that the radius was flat, 1,500 feet, and the versed sine was 20 feet. It had occasioned a great deal of anxiety during the time the dock works were in progress, as there were always ships behind it in the dock, and if the dam had failed the result would have been disastrous. Fortunately it had stood, and the reason he assigned for it apart from the care bestowed upon its construction by the contractor, Mr. McCormick, and by his engineers-was that it had been made of a circular form, according to Sir John Hawkshaw's design, and that it acted as an arch. Soon after the construction of this cofferdam Sir John Hawkshaw had to make another dam for the construction of the eastern entrance of the South Dock of the West India Docks, which was also circular, but the curve was much sharper, the radius being 150 feet, the versed sine 50 feet. There was nothing particularly novel in the design of this dam, except that, the curve being so sharp, resort was had to a new mode of making the walings, which, instead of being of timbers of large scantling, were of laminated planks, care being taken effectually to break the joints of the planks. Sir John Hawkshaw had also constructed a cofferdam in the Zuider Zee, near Amsterdam, which was a complete circle, the radius being 525 feet. It was made to construct works on the Amsterdam canal, and it was on silt and soft material throughout. It was a difficult and very interesting work, but as he hoped soon to present a Paper to the Institution on the Amsterdam Canal and the North Sea Piers, he reserved the description of that dam. The thickness of the puddle in the Victoria Dock dam at Hull was 6 feet, and in that of the West India Dock dam 5 feet. In both those dams the usual system of through bolts was employed to connect the two rows of piles, but he believed that such bolts were mischievous. The puddle could not be got tight round them; it sank and left vacuities about the bolts; and although the puddle was often well rammed, he believed that dams with through intermediate bolts almost invariably leaked. In some large cofferdams that he had recently been associated with, constructed for the purpose of carrying the East London railway through the London Docks, intermediate through bolts had been altogether dispensed with by Sir John Hawkshaw, and the thickness of puddle between the two rows of piles was reduced to 4 feet 6 inches. Instead of these through bolts, buttress piles were driven at intervals outside the dams to withstand the pressure of the puddle within. These dams had to penetrate a bed of gravel about 12 feet deep, and piles could not be driven through it. In order, therefore, to construct the dams, a trench was excavated through the gravel down to the London clay; that trench was

filled in with clay, into which the piles were driven. The clay was thus consolidated, and the plan of procedure answered. He could state as the result of a very large experience in the construction of dams that, as a rule, it was desirable to dispense with intermediate through bolts, retaining only the top row; to have as little thickness of puddle as possible; and to have no horizontal irregularities or ledges to catch the puddle, so that it would settle down uniformly. Of course dogmatic laws could not be laid down applicable to every case. If, for instance, the ground were very bad, and insufficient to enable the piles to have a firm hold, it might be necessary to have resort to through bolts at the bottom, at or near the ground line, to connect the piles together. The conclusions he had mentioned being the result of considerable experience, he hoped that they might be of service to gentlemen who might be called upon to construct such works which were sometimes of a very difficult character.

Mr. BRUNLEES observed that in the Avonmouth Dock, recently completed, he had formed a dam upon a different principle from any of those described in the Paper. The foundation consisted of a single row of piles driven closely together. Above the level of the sill the dam was formed of what might be termed a lock-gate, with planking on the outside, thoroughly caulked. The dam was strutted in the usual way from the inside. The head of water was 44 feet, so that it was necessary to be very careful. At Lynn, where he had constructed a dock some years ago, he had used the double pile and puddle in the manner described in the Paper. At Whitehaven he had used a dam of earth in front of the lock. Considering that in the one case the head of water was 42 feet, and in the others only 22 feet, he believed that the timber dam, well caulked, was far superior to the others. He might add that the timber dam was much cheaper, because the whole of the timber was fit for use after its removal. In the one case there was only labour to pay for, and in the other cases there was the material also.

Mr. BRUCE thought that in driving 6-inch piles instead of whole-balks an inefficient dam was sometimes made, and more money spent in the end, because there was much more water to contend against, and greater difficulty in keeping the dam tight. Thirty years ago at Berwick Bridge, where there was a good deal of cofferdam work, 6-inch piles had been abandoned for that reason alone. By the use of whole-balks less timber was destroyed in the end, the dam was made more easily, and the foundations were got in more cheaply. That might not be so

in every case; but he believed it was so in many instances. In the cofferdam at Birkenhead the piles were said to have been battered both ways. The puddle was narrower at the top than at the bottom, so that in sinking it would leave the sides of the dam inside, and it would necessarily be very difficult to keep it tight. He had never seen a dam made in that shape, but it appeared to him that it was not a very good one, having regard to the necessity of keeping out the water. From the fact that at Dublin the piles were not shod, he gathered that the ground must have been comparatively soft. The Author had not stated what kind of timber was used, but he presumed that it was Memel. For that class of work he thought that the American rock elm, at all events where the ground was hard, was a much better timber than Memel. Less of it was destroyed, and it was more valuable when sold again. The walings were entirely on the outside. Of course it was the right thing to have none inside, so as to allow the puddle to settle; but he imagined that if the piles got out of line it was a good thing to have, while driving, a temporary waling inside.

Mr. Woods thought that the mode of constructing quay walls referred to by Mr. Dixon was applicable under certain circumstances. He was himself constructing a quay wall at Bilbao on that principle. He simply drove sheet piles in two lines, the interval between corresponding with the width of the wall. The foundation was compact sand, which he was enabled to remove by the Woodford sand pump; and the trench was filled in with concrete. He could excavate to a depth of 12 or 15 feet. The work was not completed; but, as far as he could judge, he believed the plan would prove successful.

Mr. PRINCE wished to call attention to the strange neglect, on the part of eminent engineers, of the system of under-sinking foundations. As a rule Indian engineers avoided cofferdams, and adopted the old native system of under-sinking; and he recommended members of the Institution who were about to undertake works of that kind to look into that system, which would save an enormous amount of unnecessary expenditure.

Mr. RUSSEL AITKEN agreed with Mr. Dixon in thinking that the system of cofferdams was very expensive, and might often be dispensed with. He had himself dispensed with it in the construction of the Mazagon pier in the harbour of Bombay. He drove sheet piles, and filled in the space between, 16 feet wide, with liquid concrete; and the work had been entirely successful. He had seen the cofferdam at Hull, referred to by Mr. Hayter, when the works were nearly completed. It had assumed a most

irregular form, so much so that he had recommended the engineer in charge not to repeat it. A cofferdam with a radius of 1,500 feet received no support from the fact of its being circular. For every ton per lineal foot the pressure was 1,500 tons; so that with such a flat radius there could be no efficient strengthening power.

Mr. DOHERTY, in reply, said he had great pleasure in informing Mr. Giles, who objected to the 10-feet spaces in the dam at Dublin, that the work had stood remarkably well. Mr. Dixon had raised the question whether any dam was necessary. Such dams, however, had been used in the construction of the Thames Embankment, and if they were good for London, he saw no reason why they should not be made in Dublin. With regard to the dam at Hull, Mr. Aitken had pointed out the very slight advantages to be obtained by a curved dam of 1,500-feet radius, the dam itself, being only 470 feet long. He was glad Mr. Hayter had adopted a puddle wall 4 feet 6 inches instead of 6 feet thick. Mr. Doherty had been led to adopt 4-feet 6-inch puddle, notwithstanding the example of Hull. He believed few of the younger members of the Institution had had more experience in cofferdams than himself. He had had experience of the great dam at Birkenhead, and of all the dams erected by Sir John Hawkshaw at Hull; and he had erected a length of 2,500 feet, as described in the Paper; he was also erecting other dams at Dublin, in connection with the rebuilding of the Carlisle Bridge and another new bridge over the Liffey; and the conclusion at which he had arrived was, that if the puddle were held closely together, a wall 1 foot thick was as good as one of 10 feet for excluding water. With regard to the remarks of Mr. Brunlees about a single row of piles, he was at the present time constructing a dam in accordance with that system. He thought he had made an improvement in the ordinary mode of driving a single row of piles for the formation of a dam. He had joined three piles into one by sawing the edges straight, and fastening the three piles together by means of 3-inch elm dowels let into the joints, 3 inches into each pile, and at distances of about 8 feet apart, then fastening the piles together with 14-inch wrought-iron dogs, the interior joints being caulked previous to being driven, so that after they were driven he had only every third joint to caulk. He agreed with Mr. Brunlees that where there was not a great head of water, and where there was no danger of the dam being floated out, a single row of piles was proper. He took no credit for this plan, because it had been provided for in Mr. Stoney's specification. He had also another form of dam in

operation, where there was a great head of water, and the thickness of the puddle was 4 feet 3 inches. The dam was driven with halftimber sheeting. It was 200 feet long, and it inclosed a pier and an abutment of a new swing bridge. The foundations of the pier, extending down to the gravel bed, were 18 feet below low water. In that case there was no foreshore, but simply a head of 19 feet 6 inches of water to be resisted; that was done by a single dam with a wall of 4 feet 3 inches of puddle, with sheet piles, struts, and wales, the struts being 10 feet apart from centre to centre. He agreed with Mr. Bruce with regard to the superiority of American elm over Memel for hard driving. In 1851 Mr. Bruce described a cofferdam across the Tweed, where he had first used Memel, and then, finding it to be too soft, adopted American elm; but even then the dam was not a very great success, because it could only be used at low water. He quite agreed that in many instances a dam could be avoided altogether, but that was not the case in the work which he had described. It was only by driving a dam that the roadway over which all the traffic passed could be kept up. Water had to be excluded on one side, and the roadway kept up on the other, and between the two a river wall had been sunk to a greater depth, he believed, than any on record-28 feet below low water-and in immediate contact with the tidal river. Temporary waling inside, referred to by Mr. Bruce, had been adopted by him. Mr. Woods' plan of driving two rows of sheeting, taking away the old wall, and filling up the space between with concrete, would not have answered.

Mr. STEPHENSON, President, said that he had discovered, many years ago, in building retaining walls and large sluices in the fens, that a thick wall of puddle was a mistake. It was the custom in those days to use puddle 4 or 5 feet thick behind walls; and the result was that when it sank down it generally did so in a body; but if put in 14 or 18 inches thick, the pressure made it follow down the wall instead of settling away from it. In Holland the sea banks were made with silt that melted like sugar in water. He had been engineer-in-chief for the construction of some of the heavy banks in that country. The slopes on the sea side were 5 or 6 to 1, and on the land side 3 to 1, giving several hundred feet for the base; and the only safeguard against the water coming on to the land was a thin wall of inferior clay in the middle of the bank. The whole of Zealand, he believed, was at the present time embanked in that way: showing the efficacy even of the bad clay there used if it were only well compressed. He believed the Author was right in his conclusions on the subject of puddle.