4

THE BISMARCK

BRIDGE

which this country has been settled of a flood of equal magnitude, the scarson the trees in the bottom lands indicate that equally high floods have oc-curred. During such a flood, large cakes of ice frequently turn up edge-wise, rising 15, and perhaps 20, feet above the natural surface of the water.It seemed that it would be unsafe to build a bridge the lower chord of whichshould be less than 20 feet above the highest observed ice flood ; this wouldbe 33 feet above the assumed high water of navigation, which is only 17 feetlower than a high bridge must be placed. Add to this the fact that the nar-row openings of a draw would soon become clogged with ice, so that provi-sion must be made for the passage of the entire river through the fixedspans of the bridge, and that protection piers above and below the pivotpier must be built of sufficient strength to stand the pounding of an ice floe,and it was evident the cost of a low bridge would become very much greaterthan that of a high bridge.

It was considered very important not to raise the track across the bot-tom land between the west approach and Mandan any higher above thenatural surface of the ground than was absolutely necessary to secure surfacedrainage. This low bottom land must be regarded as the natural safety-valve through which the river will discharge in case a gorge forms at thebridge site, or, in fact, anywhere between the point where the river leavesthe west bluff, a few miles above Mandan, and where it leaves the east bluff,opposite Bismarck . If a gorge forms anywhere within this limit, which ex-tends from three miles above to one and a half miles below the bridge, thisgorge is liable to dam up the river until it can escape over the bottom land onthe west side ; if there is a free passage over this bottom land, the rise will bedetermined by the depth of water required to make a free passage over thebottom land; if, however, this passage is obstructed, the obstruction mustbe balanced by a corresponding increaste in the height of the flood; an em-bankment across the bottom land might therefore raise the flood level byan amount equal to the height of this embankment. Any such increaseof flood level would, of course, add to the general destructiveness of theflood, which would spread out, up the valley of the Heart through Mandan,and prove very destructive to that unfortunately situated town. It was,therefore, considered important that this low tract should be left as a safety-valve until a few years’ experience might show what water-way would beneeded in this place, when, if the business of the road justifies it, the trackcan be raised across the low bottom land, and a bridge built to provide forthis high water discharge. Such a bridge would not be a very expensivestructure, but must be something better than an ordinary pile bridge, as it

must be strong enough to pass a large amount of floating ice ; from presentindications, it would seem that it ought to be at least 2,000 feet long, inspans of not less than 60 feet, and with the lower chord placed at an eleva-tion not lower than i65o.

The interruption to traffic which will be caused by the overflow of thetrack across the bottom land, would not be likely to last more than two orthree days, and it would usually happen that traffic would be interruptedfrom other causes at the same time. There is no danger that such a tempo-rary diversion of the discharge of the river will lead to a permanent changeof channel, as the ground is always frozen hard at this season of the year,even though the surface may have thawed out for a few inches.

The bridge affords, practically, a water-way of 1,200 feet, obstructed bytwo piers. The actual water-way, however, excepting during the extremefloods which overflow the dike, is only 1,000 feet, being limited to the dis-tance from the east end of the dike to the east bluff. Although this widthis about one-third of the high water width of the river at this place beforethe bridge was built, it is ample to accommodate the whole discharge of theMissouri River . Whenever the channel of the Missouri has remained fixedfor a long period of years, the width of the river will be found not to exceed1,000 feet. At Fort Abraham Lincoln, 5 miles below the bridge, it is but900 feet. At Lexington, Mo. , about 1,200 miles below Bismarck , and belowthe two great tributaries, the Platte and the Kaw, the river was for manyyears but 900 feet wide. At various intermediate points, where the channelhas been permanent, widths of less than 1,000 feet are observed. The highwater width selected by the United States Engineers as most favorable to themaintenance of a permanent channel between Sioux City and the mouth ofthe Platte, is 85o feet. If it had not been for the ice floods, a width of 800feet would have been ample at Bismarck , and one of the 400-feet spanscould have been omitted. To provide, however, for the full dischargeof the ice, it was thought best to make the water-way 1,200 feet at thebridge line.

IV

RECTIFICATION WORKS

The rectification of the river at the bridge site proved a matter ofgreater difficulty than was anticipated. At the time the preliminary exami-nations were made, in the spring and summer of 1880, although the river

was scattered over a width of 3,000 feet, the principal channel was near theeastern shore, and the testimony of river men was to the effect that it hadbeen there for quite a series of years. The works to be used for narro'wingthe river consisted in a dike, running from the western shore on a due eastand west line, and terminating 1,000 feet from the eastern bluff; the direc-tion of this dike was to be nearly at right angles with the river, so that theforce of the current would be received only at the end, but inclined downstream very slightly (9 0 30'), so that any current striking it at a distanceback from the end would be deflected towards the eastern shore line, ratherthan in an opposite direction. To make this dike effective, it was necessarythat it should be continuous from the western shore to the end. It was pro-posed to finish the dike at an elevation of about 1630, so that the summerhigh water of ordinary years would just pass over it, as it was thought thatit would be buried in silt earlier, if finished at this height, than at a higherelevation.

Through some" unfortunate misunderstanding, work on this dike wasnot begun as early as was intended, and, during the first period of delay,the main channel of the river shifted to the west side, leaving only asmall channel, too shallow for navigation., on the east. Instead of adoptingthe bolder course of carrying the dike directly across the low water channel,which would have compelled the small easterly channel to deepen itself, thework was begun on the east side of this channel, and a foundation matbuilt of brush, wired together, extending from the east edge of the westerlychannel to within 1,000 feet of the easterly bluff; this mattress was 140 feetwide and about a foot thick, and built with a return, down stream, at theeasterly end ; it was laid out exactly in the proposed location for the dike,but unfortunately had no connection whatever with the west shore. Theeasterly channel gradually silted up, until, in the middle of October, therewas practically no water running anywhere east of this foundation mattress,a dry sand-bar extending from the easterly shore to the channel west ofthe dike formation. The situation at this time is shown on the map onPlate 2.

When the work was placed in my hands, early in January, 1881, thesituation was very unsatisfactory. The foundation mattress for a dike, whichwas to hold the river on the east side, had been constructed, but the entireriver was running on the west side of this foundation ; there was no waterrunning where the bridge now stands ; moreover, the ice was likely tobreak up in a little over two months, and unless a vigorous effort was made,it would tear to pieces all the work that had been done. Logs had already