Marl, clay and sand beds are likely to slip when they are superimposed, and there are some districts in which sandy soil is so charged with water that, unless the drainage of the slopes and formation suffices to drain for some distance the land outside a cutting, the sand will become overcharged with moisture and will act as a fluid and slip, the lateral support being removed by the act of excavation and its normal condition altered. Being so delicately balanced the least additional disturbing force, such as a spoil bank being tipped upon the surface, or the inducement or acceleration of a flow of water, will set it in motion and make it a quicksand. For instance, small sand islands have been removed by making cuts in them from 15 to 20 feet in width, and by men shaking bars, &c., inserted in the soil; the sand along the edge becomes loose, falls, and the current sweeps it away. As an example of the changeability of the condition of sand may be named that in sinking pits by congelation in loose sand it has been found that the grains during the freezing of the water, by means of tubes containing a freezing mixture, were additionally separated about 5 to 7 per centum.
As the sand met with in public works is seldom in very deep beds, it has not been subject to the steadying forces which many earths have undergone, and it may have been constantly moving until its final deposition, and therefore it is easily set in motion; and although sand will subside less from a load after it is saturated with moisture, the water in it trying to escape may cause it to slip upon an unsupported surface such as a slope.
Many experiments have shown that the power of absorption of sand decreases with the fineness of the grain, and that sand when thoroughly wet will contain water equal to about one-third to two-fifths of its bulk, and that almost all this can be drained; hence its varying condition and instability. If a well be sunk in sandstone and regularly pumped it will drain the rock around for some distance, the drainage space being conical, its vertex the bottom of the well, and its base the surface, varying in extent according to the nature of the soil and depth of well, showing the porous nature of sand.
The interstices of silicious sea-sand, when not compressed, have been ascertained by Mr. J. Watt Sandeman, M. Inst. C.E., to amount to about 40 per cent. of the volume of sand. For coarse or fine sand, or a mixture of the two, the interstices did not vary much. When it was compressed by a rammer in water, its bulk could be reduced to the extent of 12½ per cent. The interstices of broken red sandstone, varying in size to that which would pass through an 8-inch ring, were found to be 36 per cent. of the whole volume, but as the stones were in contact 10 per cent, must be added, and if under water 15 per cent.
| Per cent. of the interstices. | |
|---|---|
| Broken Welsh limestone to pass a 3-inch ring | 50·9 |
| Gravel, free of sand. Small pebbles to pieces gauged by a 25-inch ring | 33·6 |
| Welsh limestone and gravel as above mixed in equal proportions | 34·0 |
| Mason’s shivers of Anglesey limestone, small gravel to pieces gauged by a 4-inch ring | 48·0 |
| Runcorn red sandstone, large, varying in size from pieces to pass a 4-inch ring to an 8-inch ring | 50·0 |
| Ditto, small, varying in size from sand to pieces gauged by a 4-inch ring | 34·0 |
| The two preceding when mixed in equal proportions | 36·0 |
The experiments clearly show the known great capability of subsidence in sandy and open sandy gravel soils, their clear water space, and how easily fine sand may, by a current of water, become running sand, and their adaptability for ramming and consolidation by moisture. Tipped sand when rammed will subside if saturated with water nearly as much as it can be beaten down, which shows how greatly its bulk is affected by water, and although its rapid consolidation is an advantage in embankments, it is of importance that percolation should be equal.
The chief conditions of a safe foundation upon pure sand, namely, that it cannot escape laterally or be undermined, are obviously not to be attained in either cuttings or embankments, as the lateral support is removed, and the slopes are liable to be undermined and unequally charged with water, and the influence of water on sandy soils is the principal cause of their instability, for in excavating cuttings the face will frequently stand at a very steep slope if dry, but upon its becoming saturated the sand may flow, and in the case of gravel and sand, although the stones forming gravel do not change, the whole subsides.
As gravel is found in various conditions, it may be well to classify it as it is herein regarded.
Clean gravel is considered as that which nearly approaches the condition of a pebbly beach. If an appreciable quantity of sand is present, it is sandy gravel. If loam, or marl, or clay, it is loamy, marly, or clayey gravel.
Gravel hills are large accumulations of water-worn rocks, and may have boulders in them intermixed with the freshwater deposits of sands and marls, and by means of natural cementing material between the particles seem to be firmly set and to be so conglomerated as to appear to be in a similar condition to weak concrete; but there is always a chance of the matrix becoming dissolved, therefore it is advisable to test a mass by the application of water and to expose it to the atmosphere before relying upon its permanent stability, and with the view to determine whether it is hard cemented gravel or not.