breaking strain, according to one of these gentlemen, of 139 tons 16 lbs., and by the other, of 141 tons 18 lbs., a very nearly similar result. A variety of formulae are used by engineers for calculating the breaking weight which cast-iron beams Avill bear. We have tested the calculations by several of these formulse, and arc satisfied that with ordinarily good iron, such as alone should have been used for the purpose, the estimated strain of 140 tons is within that which each anchor should have borne, applied at the middle. It appears, hoAvever, from the calculations furnished by the Engineer-in-Chief (Appendix B), and from Mr. O'Connor's report, that the actual strain on each anchor could not at the moment of fracture have exceeded 75 tons distributed, equal to a strain of 56 tons in the middle; and in this estimate no notice is taken of the fact that the trusses of the bridge rested at each side on the solid ground; and although they were not quite equal to support themselves for the whole length of the bridge unaided, still, to whatever extent they may be regarded as rigid, they must to that extent have relieved the chains of some part of the weight of the bridge. The extent of this relief cannot be readily calculated. Mr. Carruthers states that he has roughly calculated that as much as 50 tons of the whole weight of the bridge, which was about 82 tons (without the chains), must have been borne by the trusses. No such calculation could be relied on as exact. It is sufficient, however, to show that a very appreciable part of the weight was not on the chains, and the result is that the real strain on each chain cannot have exceeded 50 tons on the middle of the anchor-plate, giving a factor of safety of 36 according to Mr, Carruthers' calculation, and more than 2*B according to the estimate of Mr. Robertson and Mr. Seager. We attribute the accident to two causes : First, tne misunderstanding as to the filling in of the concrete in front of the anchor-plates as soon as the chains had been brought to an even strain. Secondly, the defect in the anchor-plates. We do not give much importance to the faults stated to exist in the castings. Mr. Dobson detected a considerable fault, but his examination was rather hurried, as he was just leaving for Melbourne; and when the rust was removed from the surface of the fracture by washing, it appeared that he was deceived in the appearance of an air bubble. Mr. O'Connor speaks of a crack of some depth, but was unable to say how far it extended. On the whole, it may be admitted, the casting was a tolerably fair one; but the iron used was of weak quality, inferior to that which ought to have been put into a beam to be submitted to a high strain. Again, the mould was in a wrong position, the top of the casting being the part to be brought under tension, and the bottom under compression. A casual inspection of the two fragments submitted herewith will show the difference in the texture of the iron at the top and bottom of the casting. The part in tension should have been at the bottom. The casting was also undoubtedly cooled too quickly, and was in natural tension before any weight was brought upon it.

E.—2

4

Again, the anchor-plates differ from the drawings in two particulars: First, the holes in the original drawings are in no case opposite one another on the two sides of the weh; hut in two cases in the casting these holes are nearly opposite ; and it was just at these points that the anchors gave way. Secondly, the holes were designed to be directly in the line of the strain, and perpendicular to the face of the flange ; but some of them, especially where the fracture occurred, are a good deal out of the true line; and as the holts filled the holes, an appreciable cross strain was brought on the metal, tending to throw it still further into tension. We do not consider that any responsibility can be thrown on the contractors. There are no provisions in the contract as to the quality of metal to be used, or that any test should be applied to prove their strength. They were to be made to the satisfaction of the Engineer; and they were accepted as sufficient after examination; nor, if the intention of the Chief Engineer had been carried out, and the anchor-plates supported by the concrete or masonry, so that these plates would be only doing the work of washers, would blame attach anywhere. The accident must be attributed solely to the misunderstanding of the instructions given as to the time Avhen the tunnels were to be built up. But if it had been contemplated that the anchor-plates, on which the whole stability of the structure rested, until backed by the concrete, were to bear even for a short time the whole strain, we should consider that insufficient care had been taken to secure the strength of those beams. Little appears to have been done to secure good castings. A large casting, when cold, has its particles left in a state of permanent tension, so that even a sharp frost may shatter a very granular mass of cast iron. Engineers in England increase the toughness of the metal by addition of scraps of malleable iron. The practice recommended in England to secure a good mixture of iron, in contracting for a number of girders, is to stipulate that they should not break with less than a certain weight, leaving the mixture to the founder. The Engineer may then select one to be broken, and if it break with less weight than that agreed upon, the whole may be rejected. This test could not conveniently be used in the case of these massive anchor-plates; but the Resident Engineer might have been instructed to see that the pig, when put into the cupola, was mixed with a fixed proportion of malleable iron. He should also have taken care that the pattern was in accordance with the design, and that the mould was in the proper position. We have the honor to be, Sir, Your most obedient Servants, Jambs Edward EitzGerald. Charles C. Knigiit.

5

E.—2

MINUTES OF EVIDENCE.

Tuesday, 18th, August 1876. Me. Dobson examined. 1. You are a civil engineer ?—Yes. 2. Have you been accustomed to cast-iron works ?—Yes ; from 1846 to 1850, I had a great deal to do with cast iron. 3. Have you of late had any experience in cast-iron works ? —Yes, occasionally. 4. Have you seen the broken castings of the anchors of the bridge at the Grey Eiver Gorge ?— Yes; I examined them two hours ago. 5. What opinion should you give as to the character of the iron ?—lt is very weak iron; not positively bad, but very weak. 6. Did you find any flaw in the castings ?—A flaw existed in one only,-not in both. 7. Do you think that such a fault would be of any importance in a casting ? —lt would much weaken the casting. 8. Where was this fault? —In the lower part of the flange, the part in tension. 9. Did you measure the extent of it ?—No, but it was about 1| inch. The fact of there being an air bubble at all would throw very much doubt as to the casting. It simply shows that it was not a sound casting. 10. You think there is no douht that it was an air bubble ? —Not the slightest; you cannot mistake an air bubble. 11. Would the casting have been made from one furnace, or would it require more ?—Not more than one in a large foundry.

Tuesday, 29th August, 1876. Present: Me. J. E. FitzGeeald and De. Knight, Commissioners. Me. C. T. O'Connob examined. 12. Mr. FitzGerald.] "What office do you hold, Mr. O'Connor? —I am District Engineer on the "West Coast. 13. And what was Mr. McArthur ?—Mr. McArthur was the Resident Engineer at Greymouth. 14. Who had charge of the erection of this bridge ?—Mr. McArthur had charge of the erection of the bridge under my supervision. 15. Was any of the superstructure of the bridge commenced when you took charge ? —No. 16. Were the castings of the anchors made ? —No. 17. Did you see the castings made ? —No ; It would have been Mr. McArthur's duty to see the castings made. [The Commissioners requested Mr. O'Connor to give them an account in writing of the facts connected with the breaking of the bridge. Mr. O'Connor undertook to supply an account.] 18. At what intervals did you inspect the work in progress?— Generally once a month; sometimes oftener. 19. When did you last see the bridge before it gave way?— About a fortnight before. 20. Where does Mr. McArthur reside ?—At Greymouth, not at Brunnerton. 1 21. How long was it before the accident that Mr. McArthur saw the bridge ? —Well, I think it was probably nearly a month since he saw it officially. He saw it, I dare say, about a week before; but as he was then released from duty preparatory to his leaving, he could not be said to have inspected it officially. 22. You say he saw it probably about a week before ? —I think he saw it about a week before, but he did not go up then with a view of inspecting it. 23. It would be the same thing if he was in charge of the work ? —He was hardly in charge, for he was only called upon to do one special piece of work, and was going to leave. The understanding was that he was to devote himself to one special piece of work, and as soon as that was done he was to go away. He saw it officially about a month before the accident. 24. Who was the resident officer in charge of the work ? —The Inspector, Mr. Olliver. 25. Was he resident at Brunnerton ?—Yes; he was living there, and constantly in charge. 26. Did Mr. McArthur ever make any remarks to you as to the safety of the bridge ?—No. 27. Had you any reason to doubt its safety ?—No ; I had no reason to doubt its safety. I made the remark that it looked rather small, but the remark was not made iv any way as to doubting its safety. 28. Dr. KniglttJ] Was that in reference to the future arrangements for filling up the ashlar work with concrete ? Did you think that when the ashlar work was so filled up, the bridge would be of undoubted strength ? Is that what you mean ?—There is no doubt it would have been strong enough, but it was at that time that it was finally determined to fill it up with concrete work at once, when it was finished. 29. Was that to be done in consequence of your having a doubt as to its safety ?—Not in consequence of that. I simply remarked that the anchor-plates looked rather small; then the Engineer-in-Chief asked about the filling up, and he desired it to be filled up.

E.—2.

6

30. It had not been finally determined before that to fill up the ashlar work with concrete ?—lt had not been communicated to me before; but there were reasons to suppose that it was so determined when it was designed. 31. I understood you to say that it was then finally determined to fill it up ?—I mean to say as far as I was concerned. It is very hard for me to give evidence as to that point, because I was not there at the starting of the bridge contract. 32. Mr ]?itzQerald.~\ Are you cognizant of the calculations Mr. Carruthers has given us with regard to the safety of the anchor-plates ?—Yes. 33. You are aware that the factor of its safety in these calculations is represented to be 3 decimal 6 ?—Three something ; something over 3. 34. In calculating that factor of safety, has any account been taken of the rolling weight ?—No ; only the weight of the bridge itself. 35. In calculating the safety of a bridge, would you not calculate upon a factor of safety of 5 or 6 for the rolling weight, in addition to the factor of safety for the weight of the bridge ?—Yes; I should probably multiply the rolling weight by 2, and then take 3 for the whole load, which would amount to the same thing. 36. You would include an additional strain for the rolling stock, would you not ?—Oh, yes, in ordinary cases. 37. In all cases would you not ?—One would in all cases calculate for the weight of the rolling stock. » 38. If you found that Ao calculation had been made for the weight of the rolling stock, and only a calculation for the bare weight of the bridge, would you not infer that some additional strength was intended besides the anchor-plates ?—lf I had ever calculated them, I should ; but, unfortunately, I never did. 39. My question was this, Mr. O'Connor: that assuming that these calculations show an insufficient strength for the weight of the bridge and rolling stock combined, would you not infer by that that it had been part of the original design to strengthen the anchor-plates by concrete, or in some other manner ? Would you not infer that now ?—Oh, yes. 40. I understand that, as the calculations come out, it is admitted that the anchor-plates were not sufficient to carry the bridge and rolling stock with safety ? —Yes, that is admitted. 41. Did you consider it any part of your duty to calculate the strength of the anchors, and to make any representations to the Chief Engineer on the subject before they were put up ? —I did not do so. 42. I presume you did not think it your duty to do so ?—I do not know. I suppose it was my duty to have done it. If I had been in the immediate charge of that work, I certainly should have considered it my duty to do it; but I had so very many works to look after, that I had hardly time to go into the details of every one. 43. The drawings were supplied to you ?—The contract was let before I took charge. 41. Is there anything mentioned in the contract or specification as to filling up with concrete ?— No, there is not. 45. Then the contractor was not bound under the contract to fill it up ?—No, he was not. 46. Dr. Knight.'] Had you the materials i'or making the concrete there ?—We had the cement. It was principally to consist of cement. 47. Mr. FitzGerald.] Had you sufficient cement?— Yes. 48. Dr. Knight.] Did you use cement in any part of the structure ?—We had been using it, and it was kept over for the purpose. The contractor had a balance over, and kept it there, anticipating that it would be required for this filling. Of course, we should have had to pay him extra for it if he had done it. 49. Did the Chief Engineer instruct you to fill up with concrete as soon as the chains were in position ? —I did not understand it so. 50. When was it to be filled up ? —I understood it was to be filled up as soon as the bridge was completed. SL. Mr. FitzGerald.] Would it have been convenient to have filled it up before?—lf it had been filled up before, the effect of the screws would have been lost; that is to say, the power of using the screws in future would have been done away with; but, beyond that, no inconvenience would have resulted. 52. Dr. Knight.] You could not tell, I suppose, to what extent the chains would deflect ? —The difficulty was to tell how far the ropes, made in different ways, would be affected by the strain, as compared with one another. Some were twisted wire ropes and some were telegraph wire ropes. . 53. Mr. FitzGerald.] Speaking as an engineer, you think it was necessary to keep the ropes free until the final strains were adjusted?—l thought so ; and that was the basis I acted on in omitting the masonry until the bridge was completed. 54. Dr. Knight.] Would it have been safe to have put rolling stock on the bridge before the concrete had been put in ? —The fact of its breaking before it was put on shows that it would not have been safe. 55. In your opinion, would the bridge have been safe if the rolling weight had been put upon it before the concrete had been put in ? —Well, I can hardly answer that question, and leave out of consideration the fact that the bridge came down. Taking that fact into the question, it is evident it would not have been safe practically. 56. Mr. FitzGerald.] Yes; but theoretically, supposing the materials had been good, according to calculations, would the bridge have been safe ? —-Theoretically these anchor-plates ought to have borne the weight of the bridge and its load too, but they would hardly have contained the ordinary factor of safety. 57. What weight,practically, was the bridge intended to bear? I understand that the bridge was not on a level ?—lt was not. There was a considerable gradient for a railway—viz., lin 100.

7

E.—2.

58. "Would you explain how it was intended to work the traffic. It is a very important point?— The bridge was designed with a gradient of 1 in 100, sloping downwards from the mine towards the railway station, and there was a turntable designed at the mine end, just close up to the bridge. The ground was too precipitous at that side to admit of approaching the bridge by a curve, so that all the wagons would have had to be run along a line at right angles with the bridge, and put on to this turntable, and run across the bridge. I conceived, therefore, that there never could come upon the bridge a load of more than one full wagon at a time. 59. So that, in the practical working of the bridge, there would have been only one wagon on it at a time ?—Tes ; one full wagon. 60. What would be its weight ?—Ten tons in the centre—equal to a distributive weight of 20 tons. It would have been equal to an additional 20 tons distributed over the length of the bridge. 61. It was equal to 10 tons at one point?—Tes, in the centre, for of course it would sometimes have been in the centre. 62. One wagon would not have carried 10 tons ?—No ; it would only have carried 6 tons. "We call them roughly 10 tons. They are 3 tons each in weight, and the load 6 tons, but it might reach up to 7. 63. Dr. Kniglit.~\ And the load would pass over the bridge by its own weight ? —Yes; by gravitation. 64. How would the wagons be brought back ? —-By horses. 65. Would not a great many wagons bo brought back at once ?—I do not see that there would be much gained by bringing more than three. A horse could not draw more than three up that incline, and walking on smooth planks. The Engineer-in-Chief has calculated for a much greater weight than, that; but I never could see how a greater weight could in practice come upon it. 66. It is calculated for 30 or 40 tons ?—For 40 tons. 67. Have you computed the strain that would be upon each chain, arising out of the permanent weight of the bridge and the rolling weight ? —Tes ; I have computed it. But these calculations of Mr. Carruthera give it all; only they take 40 tons as the load instead of 10. He has provided for the most remote contingency; but in practice no such weight would come. 68. What about public holidays ?—That is quite another thing. There his calculations and mine agree. It is only taking so much per foot for an immense crowd ; though Ido not think that in practice it would occur. It is not necessary it should occur. 69. But as to the strength of the bridge, you would not have made it sufficient merely to carry a contemplated load of one wagon ? You think it necessary it should be of sufficient strength to bear a crowd such as on public holidays; do you not ?—I have always considered that it would be a very great mistake in New Zealand to build all the bridges on the principle they do in London. Toil have usually got the power of regulating the traffic; and where there is no absolute necessity for a dense crowd being upon it, I do not see why a large expense should be incurred in providing for so remote a contingency. A railway bridge is under the complete authority of the railway officials, and they need not allow such a thing to occur as a dense crowd coming upon it. 70. Then, if it had not been designed strong enough to sustain a dense crowd, it would have been necessary to forbid more than a certain number of people coming upon it at one time ?—Tes; that would have been the effect. As a matter of fact, however, the cables of this bridge are strong enough to sustain a dense crowd. 71. Mr. FitzGerald.~\ This bridge was not a public highway ? —No. 72. Dr. Knight.] Taking Mr. Carruthers' estimate of 70 lbs. to the square foot, do you think that excessive? —I do not think it is if you want to estimate for a dense crowd. 73. Mr. FitzGerald.] Did you consider that the bridge not being a highway, it was not contemplated to have a crowd upon it ?—Tes, I did not think it was ever contemplated to have a crowd upon it. 74. Dr. Knight.] If there was a crowd upon it, 70 lbs. to the square foot would be a fair estimate of the load ? —Tes. 75. What do you consider would be the permanent load of the bridge when finished? How many tons? —Ninety-six tons. I am now speaking from memory. I made out that statement with Mr. Carruthers, and I am perfectly satisfied that it is correct. 76. Mr. FitzGerald.] That is what produces the strain of 75 tons? —Tes, 75 on each anchor. 77. Dr. Knight.] Then, you have no doubt the permanent load ought to be multiplied by a factor of safety of 3 ?—Tes. 78. Do you know that that is the factor used by the Board of Trade in England ?—lt depends on what the material is. 79. Speaking of this suspension bridge, do you know the factor of safety of 3 is the one requred by the Board of Trade in England?—l do not know that they have laid down any rule for suspension bridges. I have never come across one laid down by the Board of Trade in England. 80. Do you know what the regulation of the Board of Trade is with respect to the strain on iron generally ?—For wrought iron, 5 tons in tension and 4 tons in compression. 81. Whatever the load may be, what would be the factor of safety for the moving load, according to the regulations of the Board of Trade?—l do not know what regulation the Board of Trade lays down. I know that most hand-books on engineering lay it down that you should take a factor of 3 for the permanent load and 6 for the rolling load. They are, I think, taken so in that statement of Mr. Carruthers. 82. Dr. Knight.] It is 5 here ?—The moving load in that is taken much higher than it would ever be likely to be in practice. 83. I understand you to say that you agree with the computations made by Mr. Carruthers ?— Tes, I agree to them all, except that matter of the moving load, which I do not think would ever be so high as he makes it. 84. Have you made any computations as to the dimensions of the section of the chains in square inches ?—Tes; I went into the calculation of that, but I put it on one side, because I found that the

E.—2

8

results were higher than experiments proved that this particular class of ropes would justify, and then Mr. Carruthers and myself put it down out of the standard books, which gave the results on this particular sort of ropes, from some experiments by an American engineer, named Eoebling, made upon ropes of various diameters. 85. Do iron-wire ropes stand a greater strain per square inch than iron bars?—lron-wire rope will stand a higher strain than iron bars of the same diameter. At the same time, iron-wire rope would not stand the same strain as an equal diameter of rope made of wires all in direct tension. Assuming it as such would give too high a result. 86. Do you know the weight of the chain from pier to pier ? —Tes. 87. If you know the weight and length of it, you can tell what the section is in square inches?— One could deduce it, bearing in mind these experiments and everything upon it. 88. Do you remember the weight of the chains ?—I think the weight of the two chains is 14 tons —31,360 lbs. for the two chains. 89. How do you obtain the weight ?—We weighed each of the ropes. 90. Of a certain length ?—We weighed several feet of them. 91. And from that you computed the total weight of the chain between the piers?— Yes. 92. And that weight was how many tons ? —Fourteen tons nearly. 93. Is that for the two chains ?—Tes. 94. So that each chain weighed about 7 tons ? —Tes, about that. 95. From computations we have had made, we found the weight of each chain ought to be about 15,893 lbs. ?—I make it 15,680 lbs. 96. So that the actual weight agrees very nearly with the computations we made ?—Yes ; they are very close. 97. And that we found gave a sectional area of 16-78 square inches ?—The sectional area of the cable, as a whole, was 12 x 2, but of course there were small spaces between the ropes which would not count. 98. Mr. FitzQerald.~\ The computation of the strains on the cable takes into account the rolling weight ? —Tes. 99. In taking the strain on the chain, did you take into consideration not only the rolling weight and the permanent weight, but the weight of the chain itself?—Oh, certainly. 100. Do you know what the section of the suspension rods is ? —lt is an inch in diameter. 101. Dr. Knight I\ From computations we made, we found that the sectional area should have been 103 square inches, which would give a diameter of I'l4 square inches. 102. Mr. FitzGerald,'] What weight are your suspension rods subject to?— Between 4J and 4J tons. 103. That is, there should be a greater number of suspension rods ?—lt is the only way in which the weight on each could be reduced if it were necessary to reduce it. Four and a quarter tons is the weight that might occasionally come upon each suspension rod. 104. Dr. Knight.] Navier, a French engineer, is of opinion that a suspension rod, with a section of 1 inch square, should not be subjected to a heavier weight than 3 tons. 105. Mr. FitzOerald.] Is there a special modulus given in these tables of Roebling for wire rope ? —No; he gives simply the breaking strain of a given diameter of rope in a table as ascertained by experiments. 106. How many of these wire bridges are there constructed, to your knowledge, in New Zealand. You said there were a good many?—l spoke about Westland. There is one large bridge, consisting of one 100-feet span and two 60-feet spans over the Arahura, and there are a large number of foot bridges reaching up to 180-feet span. 107. Is there any other bridge built to your knowledge anchored in the same way as the Grey Gorge Bridge ?—No, I do not know of one exactly the same. 108. With iron anchors, I mean ? —Oh, there are bridges with iron anchors, but the exact method of attachment is not the same. 109. There are bridges with cast-iron anchor-plates —I mean in New Zealand ? —Not in New Zealand that I know of, but there are in England. 110. You constructed the bridge over the Arahura?—Yes. 111. Did you make the plans for it ?—Tes. 112. How is it anchored?—lt is anchored to piles driven into the ground. 113. Dr. Knight.'] What was the span of that bridge ? —There was one span of 100 feet, and the other two were 60 feet. 114. Speaking of iron, I understand that iron-wire rope is the strongest of all the forms in which iron is used in the construction of bridges ?—Tes. 115. Do you know what the comparative strength of iron-wire rope is when compared with bar iron, for instance? —About as 7to 5. The strength of wire is as 7t05 as compared with wrought iron. 116. Is it usual in the casting for anchor-plates to use common pig iron, without any mixture of better iron with it ? Do you know whether engineers generally take any trouble in the matter, and whether they are anxious to have a good sound casting ? —Of course they are anxious to have a good sound casting. I never had charge of a bridge of the same description before, where cast-iron anchor plates were used. I have seen cast iron used in many cases at Home. I have had charge of works where castings have been used, and no special steps have been taken to secure their being more than ordinary marketable iron. 117. Is it not the practice at Home, when an order is given for a number of cast-iron beams or bars, to make it a part of the contract that they should be strong enough to bear a certain amount of weight without fracture ?—I have never seen it done. 118. How do the engineers at Home secure a good casting? —It really depends, as a great mauv things of the sort depend, on the reputation of the person who supplies it. They go to a foundry where they have got a reputation at stake; they pay them a reasonable price, and they undertake to give a good article.

9

E.—2,

119. Then, in such a very important structure, which really depended on the strength of the anchor-plates, would you not have thought it extremely necessary that some special care should have been taken to seethat the iron was of first-rate quality for casting? —I do not know very much about cast iron, and I was obliged very much to put myself in the hands of the man who undertook the work. But I spoke to him about it, and he saw the necessity of making a good job, and undertook to do it. These men were endeavouring to establish them selves as foundrymen. They had just established a foundry there, and seemed quite as anxious as I could be myself to make the job a good one. 120. Only the ordinary materials were used ?—Yes; but they took the greatest possible trouble to make the moulds exact. I saw the models. I saw the models put in the sand. 121. Mr. JPitzGeraJd.'] But you did not see the mould, did you? —Yes, I did; that is to say, I saw what I mean by the mould, namely, the impression of the pattern in the sand, but I did not see the anchor-plates cast. 122. You saw, then, which way the anchor-plates were to be cast in the mould ? —Yes. 123. And you saw which side was to be uppermost ?—Yes ; I saw that. 124. Are you aware that the side of the iron in tension was the side uppermost in the mould ?— Yes. 125. Is it not usual to put the tension side downwards ?—lt would have been difficult to have done it in this case in that way, on account of the web. | 126. Is not the iron at the bottom of a casting more capable of standing tension than the iron at the top ?—I should think it ought to be unless the iron is perfectly homogenous all through. 127. Have you examined the fracture of this casting ? —Yes. 128. Is there no considerable difference in the texture of the iron on the two sides of the casting? [Two exhibits produced, showing fractures in anchor-plate.] —Yes ; there is no doubt there is a considerable difference, but I believe myself that you can make almost any fracture you like in cast iron by the way you make it. I have heard practical men say, too, that it depends on the way in which it is broken what fracture it will have. 129. Do not you see a brown slag in that smaller piece of iron ?—Yes ; no doubt about it. 130. Dr. KnigJit.~\ Do you know that scraps of iron are usually mixed with the pig iron to secure a tough and good casting?—No, I do not; that is to say, I do not know that it is done with that view. 131. Did I understand you to say just now that you did not understand it was the practice to add malleable iron to the common pig iron ? —I know they frequently add a certain proportion of scrap iron, but what proportion I am not prepared to say, nor can I tell the exact effect it produces. 132. Do you not think it an important thing to secure a good casting ?—Yes. 133. Did it occur to you at all that it was necessary to give instructions to add malleable iron to the pig iron?—l did not think it necessary to do so, because this man, who has been all his life at it, knew more than I did, and I trusted to him to turn out as good a job as he could. 134. Mr. FitzGerald.] Do you know, as a fact, whether any scrap iron was placed in this casting? —No ; not from my own knowledge. 135. Dr. Kniglit.~\ Could you learn what quantity of scrap iron is put in to secure a good casting ? I understand that even one-fifth is used. —I do not know that I could find out except by a reference to books. As to asking a foundryman, he will give you his own practice. 13G. Mr. MtzGerakl.] Is it not specified in the contracts how much scrap iron shall be put in ? —I do not remember having seen it in a specification. 137. Dr. KnigM.~\ You have never had your attention drawn to the necessity of adding malleable iron to the common pig iron ? —I have never had my attention drawn to the necessity of specifying it, for really the greatest difficulty is to turn them out. Frequently castings will go in the casting; they will flaw. If they manage to turn out the casting at all, it is considered to be pretty good. At least, that is one of the guarantees that are taken. 138. Referring to the tables given of the strength of iron for tension and compression, do you know what is the margin between the best iron and the worst ?—The worst would be about two-thirds. It is about as five to nine between the worst and the best. In the calculations for these anchor-plates the worst was assumed. It generally is so. Of course, if you choose to specify the best, you may possibly get it, but you must pay for it. It is generally cheaper to put up with moderately fair iron, and put it in in larger dimensions. Even if one were to specify the most perfect iron, very few engineers would consider themselves justified in reckoning on getting the best. 139. Dr. Kni(jht.~\ You could not be secure ?—No ; not in a case like that. The reputation of the ironfounder is a better guarantee than all the specifications you can make. In the case of the Waimea pipes, for instance, the firm gave a guarantee that they would turn out a good article, and this was a better guarantee than all the specifications you could have had. It would be impossible to guarantee against putting in bad stuff if they had chosen to do it. In that case, they undertook to supply pipes which would bear a pressure of 300 vertical feet of water, and they were tested. The engineer has of course to state the dimensions he requires, and he has to see that he gets them. Bub the practical work of the casting depends greatly upon the ironfounder. 140. I understand it is the usual custom at Homo to have one bar more cast than is needed, and select one and break it. If it breaks with too light a weight, then the whole of the bars are rejected. That is the way in which I understand they test them at Home.—l have heard of that being done, but I have never seen it done. I had charge of works at Home where fifty or sixty cast-iron girders, 40 feet long by 2 feet high, were used. The only guarantee was that the man who made them had his reputation at stake in the matter. 141. Ido not see how that applies in this case. It would have been very easy to see the materials put into the cupola? —That is true, if we had understood the thing better than the man who put them in; but a practical ironfounder really understands that sort of thing better than engineers, unless they have got a speciality for that kind of work. 142. Do you know whether it is usual to bake the mould so as to get it perfectly dry before the melted metal is poured into the mould? Do you know whether it is the custom to do that ?—I do not. 2—E. 2.

E.—2

10

143. Mr. FitzGerald. ] Have you any technical knowledge of the texture of iron? Looking at these specimens before you, what would you judge was the character of the iron?—l should judge the large sample to be a very fair sort of iron. 144. And the other? —That I should say is an indifferent sort of iron. I should not say it is very good, certainly. I must say that my opinion in the matter is a good deal formed upon the opinion of Mr. Eonayne about it. He is a mechanical engineer. He is in charge of the traffic, including engines, &c, at Greymouth. He is a mechanical engineer, and served his time in the Inchicore Works, in Dublin. 145. Has he examined these specimens ? —Yes ; I asked him particularly to examine them, and I got them chipped off in order that he might do it. He was quite unbiassed in the matter. He was in no way connected with the bridge or responsible for it; and he said his opinion was that it was fair iron. He would not call it bad iron. 146. Dr. Knight.'] When did he see it ? After the fracture ? —-Immediately after the fracture ; the day on which it occurred. 147. Mr. FitzGerald] Do you think that opinion consistent with the opinion that was given to us on examination, that the iron was a very fair specimen of very low-class iron ; that was the expression used? —T do not think that Mr. Eonayne's opinion of it would bear that interpretation at all. Speaking generally, he said that he had a very large experience in castings, and I asked him to give me his opinion upon those specimens. 148. We were told that iron was what is called " No. 2 Clyde pig " ? —There is 1, 2, and 3 pig, I think. 149. Two would be about the second best ? —Yes ; I suppose so. 150. With regard to that flaw which you thought you detected in the casting, have you examined it again since the fracture to see whether your opinion was correct on that point, that there was a flaw ?—I have not examined it since, but I am thoroughly convinced that there was a flaw. It extended about half an inch into the casting. 151. Did that flaw catch your attention before the anchor-plates were put up ?—No ; it would have been quite impossible to detect it then. 152. What do you think that flaw arose from ? —Well, it must have been what they call a blow. 153. Or an imperfect union of the two flows of iron ?—-It could hardly have been that, because it was so near the top. 154. That is just where it would occur if the iron was poured in from two different sides. Do you know whether the iron was poured in from two different sides ? —I do not know. The flaw was very visible in the place I knocked it off. It was the other end of the anchor-plates I knocked it off. There was quite a plain cleavage between the two. 155. Was it in a line with the fracture ? Did it break through the flaw ? —Yes. 156. Then the flaw did not appear at the surface ?—lt did not appear. 157. It was a kind of scale ?—Yes, but it was half an inch thick. 158. Dr. Knight.] Was that on both sides? —I imagine it travelled all the way across. It was black ; that drew our attention to it first. No doubt there was a good deal of cinder in it. 150. Mr FitzGerald.] Then tho effect of that would have been, when subjected to a high tension, that the piece would have scaled off ? —lf it had extended right across it would have scaled off. It was not altogether cinder. There was a considerable amount of adhesion in it. 160. Then the effect of it would have been equivalent to a smaller beam ?—lt would have been equivalent to another plate laid at the bottom, supposing it was continuous.

Wednesday, 30th August, 1876. Mr. Caeeuthees examined. 161. Dr. Knight.] Mr. Carruthers, you are the Engineer-in-Chief ? —Yes. 162. Are all important public works carried on under the control of the Engineer-in-Chief, and the plans and specifications drawn up under his direction and supervision ?—Yes. 163. In reference to the Brunuer Suspension Bridge, who prepared the plans and specifications ? —They were done under my directions. 164. You have laid before the Board plans and specifications and other documents necessary to enable us to take out the quantities of weights and materials used in the construction of the bridge ? —Yes. 165. "Who is the contractor for the building of the bridge ?—Mr. Garven. 166. When contracts are entered into, whose duty is it to see that the works under construction are superintended by competent engineers?— Mine. 167. Who is the Superintending Engineer for the Brunner Suspension Bridge ?—Mr. Charles Yelverton O'Connor is the District Engineer, having charge of works in the district; and Mr. McArthur was the Resident Engineer, having direct local charge under Mr. O'Connor. 168. And he superintended the work from its commencement?—l think not from the beginning. I think Mr. Geisow was there at the beginning. IG9. Why did not Mr. Geisow continue in charge of that work ? —I wished to re-arrange the department, and he was appointed to another district. 170. Were any instructions given for the guidance of the Superintending Engineer ?—No special instructions were given beyond those contained in the specifications and plans. 171. No written instructions ?—No. 172. When payments are made on account of the bridge, who certifies that the work in progress has been satisfactorily performed, and that the amount claimed, as progress money, is in accordance with the terms of the contract ?—Mr. O'Connor, the District Engineer. 173. Were all the materials for the bridge supplied by the contractor ?—All except the wire ropes.

11

E.—2.

174. Where were the anchor-beams cast?—ln Q-reymouth. 175. Are you satisfied that the fouudry at Greymouth could supply a good casting for the heavy castings required for the construction of the bridge ? —STes, I think they could. 176. The stability of the bridge more or less depended on the strength of the anchor-beams. If they were not sound and otherwise sufficient, the bridge must necessarily fall?— During construction the bridge was dependent on the strength of the anchor-beams, but after the completion of the bridge, or before the completion of the bridge, the anchor-beams would have beeu supported by masonry behind them, and would become of little importance. 177. How often have you inspected the works while in progress ? —I think twice. 178. How long is it since the last inspection by yourself"?t—l do not remember the date. It was at the time the anchor-plates were cast, but not in place. 179. Mr. FitzGerald.] Did you see them on that occasion?—l saw them. 180. Dr. Knight.'] Has any other suspension bridge been built under charge of the Public Works Department, by contract, and did the contractors supply all the materials?—No other suspension bridge has been built. 181. It is observed that the chains, after passing through the anchor-plates, are screwed up by nuts instead of being fixed by wedges. Is there any danger of the threads being stripped by the pull upon them, or are they likely \o be injured by the weather? —-There is no danger of the screws being stripped, as the screws are quite as strong as other parts of the rod, or stronger. 182. And would be more convenient, I suppose, in the construction of the bridge ? —Yes. 183. In determining the tension of the rods ?—Yes, more convenient than wedging. The first design was for wedging, but I thought the want of experience of the probable tenderers ivould make it difficult to carry out the plan, and I therefore changed it. 181. We have your estimate of the weight of the structure between the piers ? —Yes. 185. And the maximum load that the bridge was estimated to carry? —Yes. 186. What is the usual estimate per square foot of platform when it is not intended for a railway ? —It varies very much much—between 50 lbs. and 100 lbs. ; 50 lbs. is enough on anything but a crowded race-day or something of that sort, where a bridge is jammed as full as it can be packed. 187. Is the truss of suflicent strength to bear the bending action on it when the load is passing over it?—lt is; with the aid of the chains. 188. Would the truss be crushed in any way by a concentrated load? —No, it would not; it is strong enough to carry the weight of half the length of the bridge. 189. Is it, indeed?—No; it would hardly do that. It would, however, transfer the weight from one part of the chain to another, for a distance of at least 100 feet. L9O. Do I understand you to mean that it does not add really to the strength of the bridge, but distributes the weight in a very favourable manner ?—To distribute the weight is its principal duty, and to resist the winds, but it very materially strengthens the bridge, as it bears a great deal of its weight. 191. Mr. FitzGerald.] At 50lbs. per foot, I find the total weight on the bridge would be something under 60 tons ? —Yes. 192. In making your calculations, what did you calculate would be the rolling weight?— Equivalent to 30 tons, but it was nearly concentrated at the centre. 193. That would be something the same as 60 tons distributed? —Yes; about equal. There is no chance of a bridge up there being loaded with a dense crowd of people, because there is nothing to induce it. 194. We have understood that the bridge was not a highway ?—lt was not. , 195. Therefore the railway officials would have power to prevent an undue weight coming upon it ?—Yes ; but, in designing it, I took into consideration the fact that it might become crowded, and, therefore, made it strong enough to carry all the people that could be put on to it. 196. We understand, from the calculations you laid before the Commissioners, that you estimated a factor of safety of 3 decimal 6 for the weight of the bridge, without loading ? —The factor of safety in designing the bridge was taken at 3* for the stable load, and 5 for the rolling load. 197. Precisely ; that is the very point. From your calculations, it appears your factor of safety for the weight of the bridge, without taking into account the rolling load, is somewhere about 3} for the anchor-plates ?—The anchor-plates would have been supported by masonry. 198. I mean without the masonry ?—Well, there was no factor of safety taken, because the anchor-plates were intended to be supported by masonry. I have calculated the strength of the plates lately. 199. lam referring to these returns you put in. The factor of safety came out slightly different, but about 3J-. In these calculations the rolling load is not taken into consideration ? —No. 200. Then, if the rolling load had been taken into consideration, the beams ought to have been much larger?—lf they were not to be supported, they ought to be ; but as they were to be supported, it was not necessary to make them larger. 201. Is there any specification in the contract for the filling up of that masonry ?—No. 202. How was that intended to be done, then ? —That was an oversight in the drawings, which was remedied by my giving instructions to the Engineer to have it done. 203. Did you contemplate originally that the contractor should do that? —In the design I had intended that it should be done, but it was just overlooked in the drawings. 204. In the specifications is there any description as to how the anchor-plates should be cast ?—No. 205. What is the provision in the specifications for the character of the iron to be used ?—I do not remember. I think you have the specifications here. [Specifications handed to witness.] There is no special mention beyond that all material should be to the satisfaction of the Engineer. 206. Mr. FitzGerald.] We have sent a telegram to Mr. Me Arthur. [Telegram and reply read.] Now, what I want to ask you is this: There being nothing in the specifications as to the character of

E.—2

12

the iron, except that it should be satisfactory to the Engineer, do you not consider that it was his duty to see the metal that was going to be used in the construction of these anchor-plates ?—Well, after the event, one is apt to think he ought to have done it. 207. Have you seen the fractures yourself?—-Yes. 208. Are you skilled in the inspection of cast iron ?—Not very much. [Specimens produced.] 209. What is your opinion as to the character of the iron ?—I think it is a fair iron, of a weak <jualit} r. 210. Is it usual iv castings that are made to bear a heavy strain to incorporate portions of scrap iron ?—Very rarely ; only where there is a very special strain brought upon them. In cannon castings it sometimes is done. 211. Not for anchor-beams ?—No. 212. Is it not usual sometimes to specify that the iron shall be of a character to bear a certain weight to the square inch ? —Yes ; on large jobs where it can be done. 213. What would you take from the tables as the modulus of the cast iron ? Did you take the highest or the lowest ?—I took the lowest given in Rankine. 214. Did you consider that ought to be for very low iron ?—Not good iron, but a fair grey iron. 215. Looking at the iron which you inspected in the fracture, do yooi think it is iron which would come within the lowest iron given in Eankine for the modulus of rupture ? —Yes, I think it is as good as that. 216. Is it usual, when iron is to be exposed to tension and compression, to take care that the part brought into tension shall be the lowest part of the mould ?—lt ought to be. 217. Would you look at these specimens of iron. [Specimens shown to witness.] One is from the edge of the side in tension, and the other from the edge of the side in compression. Do you not see a great difference in the construction of these ?—There is a very great difference. 218. The one in tension is very much the weakest?— The one in tension is the weakest, and I think, from the difference in the quality, that the casting must have been under unequal internal strains. 219. Would that arise from unequal cooling ?—Yes. 220. From a dampness in the mould ?—No ; from being allowed to cool too quickly. 221. That would be what is expressed by one of the ironmasters in his letter to us, when he says he considers that the iron was in tension when it left the foundry ? —Yes. 222. Mr. O'Connor says, in his evidence, that it appeared from the fracture of the part we have not got, that there was a flaw of some magnitude in the iron, indicated by its internal blackness, showing there had been no joining in the middle. Have you traced anything of that in the specimens we have got ? —No ; it was too rusted to see anything of that sort. 223. Mr. O'Connor described to us a fracture of considerable extent and a half-inch in depth, and I think he drew our attention to it to show that it was a kind of flake, extending for some inches. He could not trace it into the part we have got, but said that iv the other part it was very visible, and that his intention was drawn to it by its being black. Would such a fracture as that, say half an inch in depth, materially weaken the tension? —Yes; it would very considerably, and would also rather indicate the casting to have been in tension on cooling. 224. The part that cooled quickest would be the web —the part that would come into compression ? —Yes. 225. And that would throw the other side into unnatural tension ?—Yes. 226. Have you had the fragments cut ?—-There were some pieces taken off. 227. I mean, did they cut into the interior to see what character of iron was in the interior?-: — That has not been done yet. 228. I gather from the last returns that you made that the whole distributed weight, theoretically, would be about 396 tons ? —On the bridge ? 229. On the anchors ? —Oh, no, about 54 tons. The whole strength of the girder would be 396 tons distributed, equivalent to 198 tons at the centre. 230. Do you know what processes are generally used by practical ironmasters when calculating that strength ?—Generally it is a great deal of rule of thumb. 231. I suppose they have certain rules taken from books ?—Yes, rules taken from Molesworth. 232. If two persons had sent in estimates, one stating that the strength of the anchor-plate was 139, and the other that it was 141 tons odd, they would be similar, would they not ?—That would be practically quite identical. 233. You consider that the anchor ought to have held a larger weight than they give in ? —I make it 19S, and they make it 140. 234. That 140, you consider, would have had to bear a strain, practically, at the time of fracture of about 54 or 56 in the centre ? —Fifty-four. 235. That would give a factor of safety of about something under 3?— Yes, it would be under 3. 236. That is not taking into account any rolling weight?— Yes. 237. Do you consider, supposing the iron to have been in natural tension, that there could have been any effect upon it from a severe frost at that depth in the ground?—No ; the frost could not have had any influence whatever. 238. According to the design of the anchor-plates, was it intended that these holes should have been at right angles to the side that is in tension ?—Yes. 239. Have you examined these holes ?—Yes. 240. Are they perpendicular to the surface ? —No. 241. How far are they out ? —Some of them as much as a quarter of an inch in five inches. 242. Are any of these out in the direction which would bring the strain at an angle to the tension?— Yes. .

E.—2

13

243. Yv'ould that have the effect of increasing the tension upon the flange ?—Tt would have a very great effect. 244. But you could not make any accurate calculation of it ? —No ; so far as it operated, it would tend to increase the tension. 245. It would do that?— Yes. 246. Dr. Knight.'] In fact, the rod would act as a lever?— Yes; tending to burst the castings. 247. Mr. FitzGerald.] I conclude those holes were accurately shown in the drawings?— Yes. 248. I suppose where the moulder went out was in putting in the cores ?—Yes. 249. Mr. O'Connor has handed in a description of the accident, and accompanied it with a facsimile drawing of the face of the flange. Have you seen that drawing ?—Yes. 250. Do you observe that the drawing is in accordance with the drawing in the plans with respect to the arrangement of the holes ?—lt is slightly out. 251. Do you observe that upon the two ends of the anchors, two of the holes come much more nearly opposite one another in the casting than they do in the original drawing ?—Yes. 252. Is it not the case that the fracture in both eases has taken place where the holes came opposite to one another ? —Yes. 253. "What did you estimate to be the breadth of the flange in applying the formula for determining the strength of the anchors?— Eight inches. 254. How much of the iron did you take inside the line of the holes ?—Four and a half inches. 255. On both sides?— Between the two rows of holes. 256. You take 1$ inch inside the holes ? —ln taking the strain out, I took 8 inches as the width. 257. That would be taking the holes as 2 inches wide ? —Yes, it would be that. It is taking the whole width of the casting, and deducting the holes. 258. What means did you adopt to judge that a beam of that breadth which you allowed would be equivalent to a beam of the whole width perforated ? —I took the ordinary formulae in use. There is no reason to believe that the holes would have any influence beyond their own area, as there was a broad web between them. 259. You think that the beam, as cast, ought to have been as strong as a beam 8 inches wide, without holes ? —Yes, quite. 260. Dr Knight.] The actual width of the beam was a foot ?—Yes. 261. And you have taken only 8 inches of it as effective?— Yes. 262. In consequence of the holes?— Yes. 263. Mr. FitzGerald.'] The fact that the beam did break at the place where the two holes were opposite to each other would seem to indicate that the iron between and outside the holes must have been of material assistance to the beam. It broke, in fact, at the weakest place, as regards the holes ? —Yes; undoubtedly it was the weakest place through the holes. There is no question that the iron outside the holes is effective. 264. Have you made any calculations of the actual strength since the fracture of the theoretical strength of the girder at the point of fracture ? —Yes ; it is greater compared to the weight than at the centre. 265. You have satisfied yourself of that ?—Yes; I have satisfied myself of that. The weakest point is not the point that was broken, relatively to the weight. It is stronger at the point of fracture than it is at the centre. 266. Not stronger, but stronger in proportion to the weight ?—Yes; that is what I mean. 267. Dr. Knight.] Then, do you not consider that was the weakest place where the fracture actually did take place ? —lt was actually the weakest place, but where a girder is on the point of breaking, it very seldom yields at what is theoretically the weakest point. 268. But in this case both the anchor-plates broke in the same place, where the holes were opposite each other? —I dare say that had some slight influence ; enough just to define the point at which it would break. 269. Mr. FitzGerald.] You observe that they both broke in the circumference of a circle, of which the centre is the opposite bearing ?—Yes ; but I do not think that has much influence. 270. It is very curious how true they are in both cases, is it not ? —Yes, it is. 271. Have you thought of assigning any reason why they both broke in the corresponding places as nearly as possible at the opposite sides ?—No; except that the holes are just opposite each other. 272. Dr. Knight.] Are they opposite each other in the original drawings? —No. 273. Mr. FitzGerald.] Do you suppose that it was the moulding which was wrong then? —Yes, of course it must have been ; but in making the casting, so far as the holes were concerned, the moulder would not be altogether guided by the pattern, because the cores for the holes would be put in separately after tbe pattern was taken out of the mould. 274. Dr. Knight.] I observe that the holes are not opposite each other in the drawing ?—Yes; it was so planned designedly, with the intention of increasing the strength. 275. That object is lost in the casting ?—Yes ; it is not a very important matter. 276. Mr. FitzGerald.] Not until it came to the very last; then it became of importance. 277. Now, with regard to these chains, there were thirteen of them, each upon a separate screw bolt ? —Yes. 278. To what extent could you secure that they would bring an equal strain upon all the ropes ? —As long as the deflection of the ropes is the same, the tension would be the same. They could not bring an uneven strain upon the casting. 279. Would that be the case where the ropes were made of different material ?—Not unless of different weights ; but if the specific gravity was the same, the strain would be equal. 280. The ropes, I understand, were clipped together every 10 feet ?—Yes. 281. After the ropes were clipped together, and if some of the ropes had given out any, might not an unequal strain be brought upon the screws without its being indicated by the rope ? —No ; it would be indicated, because the wires would slip through the clips. The clips are not bo tight as to prevent that.

E.-*2

14

282. We understood from Mr. O'Connor that there had been a slight movement, to adjust which he was keeping the screws available to the last moment?—-Yes ; so he told me. 283. In your opinion, practically, is it not possible that the strain throughout the whole length of the anchor was not equal upon all the ropes?— Well, I think there could have been very little difference, if any. 284. But whatever difference there would be, would probably have determined on what point the anchor would have broken ?—Yes ; it might have done that. 285. It would have broken on any rope on which there would be any particular strain ? —Yes ; on that side of the centre. 286. Dr. Knight.] Did you give directions to the Superintending Engineer when he was to fill in the ashlar work with concrete ? Did you give him written directions ?—No ; I did not give him written directions; I seldom do. I gave him directions on the spot, and I expect them to be carried out at once, unless there is some reason to the contrary. 287. When you say "at once," do you mean when the bridge is completed ?—No; I certainly meant it to be filled as soon as the wires were on. Not next day, because the ropes were not in position when I was there. 288. Mr. FitzGerald.] What you expected was that as soon as the ropes were up they should be filled in ?—Yes. 289. Dr. Knight.] You did not give these instructions in writing ?—-No; I very seldom give instructions in writing. I give them on the work when passing. 290. Do you think you gave them in those precise terms that they could not be misunderstood?— It was my impression they were quite definite enough. It was for the sake of taking the weight off: the anchor-plates. It was understood that was the reason, and I had no doubt, in my own mind, that the thing was to have been done at once. In fact, two or three times I thought of telegraphing down to know if it had been done. 291. But you did not telegraph ?—No. 292. Mr. FitzGerald.] Supposing it was said that the terms of the contract required the contractor to bring the bridge to an exact form, and to bring all his strains equal, would it have been necessary, in your opinion, to have kept open the action of the screws until a later period ? —Well, I cannot believe that it was necessary. 293. When the concrete was in, the ropes would have been entirely jammed?— Yes. 294. Dr. Knight.] I understand that when you had given the proper deflection to the chain, that would not be interfered with by the weight put upon it ?—Exactly. Invariably in wire bridges they are bound together. Ido not think there would have been any difficulty in this case. 295. When the weight of the platform was put upon the bridge, would that cause a greater deflection? —-Yes ; it would bringdown the whole thing a little. 296. Mr. FitzGerald] It would stretch the wires a little ?—Yes. 297. Dr. Knight.] Would it be necessary to screw them up again ?—No ; the screws would be perfectly useless. You could not screw any of them up again. 298. Then, supposing you had contracted for a deflection of 25 feet in the curve, and if it came out to 26 feet ?—You would have to take it as it was. You could not lift it up again. 229. Would they have allowed for that in stretching the wires ?—Yes. 300. I understand they also allow a camber of 12 inches in the truss? —Yes.; that was allowed to allow the truss to take an equal bearing throughout. 301. It has been explained to us that the bridge was not on a level; that there was a gradient of 1 in 100 ?—Yes. 302. And the lowest point was on the side that broke ? —Yes. 303. Would that have had any effect in altering the strains ? —No ; no practical effect. It is too small to have any practical effect. 304. The tension would not be quite so great on the lowest side? —No. 305. Mr. FitzGerald.] I understand that the trusses rest on the abutments ?—Yes. 306. Have you made any calculation as to what the weight of the bridge would have been on the ground, arising from the stiffness of the trusses ?—I have made a calculation. It would have been about 35 tons. 307. You think that the chains would have been relieved to the extent of about 35 tons? —Yes, by the truss. 308. To what extent would that relieve the pressure on each anchor-plate in the middle ? If all the suspension rods were screwed up to the suspension girders under the truss, they would have been complete as far as taking the bearings goes ?—Yes ; I have not taken any credit in any calculations for the strength of the truss, because it is quite possible it might be left to sag out of shape or bearing, so as not to give any great support. Therefore, I have disregarded it; but I have little doubt that it would carry 35 tons if brought to a state of tension. 309. It might have been in tension for some distance from each shore ?—Yes. 310. Some part of its own weight, at all events, must have been supported ?—Yes. 311. That would diminish the pressure upon the anchors ?—lt would reduce it to 36 tons on the anchor-plates. The factor of safety would then be 51. 312. That would give you a factor safety of 5\ ?—Yes. 313. Dr. Knight.] Is it usual to take into consideration any strength that maybe derived from the girder? —It depends so much on the form of the girder. A girder as strong as that, compared with the strength of the chains, would always be taken into consideration by lowering the factor of safety. The direct object of the girder is to distribute the load.

15

E.—2.

APPENDICES. APPENDIX A. "Wellington, 19th August, 1876. Be Accident at Grey Gorge Bridge. Memorandum for Soyal Commissioners. The breaking of the anchor-plates occurred at 4 a.m. on the morning of the 28th ultimo. The anchor-plates which gave way were those at the south (Westland) end of the bridge. At the time of the accident the trusses were completed throughout, and notched into the crossbearers, which were hung to cables, with clips, complete. The horizontal cross-bracing at bottom of trusses was also completed ; the cross-ties on top were fixed, but not the braces, and the joists over two-thirds of the length of the bridge were laid. The only portions of the superstructure remaining unfixed were therefore the top bracing, one-third of the joists, and the planking ; but against this there was a large quantity of tackle and machinery (crab winch, derrick, chains, ropes, &c.) at about the centre of the bridge, and a great part of the planking strewed along it in the shape of scaffolding, so that it is probable that the strain on the cables was as nearly as possible the same as would result from the bridge completed. On the 24th of the month (three days before the accident), there was a heavy gale blowing down the Gorge, which caused the bridge to sway to the extent of about six inches from side to side, but no sign of its being in any way damaged was apparent. The fact that the cross-braces were not fixed on top would no doubt account for the swaying, which they were calculated to counteract completely, supposing wind blowing with a force not exceeding 50 lbs. per square foot. On the night preceding the accident, there was a very sharp frost, followed towards morning by a rapid thaw. Neither of the tunnels show any disturbance of the stratum or masonry; the only things except the anchor-plates which were disturbed being two out of the four cast-iron girders on which they rested — namely, the one most remote from centre line of bridge in each tunnel, which were shoved aside by the anchor-plates when they parted, and of the two so shoved aside one was broken laterally by the force of this side pressure. The tracing attached shows the exact dimensions and relative positions of the anchor-plates as they would be seen looking from the Westland side of the river towards the Nelson side by a person standing behind them. The immediate result of the accident was that the ropes ran rapidly over the piers for a length of about 80 feet, carrying the long ends of the anchor-plates with them up to the surface of the ground, and leaving the short ends remaining iv the tunnels. By this time the cables had reached the bottom of the river, so ceased to run out before the ends came up to the pier. They remained therefore hanging over in about same position as when erected, except that the saddles, which were on rollers, were pulled off by the ropes as they ran out. It was stated by one man, who was awakened, and looked out at the first shock when the anchorplates gave way, that there were two distinct reports, indicating that one of the plates went a second or two before the other, and he says that the timber work of the bridge stood sustaining its own weight for a few seconds after that, but he could form no idea which side went first. The flaw observable on the west plate would seem to indicate that it should have gone first; but as against that there is the fact that the western cable lay on top of the wood work in the river, while the eastern cable lay under everything. It is difficult therefore to connect the flaw with the accident, unless by conceiving that the second cable which broke would jump further than the first one, on account of having the whole instead of half the weight acting upon it at the time it broke. C. T. O'Cownob, C.E.

APPENDIX B. Ultimate strength of bridge— 10 large wires at 44 tons (5| circumference) ... ... ... 440 4 large wires at 36 tons (4| circumference) ... ... ... 144 12 telegraph wires at 34 tons (1 area) ... ... ... ... 408 992 Load— Weight of bridge, 96 tons. Strain, ... ... ... ... ... ... 144 tons so Factor of safety ... ... ... ... ... ... 3-J--504—504 Boiling load, 30 tons, distributed by truss over 80 feet —s=— ... 78 Factor of safety ... ... ... ... ... ... 5 390—390 894 Add for increased weight at towers ... ... ... ... 40 934

E.—2.

16

This allows nothing for strength of truss, which would be very considerable. If bridge were covered with people the load would be 300 x 8 x 65 pounds=ls6,ooo pounds=7o tons. Tons. 35-0x75 . 86 10° Strain from dead load ... ... ... ... ... ... 144 249 249 gg2=i=factor of safety, which is sufficient for a load which will seldom or never come on to the bridge. Suspension rods — Weight of bridge (less cables), 82 tons. Supported on 60 points ... ... ... 137 on each rod. Weight for one wheel ... ... ... 250 As 10:l|:: 2-5 ... ... ... ... "43 4 "30 tons maximum weight. Area of r0d=78x30=2340 of safety. The actual load will be less than the above, on account of the distributing power of trusB-Anchor-plates— Strain due to weight of bridge ... ... ... ... 144 tons. 2)— On each plate... ... ... ... ... ... 72 „ Add 4 per cent. ... ... ... ... ... 3 „ 75 „ This is distributed over 2 feet; the equivalent weight at centre is equal to 54 tons. Strength of anchors— Tons. Inches. B=Bending Moment=^=: 54 * 42=5675 I % ' M=Moment of Resistance. 1= „ „ Inertia. y=Modulus of rupture of cast iron=l6 tons. T=distance of neutral axis from lower edge of casting. T _Aff +rf Aa (H + ft) 2 • 12 + 4(A+o) _40_x 25 +15 x 25 40 xl 5 x 100 12 4 x 55 =115 + 273=388=(1). Neutral axis coincides with centre of gravity=3'4 inches above bottom of casting. = =1826 ft.-tona. =-jryg=factor of safety=3"2. Or, by another formula— Effective tensile area = 97 sq. inches — (a). Distance between centres of gravity of tensile and compressile areas = 67 in. = (b). Tensile strength of cast iron = 8 tons per sq. inch = (/"). Modulus of rupture =f x -£■ = 18= (D Distance from neutral axis to upper edge of plate = 6 *4" = (J). Distance from neutral axis to lower edge = 36" =s (c). Moment of resistance = m = a © c —=9 • 7 x 18 x 6• 7 g^ = 2,080 tons. W = weight equally distributed which anchor-plate could carry =42 ~ tons. If weight were concentrated at centre, W would be 198 tons. x> \ j- WL 54x42 _„_ , B== bending moment = —r- = — -j — = 567 tons. Effective compression, area = 3"2x3= 9"6sq. inches. Effective tension, area = 4 <sx0 <7=3"15 9 " 67 sq. inches cs (a). Distance between centres of gravity of tensile and compressile members = (I). Distance from neutral axis to outer edge of compressile member = (b).

17

E.—2

Distance from neutral axis to tensile member = (c). V- *- x° - 567 X 3-l?-4-9 *~al X ~b~— 67x9-7 6-4~* a f— tensile strength of cast iron per square inch in tons = 8. (D = modulus of rupture = f + ~ = 18. -^=^| = 3-6= factor of safety.

APPENDIX C. Mr. Eobeetson to the Chairman, Grey Gorge Bridge Committee. Sib, — Phoenix Foundry, "Wellington, 21st August, 1876. In accordance with your request, I have examined broken anchor-plate, and have the honor to submit the following report for your consideration: — I find the broken anchor-plate to be an ordinary green sand casting, of average Clyde iron, and am of opinion it should stand a breaking strain of 139 tons 16 cwts., and be capable of bearing a working load of 46 tons, at a span of 3 feet 6 inches, or, in other words, a distance of 3 feet 6 inches between supports. I am further of opinion that the casting should have been moulded the reverse way —viz., the rib uppermost. The advantage gained would be that the metal would be stronger where the tension comes on, and the contraction in cooling more equal. lam of opinion that the plate was in tension when it left the foundry. I arrive at this conclusion from the very unequal density of the plate, as on the rib side it is fine and close, and at the lower it is coarse and open. I am, &c, The Chairman of Committee, David Robebtson, Be Greymouth Suspension Bridge. (of Robertson and Co.). lieport on the Broken Anchor-Plates belonging to the Suspension Bridge at the Grey Siver Gorge Lion Foundry, "Wellington, 21st August, 1876. Having been requested to examine portions of the anchor-plates, I beg to report as follows: — The cast iron, in the broken pieces, I consider to be of fair quality, and about the average quantity of pig iron melted once from the blast furnace. The large granulation of the iron, to be seen in the centre of the fracture, is caused by the contraction of the inner portion of the metal after the outside is cold, and is generally to be seen in large bodies of metal after they are broken. The anchor-plate would have stood a greater strain if the side in tension had been cast downwards in the moulding of it, as the iron then would have been more dense and sound upon that side. The broken plate has the appearance of having been cast the other way downwards, by the rough surface of the face and the small sand hole to be seen at the edge of the fracture, caused by the sand floating on the top of the metal as it is poured into the mould. In one of the pieces there is a small flaw in the metal leading into one of the holes that is broken through on the face side of the anchor, which would be in tension when in position before giving way. It is caused either by slag getting into the mould, or the metal not being hot enough to unite when running round the core. I could tell which by seeing the other part of the plate that was broken from it. In calculating for the exact strength or strain that will break different forms of cast iron, is one of the most difficult to arrive at with certainty, for, in carrying out experiments, scarcely two pieces of metal will break with exactly the same weight placed upon them; therefore, the best safeguard a practical engineer has is a large margin for safety, by taking the ultimate transverse strength of one square inch of cast iron, 12 inches long, and weighted in the centre, for data to calculate from. I consider the breaking strain of the anchor-plate to be 141 tons 18 cwts. 1 qr. 9 lbs. at the largest section through the bolt holes. The breaking strain without the bolt holes, 172 tons 15 cwts. 1 qr. 12 lbs. I have, &c. E. Seagab, Engineer.

APPENDIX D. Telegrams. From Mr. FitzGerald to Mr. McAbthttb. 22nd August, 1876. The Commissioners for inquiry into the accident to the Grey Gorge Bridge request you to be so good as to reply to the following questions:— 1. Did you personally see the anchors cast ? 2. Did you superintend the making of the mould, and was its position determined as to which side of the casting should be uppermost? 3. Did you inspect the metal used, or take any steps to ascertain its quality ? James Edmund FitzGeeald.

E.—2.

18

Mr. McAbthitr to Mr. FitzGebaxd. 22nd August, 1876. 1. I did not see them cast. 2. I did not superintend making of mould, but saw the correctness of pattern when made, and the uppermost side is distinctly specified on contract plans —viz., that with least number of holes. 3. I did not see material used, but inspected and sounded plates after casting. D. W. McAethtjr. Mr. FitzGebald to Mr. McAbthtjb. 24th August, 1876. We do not understand from your telegram which way anchors were cast. Fracture shows that the web must have been downwards. Please state which was the bottom of the casting. James Edwabd FitzG-eeald. Mr. McAbthub to Mr. FitzGebald. I thought from your last telegram that you wished to know which side was up, when in bridge. Sorry for misapprehension. I did not remember which side was uppermost while castings were made, but I shall telegraph to Greymouth to ascertain. D. "W. McAbtiiitb. Mr. McAethitb to Mr. FitzGebald. 25th August, 1576. AxcHOB-plates were cast with web downwards. D. W. McAethue.

By Authority: Gboegb Didsbubt, GoVernment Printer, ■Wellington.—lB76. Price 9d.]