A Treatise On The Principles And Practice Of Harbour Engineering

22Ô HARBOUR ENGINEERING. its own maintenance work, and it acts generally on the lines of an ideal stream. One forcibly impressive claim which has been put forward on behalf of a rov- ing channel is, that by its constant change of course it deters the estuary from silting up in any part. This contention is one which has no little weight, because, with the reduction in capacity of a tidal basin or compartment, there is a corresponding reduction in the quantity of flood-water admitted, and a loss of scouring effeet on the subséquent ebb. The confinement of a channel within restricted boundaries inevitably leads, in the case of water heavily charged witli silt, to accretion in the adjacent submerged area. In other words, channel-training is a preliminary to land réclamation, and land reclamation is the general outcome of channel-training. Land reclamation is not an unmixed benefit; it may be attended by serions conséquences to ports situated between the locality of reclamation and the sea, and it may entail other physical dis- abilities not altogether easy to foresee. Considerable discretion is therefore required both in planning and in carrying out undertakiugs embodying any such scheme. Fixed V. Variable Channels.—Taking the question of river-training, however, as a whole, on its intrinsic merits, it seems to turn on the point of relative advantages—whether, in fact, it is preferable to have a deep, narrow, well-defined, constant channel, with adequate energy for its own maintenance, but with none utilisable for counteracting any silting tendencies elsewhere, or, on the other hand, to have a shifting channel with a more sluggish flow, sluicing à large expanse of sand so as to keep it from consolidating in any part, and so affording a broad waterway of greater sectional area, but of inferior depth, and subject to all the inconveniences of a shallow bar. The first undoubtedly represents the ideal condition, but, as indicated above, there are practical and circumstantial grounds in some cases constituting a preponder- ating argument in favour of the latter. Accretion.—Although it is oftentimes assumed that accretion is the inévi- table consequence of confining a channel within narrow limits, yet such an assumption is not legitimate on all occasions. Accretion can only arise from the deposition of suspended sediment, and this sediment can only be forthcoming from a supply in excess of that which the outgoing stream can carry. Now, there is nothing to show that any additional detritus is forthcoming from the upper reaches of a regulated river. But even supposing that there be an augmenta- tion, the increased velocity of the stream renders it capable of transpoiting a larger percentage of solid matter than before. Evidently, therefore, any deposition which takes place is hardly attributable to detritus brought down by the upland waters. The more likely and, as a matter of fact, the only possible source of accretion, is a tidal flow laden with the harvest of coast erosion. The flood-tide, entering estuaries on a sandy coast, is almost universally heavily charged with mud and fine particles which have every tendency to deposit themselves at the period of slack water, unless the down stream be so directed as to bear