ing the excavation we shored up the sides
with planks, to prevent the loose soil from falling in on us and
smothering us, as it so nearly did when we were digging our first cave.
By "shoring," I mean we lined the walls with planks, which were driven
into the ground with large wooden mallets. The planks were braced apart
with sticks at frequent intervals. As the well hole grew deeper we had
to rig up a bucket to haul the dirt out. Our bucket was a soap box
attached to a rope, which passed through a pulley at the top of the
well. The pulley was supported by a tripod made by firmly lashing
together the upper ends of three stout poles and spreading their lower
ends far enough apart to straddle the mouth of the well, as shown in
Fig. 282. After the well had been carried down to a sufficient depth, we
began laying the stone wall, which was to form the permanent lining. We
knew that the wooden walls would not do, because they would soon decay.
Our stone wall, which was built up of flat stones like the chimney of
the log house, was not very strong, I fear, and had not the soil around
it been pretty firm it would probably have caved in. However, if it
served no other purpose, it formed a fairly good finish for the well.

THE WINDMILL TOWER.

[Illustration: Fig. 283. Frame for the Tower.]

The mouth of the well was carefully covered with planks while we
constructed the windmill above it. For the tower of the windmill we
chose four long sticks. They must have measured about 16 feet in length,
and were from 4 to 6 inches in diameter. With them we made two frames of
the form given in Fig. 283, using slabs to brace them apart. These
frames were now set in position, with their lower ends firmly planted in
holes in the ground, and the tower was completed by nailing on a number
of diagonal braces. A couple of boards were nailed across the upper ends
at opposite sides, and holes were drilled through them to provide
bearings for the wind wheel shaft.

THE CRANK SHAFT.

[Illustration: Fig. 284. The Crank Shaft.]

The shaft was a piece of heavy iron rod which we procured from the
blacksmith at Lumberville. Under Bill's direction the blacksmith
hammered a U-shaped bend at the center of the shaft, so as to form a
crank, and then he flattened the rod near the ends (see Fig. 284). When
the shaft was set in its place these flat spots lay just outside of the
bearing boards, and then, to keep the shaft from sliding back and forth
in its