It must not be supposed that the transverse section of any flowing body of water is perfectly horizontal on its upper surface, because it will be found that immediately over the portion of the section where the greatest velocity prevails', the surface will be slightly raised abovo the level of the shallower portions of the stream. In small rivers it is not possible to distinguish this species of ettrelevation ; but in larger I ones it is at times strongly marked. A phenomenon of more import ance to the stability of banks consists in the existence of a zone of still water, comparatively in close proximity to the shore in consequence of tho greater resistance to the flow through the friction of the banks ; and in this part of the stream may also be observed a series of small eddies, produced by the impulsion of the current. The principal direc tion of these eddies appears to be directly opposed to that of the main stream ; and moreover it will be found that when any obstacle is offered to the onward flow of the water, the latter will in the first place heap up against this object, and finally give rise to a dis tinctly marked counter current, technically known as a back.water. If the obstacle should extend across the whole stream, the upper surface of the water will assume a sensibly parabolic longitudinal section, the apex of which will be situated at a considerable distance up stream, and the axis upon the deep waterline. If the obstecle, however, should only be upon one bank, the back-water will not extend for any great distance; but it may frequently happen that the effect of tho counter current would be so great as seriously to endanger the stability of the bank itself. The spurs, which are in many cases introduced for the purpose of defending the shores of rivers, when they extend far Into the stream, produce this particular effect; and they may often be observed to give rise to a series of eddies, currents, and back-water, which wear away the shore at the points where they are joined to the land, on the upside especially. Another effect produced by a projec tion into tho line of flow of a stream in that the portion of the current. which does not go to form the backwater, is deflected from its course, and may perhaps be made to impinge in a dengerous manner on the opposite bank. At any rate the current at the foot of a spur intc) main stream is sure to deepen the bed of the river at its foot, and to carry the channel towards the convex side of the bank; indeed this law holds universally, and in rivers it is always found that the deep water channel follows the convex aide of the meanderings, and that the channel thus naturally tends to straighten itself by the more rapid abrasion which takes place against the projections. From these re marks it would follow that whenever it is desired to regularise the conrees of rivers it is preferable to do so by means of longitudinal parallel walls), rather than by spurs ; and that the direction of the new channel should be made as far as possible rectilineal, or at lest with curves of very large radii. 31. Dcfontaine observed on the Rhine that elbows of 8300 feet radius produced but small effects upon the bed, or banks, of that river, notwithstanding that they were composed of easily transported materials; and it may be inferred from this that a bend whose radius should be not lees than twelve times the width of the maximum water line may be admitted in permanent works of thin description. The above reasoning of course applies in the most exclu sive manner to rivers whose flow is in only one direction; when the total action is brought to bear upon them, other conditions of regimen are developed to which attention will be called hereafter.
The nature of the materials of the bed a river may flow over will affect the chemical quality, and therefore the value, of its waters for such lenses* as irrigation or water supply ; and this, too, whether the material, taken up be AO taken in solution, or merely in mechanical In Warts Su ret.T, attention will be called to the chemical steel of this inquiry ; but at present it is advisable to dwell upon the merelv mechanical effects produced, for the outlined of the beds of rivers depend mainly upon them. Now the lobgitudinal profile of a river almost always forms a speciem of parabolic curve, whose apex is in the high lands, and whose base is at the embouchure ; so that the rate of flow is ovater near the source than it is near the emlxmehure, the transporting and the erosive powers being also diminished in the same proportion. Ily the experiments of Du Bunt we learn that the following materials will be carried forward by waters flowing with the named in connection with them in the table :— It thence follows that when the velocity of the river attains any of the above rates, it will be able to attack and remove by erosion the materials of its banks, which would remain in suspension at those velocities. In designing a system of river defences, however, attention must be paid exclusively to the maximum, not to the arerage velocity of the stream; because in flood time net only is the volume, but also the velocity, greater than at any other period, and the materials of which the banks are formed nmst be able to resist the transporting power of the floods. However, in consequence of their variable trans porting power, rivers in their normal state flow over materials gradually decreasing in volume as they leave the steeper parts of their courses and approach their etutialla; and as those materials necessarily become nt last very fine, especially in long flat plains, they offer little resistance to occasional floods, as may be observed in the deltas of such rivers as this Nile, Danube, Ganges, dtc., which are composed of the fine mud
le-ought down from the uplands, and in which the course of the stream continually shifts.
Sonic curious effects may be observed to take place with regard to the deposition of alluvial matters near the junction of streams draining areas of country situated under marked differences of thermemetrical off hydrographical conditions; because the floods (or as they are technically called the freshen) frequently occur at distinctly different epochs, anti thus, by reason of the freshet of one affluent pending back the other, a deposition of the matters in suspension in the retarded stream will be produced. It often happens that, under such circum stances. the accumulation of alluvial matters takes place in a manner to direct the loud current against the banks in a direction essentially different from the main line of flow, the effect of which would be to divert the river from its original bed, to a greater or lesser extent, according to the resistance of the banks. There is another law with respect to the mode in which the alluvial matters brought down by Meal. or freshets takes place, namely, that when the waters overflow their banks they throw down those natters in the order of their specific gravities, the heavier ones being thrown down on the edge of the deep stream, where the velocity is greater, and the lighter ones being thrown down on the edge of the shallow water where on account of the friction the velocity is the leaat. The transverse section of a valley subject to freshets wilt be in fact something like the sketch here given. The banks will form slightly elevated ridges close to the water'. edge, and the surface of the alluvial deposits will fall away gra dadilly on either aide: this condition may remain stable for some time, but there is always a danger from the tendency of the stream to create for itself a new bed hi the lower level. On the Po, where the banks have been carefully maintained for centuries, the accumulation of the alluvial matters has taken place In the manner above sketched, but to such an extent as to create a real source of danger ; for the banks of the river have been gradually so raised by the accumulation of alluvial mattered, as to be in some casts as much as 40 feet above the extreme depression. [Athuvleu, hl NAT. DIST( Div.) A few final general remarks with respect to non-tidal rivers may here be inserted, namely, that their volumes, even when they are of a comparatively stable character, vary within very wide limits. Thus the Rhone at Lyon, at times, flows with a volume of 7000 cubic feet per minute, at others with a volume of 203,000 cubic feet ; the Rhine at Strasbourg varies between the limits of 9,500 and 164,000 cubic feet per minute ; the Thames at Teddington between • 5,400 and 15,000 cubic feet per minute. For the practical purposes of navigation, a current is considered to be slew when it does not exceed 1 foot 6 inches per second ; a velocity of between 2 feet and 3 feet 6 inches per second is considered to be an average one; beyond the latter limit the velocity of the current is considered to be decidedly unfavourable to an upward navigation, and if the stream should flow at the rate of 6 or 7 feet per second, no boats can economically be made to work against it. When the rate of inclination of the bed of the river attains between 5 and 6 in 10,000, the velocity of the current becomes too great to allow of the navigation being carried on against it by the ordinary modes of trac tion ; but by the use of steam tugs worked by high-pressure expansion engines, the Rhone has been navigated against the stream, even though its inclination has been on the areraye of its course from 7 to 8 in 10,000. Rivers which do not present a permanent depth of 1 foot 2 inches, with an available width of at least 14 feet, are not adapted to receive any kind of navigation, and they can only be rendered available for commerce by being canalized, or by being converted into what the continental engineers call floatable rivers. Streams of this kind occur in mountainous and woody districts, and are much used for the pur pose of floating down rafts of timber, or such natural productions as are not likely to be injured by being occasionally submerged. When the width and depth of a river exceed respectively 18 feet and 18 inches, a barge navigation may be established, provided of course that the rate of flow be not unfavourable ; and when the width and depth of the stream become respectively 60 feet and 8 feet, the river becomes navigable for sea-going vessels. The modes of propulsion to be adopted must depend upon local circumstances. In still water animal power is the most economical ; steam is frequently used in deep water canals and river navigation ; whilst sails, oars, and tides are frequently re sorted to for this purpose in rivers adapted to their use.