If the strata consist of sand, water will pass through them with facility, and they will also hold a considerable quantity between the interstices of their component grains; whereas a bed of pure clay will not allow of the passage of water. These are the two extremes of the case; the intermixture of these materials in the same bed will of course, according to their relative proportions, modify the transmission of water. Prestwich found by experiment that a silicious sand of ordinary character will hold on an average rather more than one-third of its bulk of water, or from two to two and a half gallons in one cubic foot. In strata so composed the water may be termed free, as it passes easily in all directions, and under the pressure of a column of water is comparatively but little impeded by capillary attraction. These are the conditions of a true permeable stratum. Where the strata are more compact and solid, as in sandstone, limestone, and oolite, although all such rocks imbibe more or less water, yet the water so absorbed does not pass freely through the mass, but is held in the pores of the rock by capillary attraction, and parted with very slowly; so that in such deposits water can be freely transmitted only in the planes of bedding and in fissures. If the water-bearing deposit is of uniform lithological character over a large area, then the proposition is reduced to its simplest form; but when, as in the deposit between the London clay and the chalk, the strata consist of variable mineral ingredients, it becomes essential to estimate the extent of these variations; for very different conclusions might be drawn from an inspection of the Lower Tertiary strata at different localities. [Illustration: _a_ London clay, _b_ Sands and clay, _c_ Chalk. Fig. 5.]