INTAKE TOWER AND TUNNEL AT THE ST. LOUIS WATER WORKS’
Since the completion of the Low Service Pumping Station at the Chain of Rocks in 1892, the City of St. Louis has depended for its entire water supply on one intake tower and a tunnel seven feet in diametter connecting this tower with the wet well near the Pumping Station. This tower is located on the western edge of the channel about 1,500 feet from the west bank of the Mississippi River. Under normal conditions the capacity of this intake is 125,000,000 gallons per day. On account of floating ice together with extreme low water which have existed at various periods, the capacity has been reduced to such a degree as to threaten the city on several occasions with a water famine. In 1900 a report of the water supply was made by a Hydraulic Commission, consisting of Messrs. Geo. Wisner, Benezette Williams and Allen flazen. In this report it was recommended that a new intake tower and tunnel should be constructed without delay, since, in their opinion, it was extremely hazardous for a city the size of St. Louis to be dependent on a single intake for its entire water supply. Little or no effort was made to improve the intake conditions until Mr. Edward E. Wall was made Water Commissioner, who at once started on the preparation of plans for a new intake and an 8-foot tunnel, and after thorougly investigating the natural conditions the plans were completed, and the consent of the United States Government was obtained for the location of the tower and the necessary money appropriated by the Municipal Assembly. These additional improvements of the intake conditions will increase the intake capacity 150,000,000 gallons per day, making a total capacity of approximately 275,000,000 gallons per day, which means an ample intake supply for many years. The contract for this work was let in June, 1913, to the Fruin-Colnon Contracting Company, for the sum of $190,355, and at present the work is completed except the superstructures for the screen chamber and intake tower, and barring any unforeseen difficulty these improvements will be in operation the first of the year.
The new tunnel is approximately 2,767 feet long, extending from the wet well near the pumping station to the new intake tower located some 700 feet West and 300 feet North of the present intake. From the screen chamber the tunnel has a descending grade of 6 per cent, for 560 feet to the shore shaft and beyond this an ascending grade 1 foot in 1,000 to the down take shaft, thereby affording drainage from both directions to the shore shaft. The tunnel is straight with the exception of a curve with an 80-foot rail, located at the extreme west end and near the screen chambers. At the short shaft the bottom of the tunnel is 98 feet below the surface or 60 feet below the river bed. while at the downtake shaft it is approximately 35 feet below the river bed. From the shore shaft a connecting tunnel 4 feet in diameter was driven to the present pump pit shaft through which the main tunnel can be drained in order to make any necessary repairs. The main tunnel, circular in sections, was driven with a diameter of 10 feet. In the screen-chamber excavation, trouble was anticipated due to caving. To guard against this, steel sheet piling was driven 4 feet outside the dimension lines. The piling was braced horizontally with sets of 10 x 12 in. timbers, placed approximately 7 feet c. to c., and set as the excavation advanced. This excavation was carried 20 feet in earth and 9 feet in quicksand without serious difficulty. The shore shaft, located on the bank of the river, approximately 100 feet from the water’s edge, was used as a working shaft in driving the tunnels. The upper part, in earth and quicksand, was excavated with a diameter of 15 feet 4 inches to a depth of 21 feet in earth and 12 feet in quicksand by means of an open concrete crib. This crib was built in place as the excavation advanced. No water appeared in the quicksand and consequently but little trouble was encountered. The rock portion of this shaft was excavated with a diameter of 12 feet and did not vary from the usual methods employed in sinking shafts. The progress was greatly retarded by running into a number of water seams near the surface of the rock, which gave a flow of approximately 100 gallons per minute. An attempt was made to grout these seams and to a certain extent was successful, although more or less water appeared at different elevations and a 3-inch discharge pump was required to take care of this water.
The power and compressor plant, located just north of the shore shaft, consisted of three 75 h. p. boilers and a two stage air-compressor with a capacity of 950 cubic feet per minute, maintaining a pressure of 100 pounds in a receiver 3 x 10 feet placed outside the compressor houses The air line was 4 inches in diameter from the receiver to the bottom of the shaft, and 3 inches to the face of the heading, reduced at the end by a 2-inch manifold to receive the air hose for the drills. A rotary pressure blower with a capacity of 4.8 cubic feet per revolution, running at 325 revolutions per minute, was used for ventilating the tunnel. An 8-inch ventilating pipe was laid from the blower house through the tunnel to within 100 feet of the face of the heading, and was extended as the work progressed. The average time to clear the tunnel from smoke and gas in order that the shift could get back in the heading was from 12 to 20 minutes. For transporting muck to the shaft a track of 24-inch, gauge was laid on the south side of the tunnel and wooden side-dumping cars of 24 cubic feet capacity were used. These were hauled by a 25-h. p. storage battery locomotive weighing 3 tons and having a hauling capacity on level track of 10 tons. The muck was hoisted up the shaft by means of two cages, each large enough for a car, operated by a two-cylinder link-motion reversible hoist, equipped with a drum 48 inches diameter and 48 inches face. More or lesswater was anticipated in the east heading, as this end of the tunnel is located under the bed of the river, and facilities for handling 900 gallons per minute were provided. However, the maximum flow of approximately 350 gallons per minute was obtained at the time the tunnel was completed. This water gave little or no trouble,, but on account of the slight grade in the east heading a pump of 100-gallon capacity was necessary to keep the water from the face of the heading.
The tunnel force consisted of three shifts working eight hours each, each shift composed essentially of: 1 superintendent, 1 heading boss, 4 drill runners, 4 helpers, 1 muck foreman, 8 muckers, 1 nipper, 1 signal man, 1 motorman, 1 switchman, 2 cagemen, 1 electrician, 2 enginemen, 1 fireman, 1 pump man, 3 laborers. In. addition, on the day shift there was a blacksmith and helper, 1 foreman , four laborers working on track, and one mechanic repairing drills and other machinery. At the beginning of the work one round was drilled and shot per shift. The drills were then mounted on two columns, one on each side of the heading, with two drills mounted on each column, the 10-foot cut holes and 8-foot side holes were drilled, and this round pulled from 4 to 5 feet per shot. After the organization was more complete, four shots per day were obtained. Columns were still used during this period, and a round of 10-foot cut holes and 8-foot side holes were drilled; this broke an average of 5.16 feet per shot. During this latter period the 8 A. M. to 4 P. M. shift found the heading as left just after the last shot on the previous day. They mucked out, set up the drills, drilled the round, and shot it between 2 and 2.30′ P. M. The shift went in immediately, mucked out the heading again and had the drills set ready for the 4 to 12 P. M. shift. This latter shift drilled the round and shot it between 7 and 8 P. M.; immediately mucked back, set up the drills and drilled another round. Then the 12 to 8 A. M. shift shot this round between 1 and 2 A. M.; drilled another and shot it between 7 and 8 A. M. The average daily progress in this schedule was 20.5 feet. The efficiency of the organization was further increased, and during the last two months two rounds were drilled and shot per shift, or six shots in 24 hours. With this schedule, the drills were set on arms mounted on a horizontal bar approximately 6 inches above the center of the tunnel. This was held in position by a jack screw, tightened against the sides and supported in the middle with a jack screw against the bottom. Four drills were thus mounted, two above and two below spring line. The entire round of 16 holes was drilled from this one position of the bar, with cut-holes 6 to 8 feet and side-round holes 4 to 6 feet deep, depending upon the time allowed to get the second shot. Figure 2 shows the general method of drilling, and each shot resulted in an average advance of 4.27 feet or 25.6 feet per day. Each man drilled four holes. The cut-holes (1. 2, 3, 4. 5. 6) were shot first: next the side-round holes (7. 8. 9. 10. 14, 15, 16) were shot; then the breakdown holes (11, 12. 13, 17) were shot last. Hole No. 7 was used only when the bottom was. high, and this hole was shot with the side round. The holes were started with a diameter of 2 1/2 inches and bottom with a diameter of 1 inch. All shooting was done by means of electric current. Three strengths of Mating gelatin were used at various times; 40 per cent, was found to give too much smoke and gas; 60 per cent, was too strong—the explosion was too quick and consequently would break near the surface of the holes and not at the bottom. Finally, 50 per cent, was used, and with this the best results were obtained. The charge varied both as to the depth of the hole and the quality of the rock, but on an average one stick (1 1/4 x 8 inches) per linear foot of hole was used. An average of 6.7 pounds of powder was used per cubic yard of excavation. The rock encountered was St. Louis blue and gray limestone, easily drilled and shot. A few horizontal seams and occasionally soft places were run into. These gave but little trouble, and timbering was unnecessary. Both headings were turned September 26, 1913. The west heading was holed December 17, 1913, and th east heading holed March 4, 1914. No effort was made for a record on this work, but for the week ending February 19, 1914, 184 feet of tunnel was driven, and from January 21 to February 21 inclusive 745 linear feet were driven in the east heading. The writer believes that this latter set a new record in the United States for tunnel driving. The heading was driven to a finished line and grade, and the actual excavation ran approximately 10 per cent, under that which was paid for under the specifications. A bonus system of payment was used. A minimum progress of 72 linear feet per week was established, and over this the rates were paid per linear foot of tunnel driven, the payment being divided equally between the three shifts. This amounted to approximately $4.6 per linear foot. No man who had lost over three shifts during the week was allowed a bonus. About 905 linear feet of tunnel came under this system and was paid for by this bonus.
* Abstract of paper read before Association of Engineeing Societies, November 18, 1914.
The tunnel was lined with concrete with a mixture of 1:2:4, no reinforcement required. The concrete was mixed and placed by means of compressed air: the necessary air for this work was furnished by the compressor plant originally provided for excavation of the tunnel plus one 650-cubic foot compression making a total air capacity of 1,600 cubic feet per minute. Air was piped down the shaft through a 4-inch line and reduced to 3 inches at the bottom; the three smaller lines shown entering the mixer are connected to this 3-inch line. The lower pipe is 2 inches in diameter and connects directly into the machine end of the 8-inch discharge pipe by means of which the concrete was conveyed to the forms. The upper pipe line, 1 1/2 inch in diameter, enters the mixer near the top and the relative positions of these two lines prevents the bottom of the machine from becoming clogged and also thoroughly mixes the concrete during transportation. The small air hose shown supplies the air to the small cylinder on the side of the mixer. The function of this cylinder is to close the door at the top of the machine when the batch of concrete material has entered the mixer. This door prevents the escape of air from the mixer top when the transporting air blasts are applied. Two sets of forms 30 feet in length were used on this work and were built of structural angles and channels with wooden laggin covered with light sheet metal securely nailed to the laggin. However, this sheet metal was removed at an early stage and a better finish of the concrete was obtained by using plain wooden laggin kept well oiled. The forms were portable and supported on two small trucks running on a 24-inch gauge track placed on the finished invert. The forms were provided with screw jacks built on the trucks for lowering and raising and with turn buckles which were used to pull the wings in laterally to provide the clearance when moving the forms.
The land tunnel was lined first. The pneumatic quarter cubic yard mixer and conveyor was set at the bottom of the shaft. The cement, sand and gravel were loaded directly into the mixer from the top of the shaft by building a hopper at the shaft top with a 10-inch pipe leading down and discharging directly into the mixer. This hopper was made with a sliding horizontal door at the bottom so that upon a signal from the man at the mixer this door was opened by means of a lever, allowing the measured batch to drop directly into the machine. The man who operated at this door also let the water in from a barrel at the same time the batch was dropped. This water had a tendency to clog the chute and a 2-inch water pipe was provided and the water entered the machine directly into the mixer hopper at the bottom of the shaft. One set of forms 30 feet long was set up approximately 70 feet from the west end of the shore tunnel, the 8-inch discharge pipe with flange couplings was laid extending through the forms and approximately 70 feet of intake was laid. After this section of the invert was placed the track was laid at the proper elevation, carefully centered and the forms pushed back over the completed invert and made ready for the concreting at the arch. Heavy bulkheads were built at both ends of the arch with sand bags, and this type of bulkhead proved to be very effective. The discharge pipe was laid along the tunnel from the mixer to within 2U feet of the forms and here 22J4 degrees elbow was inserted, then a 20-foot length of pipe was added which brought the discharge pipe up to the roof of the tunnel and here another 22 1/2 degrees elbow was placed, then a piece of galvanized pipe about 18 feet long formerly used as a blowpipe in the tunnel was placed which carried the end of the pipe over the forms to about the middle. This galvanized pipe, although easily handled and placed, had to be renewed every third day on account of the wear and tear due to the gravel. While the shore tunnel was being lined the river tunnel was made ready to receive the lining. A well devised scheme of placing the concrete in the invert by hand throughout the river tunnel was adopted by the contractor, and this proved to be quite a saving in the time as well as expense. No forms were used for this portion of the work, the concrete was mixed at the top of the shaft, loaded into a hopper and dropped down the shaft through a 16 x 16 inch chute, staggered with one-inch bolts placed two feet center to center which bolts also afforded further mixing. Cars 24 cubic feet capacity were placed directly under the chute ready to receive the concrete and be immediately hauled by the electrical locomotive to the place of depositing. The work was begun May 4 and completed June 2, a total of 25 working days. An average progress of 87 linear feet per day of 9 hours was made, the concrete running 49 cubic yards per linear feet of invert. The invert in the river tunnel was finished a few days before the linnig of the shore tunnel, which made possible the immediate removal of the mixer and forms to the river tunnel. With the air capacity available, it was decided no effort would be made to shoot the concrete more than 1,100 feet; thus the mixer was moved to a point approximately midway between the shaft and the end of the tunnel, and placed in a hole four feet deep so that the top of the machine came approximately 18 inches below the center of the tunnel where a platform was built. The tracks were elevated so that cars were run up and unloaded directly into the mixer, thus dispensing with more or less labor. The material for the concrete was measured in the hopper originally provided for concreting at the invert, and dumped through the chute into the cars and thence hauled directly over the mixer; the cars were side-dumping and the transferring of the material to the machine was very rapid and but little labor was required. A oneinch water line was run through the tunnel to the mixer and water added directly into the hopper of a machine. Two sets of forms were set up, one at the end of the tunnel, 1,100 feet from a machine, and the other 500 feet from a machine; the discharge pipe was elevated 3 feet above the invert so that it would run through the first set of forms and in no way interfere with the moving. The first effort to shoot concrete into the forms at the end of the tunnel was more or less a failure and it was decided as a remedy to place two air receivers, 125 to 100 cubic feet capacities, in the tunnel at the mixer, which relieved the situation. Little or no trouble was thus experienced in shooting the concrete the required distance. The forms were poured alternately requiring from two to four hours, depending upon the distance between the mixer and the forms. While this portion of the tunnel was being lined a perceptible difference in the quality of the concrete between the rear and the front of a form was noticed. Upon investigation it was noted that the gravel in the concrete would discharge at a rapid velocity and deposit itself in the rear part of the forms and the mortar would drop almost directly from the end of the pipes—about midway of the forms. A scheme to overcome this condition was devised by which a 10-inch galvanized pipe 12 feet long was slipped over the 8-inch discharge pipe leading from the end of the arch to the center and this pipe was made loose so as the forms were filled the pipe could be slipped back, thus keeping the end of the discharge pipe no more than 4 feet from the place where the concrete was being deposited. This proved quite effective, and the quality of the concrete was more uniform throughout the form and much more satisfactory results were obtained. lAfter the part of the river tunnel between the mixer and the end was completed the mixer was again moved to the bottom of the shaft and one set of forms pulled ahead on the track to a point 500 feet from the machine and the concreting handled in the same manner as was employed in the shore lining.
The work of lining the shore tunnel was begun April 28 and completed May 22, a total of 20 working days. During this time 446 linear feet of tunnel lining, including both invert and arch, was poured, an average of 23.3 feet per day. The progress was rather slow, due to the fact that only one set of forms could be used. The work of lining the river tunnel was begun June 5 and completed August 3, a total of 09 days. Of these 10 days were Sundays on which no work was performed ; 6 days were lost on account of fire which destroyed the compressor house, and three days were consumed in moving the machine* thus leaving a total number of 50 working days, during which 2,107 linear feet of arch was completed, an average of 43.4 linear feet per day. Two sets of forms were used and poured alternately.
The concreting force employed during this work varied depending upon the number of arches poured per day. The average force per day consisted of one superintendent, one compressor engineer, one fireman, two men on loading materials from the cars to the storage yards, four to five men wheeling material from the storage yard to the hopper at the top of the shaft, one motorman, who also took care of the bull gang of two to three men doing odd jobs and maintaining the track, one foreman and six laborers moving forms and adjusting the discharge pipe preparatory to concreting, one man on the forms while the concrete was being poured and one concrete finisher.
After the lining was completed, quite a number of hollow places were found between the top of the concrete and the roof of the tunnel. These hollows were caused by small rock projections from the roof, the concrete while traveling at a rapid velocity would hit these projections and deflect to either side. The space in these hollows varied from 3 to 8 inches in thickness. The concrete was carefully sounded throughout the tunnel and where hollows were discovered a twoinch hole was drilled with an Ingersol Rand jack hammer drill, grout pipes inserted and the hole filled with grout; one part cement and two parts sand. A Ransome Cannif grout machine was provided and a pressure of 90 pounds was maintained. Where honeycomb places existed in the concrete this grout was forced into the voids and thereby strengthened any weak spots. Delays were quite frequent, due to the insufficient air capacity, while the conm Crete was being shot a distance of 500 to 1,100 feet. In the writer’s opinion this was due to too large discharge pipes in connection with the one-quarter yard mixer used. The amount of air consumed varied from 1-2 to 1-7 cubic foot of free air per linear foot of discharge pipe. A pressure of 110 pounds was maintained at the receiver and by the time the batch had been discharged the pressure was reduced from 110 pounds to 25 to 40 pounds, depending upon the length of the discharge pipe. There is no doubt if a 6-inch discharge pipe had been used instead of an 8-inch the consumption of air would have been materially reduced, thus effecting a better progress and at the same time saving expense.
The tower covering the downtake shaft is located on bed rock at elevation 60 or approximately 12 feet below extreme low water. Below elevation 80 the tower is 70 feet long and 26. feet 8 inches wide with vertical sides, triangular nose and semi-octagonal back; between elevations 80 and 106 the back is semi-circular, the side battering 1/2 inch per foot, and the nosing slopes back to form an ice breaker with its elevation at 73.2. Above this point the front of the tower is semi-circular and sides are vertical; above the 106 both ends are semi-circular and the sides are vertical. A stone balcony at elevation 117 overhangs the substructure. Above this point is the superstructure containing the operating chambers, the balconies with sleeping quarters for the operating force and a room above for the tower keeper. The substructure is of concrete faced with granite on the nosing and below elevation 80 and a facing of sawed Bedford limestone, elsewhere. The superstructure and balcony is of Bedford limestone with interiors of white enameled brick and roof of green slate. The tower was designed in the same manner as a bridge pier to resist wind, ice and current. In the interior below the operating floor are two wells one of these 10 feet x 11 feet with semi-circular end forms an upward extension of the downtake shaft; the other is a rectangular chamber 8 x 16 feet, the two being connected by a 6 x 8 part in a partition wall and filled with rectangular shore gate. All ports open into one of these wells and are closed by 4×6 shore gates located in the walls and operated by hydraulic lifts, all ports opening into the river, except the upper one. These ports are fitted with cast iron gratings which prevent the entrance of large blocks of ice. logs or drift of any kind. As the lowest stage of the river recorded at the Chain of Rocks is 72.8 the seven lower gates at elevation 83 should at all times insure against any water shortage. The difficulties of unwatering the site for the intake tower proved to be quite a problem as it is located on bed rock in about 18 to 22 feet of water with a current of 3 to 4 miles. The method of constructing an angular caisson was adopted. The caisson was built of yellow pine containing approximately 200,000 feet. It is 94 feet long and 52 feet wide, which aliows 4 feet all around for working space. The inner and outer walls of this caisson are about 8 feet apart and the working chamber is 7 feet high. It has two-man locks and supply shafts for four material locks through the deck. The walls of the work chamber are of 8-inch timber and they are continued at this thickness to a point about 9 feet above the deck. From there to the top, which is 26 feet above the cutting edge, the walls are of 6-inch timber. The outside of the entire caisson is sheathed with 2-inch planking and caulked. The deck and sides of the working chamber are also caulked, but not sheathed inside. The entire inside wall of the caisson is sheathed but not caulked. The caisson is braced across from one side to the other with 12 x 12 inch timbers laced together with 3 x 12 inch planks and the walls were also braced apart with 6 x 12 inch timbers and 3 x 12 planks. The new intake tower is located in 20 to 22 feet of water during normal water stages, on a very uneven rock bottom and in a current of 3 1/2 to 4 miles per hour. It was therefore quite a proposition to anchor the caisson. For the up stream anchors four 3 1/4-inch holes, 4 feet deep, were drilled in the rock bottom of the river and four 3-inch eye bolts, 4 feet long with a 10 x 12-inch ring in each, were driven down into the holes. To these, four 1-inch steel cables were shackled. For the west side, two 2-inch eye bolts were used, one for the up stream and one for the down stream ends of the caisson. To each of these, two lines of 5/8-inch steel cable were attached. As all of the rock bottom east of the tower location was covered with sand from 2 to 6 feet deep, holes could not be drilled for the east side anchors, so three 6. x 6-feet concrete blocks were used for each anchor. For drilling the holes a platform was built upon a 22-foot section of 12-inch pipe, imbedded in a 6 x 6 x 3 feet concrete block, and with a 6 x 6 foot plank platform on the top of the pipe. This was set on the bottom of the river with a derrick boat and a stream drill mounted on a tripod was placed on the plank platform. The drill was operated by steam obtained through a steam hose from the boiler on the derrick boat. The drill which was 24 feet long was let down through the pipe, the hole drilled and the drill moved away. After that the eye bolt with a small cable attached was let down through the pipe and driven firmly into the hole with a heavy ram which was small enough in diameter to be lowered through the pipe. The drill platform was then moved and a diver shackled the anchor cables to the rings. When the caisson had been built 16 feet above the cutting edge and caulked it was launched. It had been built on greased ways laid on the decks of the barges and before launching all bracing between the barges was removed. A derrick boat was anchored on each side of the caisson and about 75 feet distant from it. These barges were anchored up stream to the upper caisson anchors and were also made fast to the side anchors on their respective sides. To haul the two barges out from under the caisson on each side, a set of double blocks rove with 1 1/2-inch Manilla line was used for each barge, with the hauling parts of the lines run to the winch heads of the hoisting engines on the derrick boats. One block was hooked into a steel cable sling which passed around the barge under the caisson and the other block was made fast to the timber head on the derrick boat. These lines were hauled taut and the barges under the caisson were then scuttled. A steady strain was kept on the hauling lines and when the buoyancy of the caisson began to relieve the load on the barges the latter slid out nicely with the exception of one which was jammed between the cutting edge of the caisson and a high point on the rock bottom. Even that one gave very little trouble, however, as after the slack had all been pulled out of the anchor cable opposite to it, the barge came out with no more damage to it than a few splinters knocked off. The caisson was then built up to its full height and ballasted with several feet of clay placed on the deck. An endeavor was then made to put air on it but the severe racking it had received from the launching and from its resting on the uneven bottom had opened the seams and with the compressor on hand (673 cubic feet per minute capacity), it was impossible to raise the pressure enough to get inside the working chamber. To tighten it up 2-inch pipes were driven down to the deck through the clay and alongside the vertical timbers where the most leakage seemed to be. These pipees were cleaned out with an air jet and grouted under 40 to 60 pounds pressure. After the grout had set it was possible to get air enough in the working chamber to force the water about 2 feet below the deck, after which men could get inside and stop the leaks. This allowed the air to be raised until the cutting edge was exposed. The greatest pressure carried at any time was 8 to 9 pounds. The rock inside the working chamber was then cleaned off and the edge sealed down to the rock with concrete. This concrete was brought up about 2 feet above the cutting edge and allow’ed to set, after which an effort was made to pump out inside the coffer dam. An examination by a diver showed that there were several large leaks under the edge of the caisson. Men were set inside the working chamber and wherever a leak had been located a part of the concrete was cut out and this usually disclosed a gravel pocket under the concrete. These pockets were cleaned out and refilled with rich concrete, and after that had set the coffer-dam was readily pumped out and the air finally taken off the working chamber. When the loose mud and sand inside the coffer dam had been removed and the surface of the limestone rock cleaned up, a rather startling condition of affairs was disclosed. The rock was criss-crossed with seams and honeycombed with cavities up to 3. feet in diameter and 2 to 4 feet deep and as the rock was excavated it was found that beween the different strata there were numerous channels of all sizes resembling in appearance worm holes. Two of the seams were especially bad and the leakage from them amounted to about 3,000 gallons per minute. A fairly satisfactory rock was found about 4 feet below the surface and after the rock had been excavated to that level the work was shut down on December 28 for the remainder of the winter before starting to place masonry for the tower itself. This was done as the high water and ice were due at any time. Work was resumed in July, 1914. The caisson had stood the wear and tear of the drift and floating ice, also a collision which sunk the excursion steamer “Majestic.” Practically the same pumping capacity was required to unwater the caisson and within two weeks from the time of beginning, the foundation was made ready for masonry. After the masonry had been laid above high water, the river shaft was sunk and the connection with the tunnel made, also the coffer-dam removed. After the caisson had been set on bottom a power barge 30 x 120 feet was moored to the down stream end of it and remained there until the work was shut down. On this were three 60 h. p. locomotive type boilers,one single stage 675 cubic feet per minute Ingersoll Air Compressor, one 7 1/2 KW. 110 volt direct connected lighting set, pumps for water jet, cooling water, etc., and a dry house for the men. The boilers supplied steam for running the compressor, lighting plant and for the pumps which were used for keeping the water out of the coffer-dam. These latter included a 12-inch and a 6-inch direct connected centrifugal pump, a 5-inch belted centrifugal pump and a 5-inch and an 8-inch Emerson pump. Two derrick barges were also used and on one was a 150 cubic foot capacity Westinghouse compound air compressor which was used for grouting and to supply air for the hammer drills with which all drilling inside the coffer dam was done. A steamboat was on the work at all times and two gasoline launches were also used to transfer men and material between the caisson and the loading dock.