*Money cost of this work included in cost of material. $194,815.566,600 (linear feet) $29.517 per linear foot. *NOTE. $14,318.38 is taken up in the above statement as the estimated cost of towing for stone received from the general service.

LAKE PROVIDENCE REACH.

This reach extends from Carolina Landing, Miss., 517 miles below Cairo, to Island 95, a distance of 35 miles.

Before its improvement was undertaken a depth of but 4 feet at low water had been reported on some of the crossings. (Report of Mississippi River Commission, 1883, page 408.)

The plan of improvement has consisted in the construction of permeable dikes, to close chutes, and contract wide portions of the river, and of bank revetment, to prevent caving.

Strong levees have also been constructed on both banks through this reach. Dikes have been built at Duncansby Crossing, Cottonwood, Mayersville, Elton, Baleshed, and Stack Island, and are fully described in previous reports of the Commission. Their effect has been beneficial. A deposit of from 6 to 18 feet was formed behind them the first season, chutes were closed, and the river at low water contracted and confined to a single channel. The least depth that has been recorded on crossings in Lake Providence Reach since their construction, has been 7 feet. The following are the soundings obtained at crossings during the extreme low water of the past season:

At a higher stage, but 74 feet was reported on the Ben Lomond crossing, by pilots. The extensive caving of the banks, however, which has extended from Louisiana Bend through this reach, has successively flanked and destroyed those dikes, until the Baleshed and Stack Island systems alone remain, while opposite them the caving is removing the channel of the river, each year, farther from their sphere of action.

Bank revetment has been undertaken at Mayersville Island and Louisiana Bend. The first attempts were unsuccessful. The caving continued above the points protected, and flanked and destroyed the works. But the revetment constructed at Louisiana Bend in 1889, 6,024 feet in length, has successfully withstood the action of succeeding freshets.

During the past season this revetment has been extended to Station 1,124, 5,224 feet having been constructed. As the mat boats of the district were in use at Ashbrook Neck and Greenville, two large barges were utilized for the purpose. The width of mat was 270 feet, as wide as could be constructed upon them. They extend to a depth of water, at the time of construction, of from 65 to 85 feet. These mats were connected to the irregular bank by twelve pocket mats. The upper bank revetment was a riprap of stone, 10 inches thick, to a 27-foot stage, and extended from the work of 1889 to the mouth of Old River. The mouth of Old River was protected by a revetment of brush and stone. The work cost $26.49 per linear foot.

A detailed report of the season's operations will be found in the accompanying report of Assistant Engineer E. C. Tollinger, who was in local charge.

REPORT OF ASSISTANT ENGINEER E. C. TOLLINGER ON OPERATIONS AT LOUISIANA BEND, LOUISIANA, FOR THE SEASON ENDING JANUARY, 1892.

Construction.-Work was completed from Station 0 (▲ Owens) to Station 60 in 1889. No work was done in 1890. An examination made in September, 1891, of the work done in 1889, showed it to be in good condition, except about 200 linear feet of the extreme lower end that had been injured by the action of the water cutting the bank away. Work was begun September 17, 1891, and completed January 22, 1892, and extended from Station 58 to Station 110+24 feet, a distance of 5,224 linear feet. The entire plant was moved to Greenville, Miss., the last arriving there on January 25, 1892.

General plan of the work.-A deep-water mat 270 feet wide was first made and sunk on a line that reached to about the deepest water. These were overlapped with pocket mats north of Old River, and connected with brush revetment work across the mouth of Old River, extending to the top of the bank or a 15-foot stage of water. North of Old River, and extending from the line of the pocket mats, the bank was riprapped, 10 inches thick, with rock on a slope 5-4 and 3 to 1, to a 27-foot stage of water. Seven-tenths of cubic yard of rock per square was used on the deep water, and 14 yards per square on pocket mats and brush revetment.

Towboats.-The steamer Vidalia (with the exception of five days at Greenville and Offuts, while the steamer Osceola and tug Parker were having burnt boilers repaired) was constantly employed. A single crew was employed until the 9th of December, when it became necessary to put on a double crew. The unsound condition of her timbers above water line involved much care upon the captain and engineer to keep her ready at all times for active duty.

The steamer Vedette is the most economical for harbor work, if the conditions are favorable, than any of the towboats in the employ of the Third District. She is exclusively a harbor boat.

Grading (hydraulie).-Grading was commenced September 15 and finished December 21; 3,390 linear feet was graded. Our first experience was with a grader that had been constructed from the several parts of broken down machinery. After repeated breakdowns the weak parts were replaced with new ones, but the season had so far advanced that it was necessary to send Grader No. 1 from Greenville Harbor, after completing her work there, to take the place of Grader No. 77. She arrived with a double crew, ready for work, November 13, 11:30 p. m. The water cylinders were leaking, and the pressure had to be reduced from 160 to 125 pounds per square inch. Without the foreman in charge of night force has had special training in this particular line, the bank graded is left in such a bad condition that little or nothing is accomplished.

Deep-water mats.-These were begun September 21 and the last one sunk December 22. Five thousand three hundred and sixty linear feet of 270 feet wide was made and sunk. Eight longitudinal g-inch wire cables were used the full length of the mat. The -inch transverse wire cables were used every 50 feet. The shore lines being very irregular, the mats were built straight from point to point that projected into the river, that they might reach about out to the deepest water. The mat boats were only 270 feet long. The depth of water was from 65 to 85 feet on the outside of the mats, at a low stage of the river.

Wiring the mat.-This is one of the most important parts of mat construction. There is so much depending upon this part of the work that a thorough examination should be made of every square before ballasting.

Cables are intended to strengthen the mat and to prevent breaking while sinking. The longitudinal ones answer a good purpose. The transverse ones are of doubtful value after the mat has been sunk. As holding-in cables they are of little or no value should the mat start out.

Ballasting should receive careful supervision. Careful notes should be made of the direction and force of the current, and the work done in accordance with the difficulties you are expected to meet.

Sinking.-Stone should be put on the mooring barges to sink the head of the mat. This should be done at the time the mat is being ballasted. Every little detail should be computed and every precaution taken that will insure success. You are not warranted, under the most flattering conditions, of taking chances.

Pocket mats.-The hard, projecting parts along the shore made it necessary to construct 12 mats in the first 3,000 linear feet. These were, with the exception of 3, built in lengths of 270 feet, and vary in widths from 75 to 225 feet.

I

Riprap (stone).-The bank for the first 3,390 linear feet was graded to a slope of 5, 4, and 3 on 1, and covered with rock 10 inches thick to a 27-foot stage of water. am of the opinion that 6 inches, well laid, to a and stage, is sufficient.

Brush was obtained close by. In size it was too large, but the want of barges did not warrant making a change. By constant urging the contractor did fairly

well, but the little shortage from day to day added to the cost of the work. contractor was at all time promptly supplied with barges.

The

Poles.-The first poles were loaded on barges 75 miles up the river. The quality was in accordance with the specifications. The scarcity of barges and the great distance prevented promptness at all times on our part.

Rock.-The rock was unloaded on the bank at Ashton, La., prior to construction. Most of it was unloaded at points not accessible at extreme low water, thus materially adding to the cost of loading. The unfavorable conditions were improved at a 12 and 15 foot stage of water. Twenty-five thousand seven hundred and four and eight-tenths cubic yards were loaded on barges and towed to the work. About 70 per cent loaded with barrows, 30 per cent with rock cars; these proved very satisfactory.

Conclusion. The work is, I believe, the best of the kind. The results were obtained by constant and persistent energy, experience, and thoughtfulness on the part of the principal employés.

I make special mention of the valuable service rendered by W. M. Keller, in charge of the office; C. H. Wilson, superintendent of construction; N. Stephany, in charge of the mattress work, and C. A. Lacy, timekeeper.

Itemized accounts, tabulated statements showing amounts charged and expenditures and the cost of the different kinds of work, and progress sketch, accompany this report. (Omitted.)

GENERAL REMARKS UPON BANK REVETMENT.

In the following table is given the amount of revetment which exists in the third district:

In addition to the above, about 3,500 feet of Greenville Front is protected by spur dikes. It will be noted the large proportion of this work which was constructed during the last season; but it may also be stated that if any material improvement in the low-water channel is soon to be expected in the 220 miles of river in the Third District, this rate of progress must be largely exceeded.

The amount of work done last season would have been an impossibility if forty barges and the steamboats Etheridge and Vedette had not been temporarily transferred from the general service to this district.

With this addition the plant was employed to its utmost capacity. Barges were converted into mattress boats, and a grader extemporized from the machinery of a condemned pile-driver. The graders were worked night and day, and while for some time six towboats were in commission, it was necessary on the three largest to employ double crews. Ashbrook Neck suffered especially from lack of plant, while at Greenville, it was with great difficulty that stone could be obtained with sufficient rapidity to complete the work.

The following is a list of the aggregate amounts of the principal items of material expended during the season.

About 7,700 cubic yards of stone were towed a distance of 248 miles, 5,700 cubic yards a distance of 40 miles, some of the brush and poles a distance of 90 miles. The necessity of accumulating large amounts of material, particularly of stone, near the work before beginning operations was again demonstrated. During the months of December and January all the stone which could be delivered by three railroads was accepted, and was insufficient to meet the demand.

The futility of depending upon delivery by barges from the upper Mississippi or

Ohio rivers was also again shown, but 2,429 cubic yards being received of 10,000 cubic yards which were to have been delivered. The failure at a critical period of this expected supply was the cause of the difficulties experienced at Greenville.

The reduction in cost of work this season is also very satisfactory. The estimated cost per running foot of the revetment constructed at Louisiana Bend in 1889 was over $40. Its extension this season was at a cost of $26.45 per linear foot. It should be noted, however, that across the month of Old River the bank only extended to a 15-foot stage. I estimate the cost of the 3,380 feet between the old revetment and the mouth of Old River at about $28 per linear foot. It is believed that a further reduction in cost can be obtained.

In the form of revetment now adopted about 6 cubic yards of stone are used per running foot. The cost of stone delivered on the bank at Louisiana Bend, either by the general service or by contract, has heretofore been over $2.25 per yard.

After the close of last season's work the experiment was tried of towing stone with the plant of the third district. Bids were received for stone loaded on barges in the Little Red, a branch of the White River, at 49 cents, 60 cents, and 65 cents per cubic yard. As these quarries are over 200 miles nearer the work than those on the Upper Mississippi and Ôhio, a reduction in the cost of towing is anticipated, though the tow boats of the third district are not well adapted to the purpose.

In a fine, sandy soil, such as was found above Greenville, the 10 inches of stone which were substituted this season for the upper bank revetment of brush and stone previously employed appear to protect the sand from the river currents, but they do not afford an adequate protection, either from wave action, caused by a strong wind blowing on shore, nor from the action of rain water flowing down the slope, during a low stage of the river. In both these cases the water has sufficient force to carry through the interstices of stone large quantities of the sand, forming gullies and cavities, into which the stone settles. An underlying layer of gravel, it is believed, would check this action, and if covered by a thin layer of stone would be in no danger of being carried away by river currents. At Greenville satisfactory results were obtained for a short length of bank by using a revetment of quarry spalls and the refuse of the stone pile.

DELTA POINT AND VICKSBURG HARBOR.

The works for the improvement of Vicksburg Harbor consist of the revetment of Delta Point, and of the dredging of a canal and basin in front of the city.

The object of revetting Delta Point was to prevent its further recession, which threatened to largely increase the width of the bar in front of the town. Between 1878 and 1884, 10,700 feet of bank were revetted. This work has not been repaired for several years. During the last low water an examination was made, which showed that the subaqueons mats are still in good condition, but several breaks were observed in the upper bank revetment, due to the rotting out of the brush. It is proposed, during the next working season, to repair these breaks, replacing the decayed brush with a layer of 10 inches of stone. Above the portion protected caving is slowly taking place. It will ultimately be necessary to further extend the revetment up the river, to prevent the flanking and destruction of the work constructed.

The work of dredging the canal and basin in front of the city was begun in 1887. In the project then adopted it was proposed to inclose the canal and basin by a dam constructed across Centennial Lake from the city to De Soto Island, and along the island parallel to the canal, for the purpose of limiting the flow of water into the basin during floods, and thus reducing the annual deposit of sediment. A permeable pile dike was first constructed along the axis of the dam, and as the dredging progressed the dredged material was deposited on both sides of the dike, so as ultimately to form a solid embankment of earth.

The canal was to have a bottom width of 75 feet, and slopes of 1 on 3 on the land side and 1 on 5 toward the river. The basin was to be 300 feet wide by 1,700 feet long, with slopes the same as the canal.

At the close of the last fiscal year the canal had been dredged to a depth of 48 feet on the Vicksburg gauge, 954,514 cubic yards of material having been removed. Dredg ing was then in progress under a contract with the Alabama Dredging and Jetty Company, at 11.9 cents per cubic yard measured in scow, the cut being inade to zero on the gauge.

Work was suspended September 7, due to the low stage of the river. A survey then made showed a fill averaging 5 feet through the canal, which had principally occurred during a flood of the June preceding.

Dredging was resumed February 8, 1892, and on May 15 one cut 40 feet wide had been again excavated through the canal to the zero of the gange, and 3,000 linear feet of a second cut. The amount of material that has been excavated during the year has been 315,079 cubic yards.

It has been proposed by Capt. J. H. Willard, Corps of Engineers, in charge of the improvement of the Yazoo River, to divert that river through Centennial Lake. Should that project be approved by Congress, a modification of the existing plans for the improvement of Vicksburg Harbor will be necessary.

The following report of Assistant Engineer H. St. L. Coppee gives in detail the work done during the season:

REPORT OF ASSISTANT ENGINEER H. ST. L. COPPEE.

Vicksburg Harbor, Mississippi.—At the time of submitting the last annual report the dredge Herndon, of the Alabama Dredging and Jetty Company, was at work in the harbor operating under the contract of January, 1891; the price being 11.9 cents, the requirement 4,000 cubic yards per day; the limit of contract December 31, 1891, and the amount of money available as per specifications $75,000, the Government reserving the right to reduce it to $60,000.

Up to May 1, 1891, the dredge had excavated on the contract as measured in scows 108,130 cubic yards. This excavation was to the zero plane, with the expectation that the slopes would slough down, leaving a cut to about the +3 foot plane; up to that time one cut 40 feet wide to zero had been taken out from Range 0 to 55 of canal, and one cut of outside slope from 0 to 10 (also in the canal).

The dredge Herndon continued to work in canal and basin to same depth until September 7, when, owing to low water, she was withdrawn; a long interval of low water occurring she was laid up for repairs, her tug and scows being sent to Mobile for use on other contracts. The contract time having elapsed, an extension was obtained and the dredge resumed work February 8, 1892.

At close of season's work September 7, 1891, the dredge had completed two channel cuts in canal and basin, to the zero of the gauge, and taken one cut from outside slope, all but 1,000 linear feet that had been abandoned on account of falling water. Since that date she has excavated one entire cut through canal to zero, and half of another, and is now continuing the same to the river.

The excavation measured in scows up to date on present contract is as follows:

On the preceding contract (1890) the canal had been cut approximately to +8 feet on the gauge, and, as stated above, the purpose was to excavate the canal and basin on the present contract to an available depth of +3 feet on the gauge. In order to obtain this depth, as most of the slope was uncut, it was necessary to excavate to the zero plane to allow for sliding in of bank. During September (1891), after the dredge had been taken from harbor, the ranges were sounded throughout the entire canal and basin, careful levels being run over cross sections where it was impossible to sound, and the results plotted and compared with profiles of former years. These results indicated that a very considerable fill had taken place since cross sectioning in July, 1890, just after season's work, and in February, 1891. A report of these results was forwarded to your office October 14, 1891, which stated that "since 1888 up to the season of 1891, the fill in the harbor (canal and basin) had been comparatively slight, and could be accounted for by the slide of the sides of excavated channel and normal deposit for the aggregate rises during the seasons, as stated in reports and estimates submitted at the time." But the change in 1891 was in considerable excess of former experience. The report further states: "The excavation as determined by using ordinates between the bottom lines of February, 1891, and September, 1891, was 162,361 cubic yards, which is but about 50 per cent of the material actually removed by dredge as measured in scows, or about 60 per cent of the material measured in place."

Taking total cost of the season's work of 1891 as $39,413.28, we get the cost per cubic yard measured in situ=$39,413.28-162,361 approximately 24 cents, or in other words, the excavated channel, instead of being to the +3 foot plane, as contemplated, and costing about 14 cents measured in situ, averages between the +5 and +6 foot plane, costing approximately 24 cents. Between the middle of June and July of 1891 there was a 10-foot rise in the river of a month's duration, and of