But I am convinced that for an annual outlay of $20,000 on the Osage and of half that amount on the Gasconade both rivers may be kept navigable the year round. When this outlay is compared with the enormous sum required for canalization, it should leave no doubt as to which method should be adopted.

It is of course important that the lock and dam now under construction on the Osage be completed, and one lock and dam on the Gasconade just below Priors bend would be the best method of eliminating the difficulties in that stretch of river; but with these two exceptions both streams can be maintained in good navigable condition for the annual outlay above indicated.

ANNUAL REPORT OF MR. F. B. MALTBY, ASSISTANT ENGINEER.

ST. Louis, Mo., June 30, 1898. COLONEL: I have the honor to submit the following report of operations on the Osage River during the fiscal year ending June 30, 1898:

In obedience to instructions received from Capt. H. M. Chittenden, Corps of Engineers, U. S. A., I assumed charge of the work of construction at Lock and Dam No. 1 on July 8, 1897, and on the same day a small force was organized for overhauling machinery and plant preparatory to beginning active operations.

During the year 1896 the excavation for the main body of the lock had been very nearly completed (about 5,317 yards having been removed during the year) and 1,704 foundation piles driven. Active operations were suspended January 2, 1897, on account of high water, and had not been resumed at the end of the fiscal year. During the spring of 1897 sand and gravel in sufficient quantities for constructing the concrete work for the lock and dam had been stored on the bank convenient to the work, contracts had been let for 5,000 barrels of Portland cement and about 135,000 feet of yellow-pine lumber for concrete forming. Cast-iron quoins, frames for filling valves, and a concrete mixing plant had been delivered.

During the month of July, 1897, the centrifugal pumps were thoroughly repaired and new frames made for them, pumping engines were overhauled and put in good condition, fire boxes for boilers were lined, grates raised, and an elevated tank for general water supply built.

On August 6 the pump was started and the cofferdam pumped out. The pumps were run continuously till November 6, a period of ninety-three days, during which time the cofferdam was not allowed to fill with water. On the latter day work was suspended for the season.

The cost of pumping has been a very large item of expense, owing to the unex pected and unusual flow of water encountered, as explained in Annual Report for 1897. Three centrifugal pumps with discharge pipes 8, 12, and 15 inches in diameter, respectively, were used, each being driven by an independent engine. These pumps had a combined nominal capacity of 30,750 gallons per minute. Their actual capacity, as measured, was about 12,000 gallons per minute, pumping against a head of about 18 feet. The latter figure gives the approximate maximum inflow of water, as at times the pumps were barely able to hold the water down. After the foundations and floor of the lock were completed the flow was about 5,000 gallons per minute, and at this rate was quite constant.

In 1896 the pumps were operated from September 18 to January 1, with numerous interruptions, a period of one hundred and two days, at a cost of $8,091.30. In 1897 the pumps were operated ninety-three days, at a cost of $5,243.64, or an average cost of $69.45 per day. The above cost includes labor and subsistence, fuel, repairs, small supplies, and fittings, but does not include first cost of engines, pumps, and

boilers.

The cost of pumping has so increased the cost per unit of the various kinds of work performed that in the following statements of cost the cost of pumping has been given separately.

EXCAVATION.

During 1896 the excavation for the main body of the lock was nearly completed. During 1897 about 6 inches was taken out from the chamber of the lock and almost the entire excavation for the wing walls. During 1897 the excavation was all done by hand, loaded into boxes, raised with a derrick, and dumped outside the cofferdam. Upon pumping out the coffer it was found that sediment had been deposited during high water to an average depth of about 5 feet over the entire area which had been excavated. This deposit, amounting to about 4,000 cubic yards, was all removed through the centrifugal pumps, strong jets of water through ordinary fire nozzles being used to stir it up and wash it to the pumps. The estimated cost of removing this sediment was 16 cents per yard for labor and subsistence and 17 cents per yard for pumping, or a total of 33 cents per cubic yard.

Omitting from the above the 4,000 yards of sediment removed at 33 cents per yard would leave 7,063 yards of excavation, costing $2.25 per yard. A fair estimate of the proportion of general expenses, superintendence, plant, etc., chargeable to excavation would be 75 cents per yard, or a total cost of $3 per yard. It is but fair to note that during negotiations with the contractors for the lock under the original project for an agreement under the revised and present project $3.50 per yard was demanded for excavation.

FOUNDATIONS.

The foundation of the lock consists of piles driven over the entire area covered and from 3 to 4 feet apart, center to center. Under the walls these piles are capped with 10 by 10 inch yellow pine timbers, and these timbers covered with a 4-inch oak floor, upon which the walls rest. Under the lock chamber the piles were cut off flush with the floor. The earth was excavated around them to a depth of 3 feet and filled with concrete. A line of Wakefield sheet piling, consisting of three pieces of 3 by 10 inch yellow pine bolted together, was driven across the lock under the upper miter sill wall and extending under the upper wing wall to its outer or land end. These piles were driven to a depth of from 16 to 18 feet below the lock floor, and it is proposed to continue the same line of piling across the river under the dam. Cost and amount of pile driving.

Cutting off, removing, and storing 1,907 pileheads cost $0.683 apiece. Seventyeight thousand seven hundred feet B. M. of timber was placed in foundations at a

A cost of $8.14 per thousand. Pumping cost, $274.68, or $3.49 per thousand.

CONCRETE WORK.

The walls and floor of the lock are of concrete deposited in place. The floor of the chamber is of natural cement concrete deposited to a depth of 3 feet around piles, which were cut off at the level of the floor. This concrete is made of the folfowing proportions: One barrel of cement, weighing 280 pounds net; 7.5 cubic feet sand, and 14 cubic feet gravel, measured loose. The floor contains 1,395 cubic yards. The walls are built of American Portland cement concrote, the main body of the walls being made in the proportion of 1 barrel cement, weighing 380 pounds net, to 14.6 cubic feet sand, and 32 cubic feet gravel, measured loose. To insure a somewhat smoother appearance, the exposed faces of the walls for an average thickness of about 8 inches is composed of a proportion of 1 barrel cement, 7.3 cubic feet sand, and 9.5 cubic feet gravel. It is estimated that this facing contains 360 cubic yards of concrete. The interior and bottom of a part of the land wall contains 970 barrels of Milwaukee cement, which was on hand, and which was mixed in the proportion of 1 barrel cement, 7.3 cubic feet sand, and 16 feet gravel.

The breast wall is also built of Milwaukee cement concrete, in proportion of 1 barrel cement, 7.3 cubic feet sand, 17 cubic feet gravel.

The natural cement used was purchased in 1896 from the Milwaukee Cement Company, of Milwaukee, Wis., and cost, f. o. b. Osage City, $1.02 per barrel of 280 pounds net.

Portland cement was purchased under specifications requiring 99 per cent to pass a No. 50 screen and 88 per cent to pass a No. 100 screen. It was required to have a tensile strength for neat cement in seven days of 500 pounds per square inch and 600 pounds in twenty-eight days; for 1 to 3 of sand in seven days, 160 pounds, and in twenty-eight days, 240 pounds. Every tenth barrel received was tested and weighed. A contract for 5,000 barrels was entered into with the Atlas Cement Company, of New York, at $2.46 per barrel of 380 pounds net, f. o. b. Osage City. Owing to the failure of the contractors to deliver cement as fast as required, their contract was reduced to 3,750 barrels, and 1,100 barrels of Portland cement was purchased in the open market in St. Louis at an average price of $2.86 per barrel, f. o. b. Osage City. All cement was received in barrels, except one lot of 400 sacks, equal to 100 barrels. The sand used was dredged from the Missouri River from 8 to 10 miles from the lock. It is clean and sharp, but contains rather too large a proportion of fine material for the best results; about 30 per cent will pass a No. 50 screen. It cost, stored on the bank at the lock, 80 cents per cubic yard.

The gravel used in the concrete was taken from the Osage River within half a mile of the lock. A ladder dredge was used, which elevated the material into revolving screens, where it was washed and thence delivered on a barge. It is peculiarly well adapted for concrete work. It is very hard, of a flinty nature, and though the facets are smooth, the stones are irregular in form. It cost, stored on the bank at the lock, $0.571 per cubic yard. The following description of the methods employed in the concrete construction may be of interest.

FORMING.

The forms for the walls were made of frames or bents of 8 by 10 timber, braced thoroughly with 6 by 8 and 7 by 9 braces. The posts were tied together at the top with 4 by 10 oak pieces. These pieces also formed the supports for a track, on which the concrete was brought from the mixer. The bents were sheeted or lined with 4 by 10 plank, which was sized at the mill to 34 by 9 inches. All the timber used in forming was long-leaf yellow pine, and the posts for the 35-foot wall were single sticks 36 feet long. The forms were divided by bulkheads into sections from 20 to 24 feet long and alternate sections filled. The bulkheads were then taken down and the intermediate sections filled. A section was usually filled without interruption, though in a few instances work was suspended from 12 midnight to 7 a. m.

Timber enough for forming only one-half of one wall was provided. On comple tion of a portion of the wall the forming was taken down and again erected in position for another portion. The timber was thus used on an average of four times, and thereby reduced the cost of forming per cubic yard of concrete. It was found that the loss and damage from use and handling was very small.

MIXING AND PLACING.

The mixer used was a cubical one with 4-foot sides, and was erected at the center of the length of the lock and on the edge of the bank about 40 feet back from the land wall.

Tracks were laid to the sand and gravel piles, which were in convenient proximity. The sand and gravel were loaded by hand into 2-foot gauge side-dump cars, and in quantities sufficient for one batch of concrete.

The cars from the sand pile passed the cement house, where they received a barrel of cement, the size of the batches being that required for one barrel of cement. The cars were run to the mixer, and the sand, gravel, and cement were dumped into a box which was hoisted and dumped into the hopper above the mixer.

At this point the water, which had been drawn and measured into a tub standing on the hopper platform, was introduced and the whole mixture drawn into the mixer below, which was then revolved fifteen times. It was then dumped into a car standing on a track below. This track was on a level and connected with the track on the forms. The car was run out to the section of the wall being built and dumped into place. It was spread into layers of about 6 inches and tamped thoroughly with rammers weighing 22 pounds each.

The mixer was run by a belt from a pulley put on a Mundy hoisting engine in place of one of the winch heads. The same engine also hoisted the material ento the hopper platform. In mixing it was the practice to use all the water possible and avoid quaking under the rammers. The capacity of the plant was only limited by the progress of the tampers, and it was found that in the size sections built only eight to ten tampers could be used economically at one time. From 12 to 15 yards were ordinarily placed per hour.

The following tables give in detail the cost of the total quantity of concrete made:

Cost of 6,708 cubic yards concrete built in 1897.

Cost per yard.

6,010 yards gravel, at $0.571 per yard.
3,021 yards sand, at 80 cents yer yard.
3,000 barrels Milwaukee cement, at $1.17 per barrel at lock; 210 barrels Mil-
waukee cement, at $1.02 per barrel at Osage; 1,110 barrels Portland
cement, at $2.86 per barrel at Osage; 3,443 barrels Atlas cement, at $2.46
per barrel at Osage

Material used in erecting mixer, tracks, etc.

Storing cement

Fuel.

The above table gives the total cost of all concrete made during the year, and is made up of the items given. Of the above amount of concrete, 2,030 yards were made of natural cement, costing as follows:

There were 4,678 yards of Portland cement concrete, costing as follows:

It will be noticed in the above that the item "Mixing, placing, forming, etc.," $2.16 per yard is the same in each case. This item is taken from the preceding table of general cost, as it is impossible to give this cost exactly for each kind of concrete, and there seems no reason why it should vary materially for the two kinds of concrete made.

Owing to the exhaustion of funds active operations were suspended on November 6, and all the force not required for the care of the plant was discharged.

The following work yet remains to be done to complete the lock: About 1,650 yards of concrete; all the machinery and appurtenances, consisting of gates, filling valves, maneuvering gear, miter sills, etc.; back filling, surfacing, fencing, riprapping bank, dredging, etc. Nothing has been done toward the construction of the dam.

There is submitted herewith a tracing showing the arrangement of the plant at the lock.

During the winter and spring months of this year the detailed plans for completing the lock and dam have been worked out. A very careful comparative study of the various types of movable dam has been made, but nothing has been learned that tends to change the belief that for the conditions existing and required for the Osage River the drum weir, as proposed by Capt. H. M. Chittenden, Corps of Engineers, U. S. A., best meets the requirements of the situation. The estimate of the cost remains the same.

During the months of December, 1897, and January, 1898, a small survey party under W. P. Pollard, overseer, was organized and sent into the field. A duplicate line of levels was run from Lock No. 1 to the mouth of the Osage River below Bonnots Mill, connecting with the water surface at the upper end of the proposed canal across Osage Point; also with the water surface in the Missouri River at the lower end of the canal.

The party then proceeded up the river, measuring the low-water slope over the shoals. A bench mark was set at each shoal and levels run to the pool above and below the shoal. These bench marks will permit the reduction of soundings, taken in future surveys, to low-water depths. This party also erected a new gauge at Tuscumbia. The examination of shoals was carried up as far as Warsaw, 171 miles from the mouth, and the party disbanded. An inspection of the results of these examinations shows that the slope or fall over 90 shoals was determined. The total fall over the shoals is 111.68, or an average of 1.24 feet in each shoal. The total fall of the river between Warsaw and the mouth is 116.55 feet. The total length of the shoals is 28.7 miles, or 95 per cent of the fall is concentrated in 17 per cent of the distance.

The examination of these shoals and a comparison made with surveys made in 1880 show them to be possessed of considerable stability, more in fact than would be thought possible by those familiar with the conditions on the Missouri and Mississippi rivers. The condition of many of the simple and inexpensive low, loose stone dikes or dams used for contracting the flow over the shoals emphasizes this fact. Some of these dams are from twenty to thirty years old, and many of them show no injury or signs of deterioration. The majority of them have done and continue to do work expected of them, and have brought about and maintain the conditions required-i. e., the contraction and maintenance of a channel over the shoals restricted in width to give with the available flow a depth necessary for navigation purposes.

At a number of the shoals where the dams themselves are intact and a channel maintained along their length, there has been a shoaling at their head and foot; in these cases the remedy is simply extending them for a few hundred feet. In a few instances breaks have been made in the dams and their efficiency destroyed. A comparatively small amount of rock will repair them.

In conclusion, from a personal examination of the stream, I am lead to believe that not only is it capable of radical improvement without canalization, but at a cost which is insignificant in comparison with the cost of canalization, and very small in comparison to the importance of the stream as an outlet for the country through which it flows.