ffers a passage. Let us suppose that a
concave clay basin of considerable depth is filled with water-soaked
sand. At the very lowest point on the edge of this basin a stream will
slowly trickle out, and will continue to flow, as long as water from
above keeps the bowl full.

It is not uncommon to find on hillsides, in many regions, little brooks
whose beginnings are traceable to springs that gush out of the ground.
The spring fills a little basin, the overflow of which is the brook. If
the source of this spring could be traced underground, we might easily
follow it along some loose rock formation until we come to a clay basin
like the one described above. We might have to go down quite a distance
and then up again to reach the level of this supply, but the level of
the water at the mouth of the spring can never be higher than the level
of the water in the underground supply basin.

Often in hot summers springs "go dry." The level of water in the supply
basin has fallen below the level of the spring. We must wait until
rainfall has added to the depth of water in the basin before we can
expect any flow into the pool which marks the place where the brook
begins.

Suppose we had no beds of clay, but only sand and gravel under the
surface soil. We should then expect the water to sink through this loose
material without hindrance, and, finding its way out of the ground, to
flow directly into the various branches of the main river system of our
region. After a long rain we should have the streams flooded for a few
days, then dry weather and the streams all low, many of them entirely
dry until the next rainstorm.

Instead of this, the soil to a great depth is stored with water which
cannot get away, except by the slow process by which the springs draw
it off. This explains the steady flow of rivers. What should we do for
wells if it were not for the water basins that lie below the surface? A
shallow well may go dry. Its owner digs deeper, and strikes a lower
"vein" of water that gives a more generous supply. In the regions of the
country where the drift soil, left by the great ice-sheet, lies deepest,
the glacial boulder clay is very far down. The surface water, settling
from one level to another, finally reaches the bottom of the drift.
Wells have to be deep that reach this water bed.

The water follows the slope of this bed and is drained into the ocean,
sometimes by subterranean channels, because the bed of the nearest