As has been stated, clays come from the decomposition of crystaline rocks. The purest clay known (kaolin) is composed of alumina, silica and water. The smaller the proportion of silica the more water it will absorb and retain. Dry clay will absorb nearly one-third of its weight of water, and clay in a naturally moist condition 1-6 to 1-8 its weight of water. The eminent English engineers, Baker and Latham, put the percentage of absorption by clayey soils as high as 40 to 60%. Pure clays shrink about 5% in drying, while a mixture by weight of 1 clay to 2 sand will shrink about 3%. It follows, then, that the larger the percentage of clay there may be in a mixture the greater will be both the expansion and the contraction.
Clay materials may be very deceptive in some of their physical properties, being hard to pick under certain conditions, and yet when exposed to air and water will rapidly disintegrate. Beds of clay, marl and very fine sand are liable to slip when saturated, becoming semi-fluid in their nature, and will run like cream.
The cohesive and frictional resistances of clays becoming thus very much reduced when charged with water, a too liberal use of this material is to be deprecated. The ultimate particles forming clays, viewed under the microscope, are seen to be flat and scale-like, while those of sands are more cubical and spherical. This is a mechanical difference which ought to be apparent to even a superficial observer and yet has escaped recognition by many who have vainly attempted a definition of quicksand.
Mr. Strange recommends filling the puddle trench with material having three parts soil and two parts sand. After the first layer next to bed rock foundation, which he kneads and compacts, he would put the layers in dry, then water and work it by treading, finally covering to avoid its drying out and cracking.
Prof. Philipp Forchheimer, of Gratz, Austria, one of the highest authorities and experimentalists, affirms that if a sandy soil contains clay to such an extent that the clay fills up the interstices between the grains of sand entirely the compound is practically impervious.
Mr. Herbert M. Wilson, C. E., in his “Manual of Irrigation Engineering,” recommends the following as an ideal mixture of materials:
| Cu. yds. | |
| Coarse gravel | 1.00 |
| Fine gravel | 0.35 |
| Sand | 0.15 |
| Clay | 0.20 |
| Total | 1.70 |
This mixture, when rolled and compacted, should give 1.25 cu. yds. in bulk, thus resulting in 26½% compression.
Mr. Clemens Herschel suggests the following test of “good binding gravel:” “Mix with water in a pail to the consistency of moist earth; if on turning the pail upside down the gravel remains in the pail it is fit for use, otherwise it is to be rejected.” For puddling material he would use such a proportion as will render the water invisible.