oint of
the needle into the form of a small drill and repeat the operation we
find that the oil no longer ascends. It rises from the point to the
extreme width of the drill portion, but refuses to go beyond. It clings
to that portion of the needle which would correspond to the ridge just
back of the slope in a conical pivot. Water, oil, etc., when placed in a
clean wine glass, do not exhibit a perfectly level surface, but raise at
the edges as shown at _a_ in Fig. 6. If a tube is now inserted, we find
that the liquid not only rises around the outside of the tube and the
edges of the vessel, but also rises in the tube far beyond its mean
level, as shown at _b_. These various effects are caused by one of the
forces above described, _i. e._, the adhesion, or mutual attraction
existing between the liquid and the substance of the vessel and rod. The
word capillarity is of Latin derivation, and signifies hair-like
slenderness. The smaller the tube, or the nearer the edges of a vessel
are brought together, the higher in proportion will the liquid rise above
the level. An ascent of a liquid, due to capillarity, also takes place,
where the liquid is placed between two separate bodies, as oil placed
between two pieces of flat glass. If the plates are parallel to one
another and perpendicular to the surface of the liquid it will ascend to
the same height between the plates, as shown at _c_ in Fig. 6. If the
plates were united at the back like a book and spread somewhat at the
front, the oil would ascend the higher as the two sides approach one
another, as shown at _d_, Fig. 6. If a drop is placed somewhat away from
the intersecting point, of the glasses, as shown at _m_ it will, if not
too far away, gradually work its way to the junction, providing the
glasses are level. If, however, the glasses are inclined to a certain
extent, the drop will remain stationary, since it is drawn in one
direction by gravity and in the other by capillarity. When a drop of oil
is placed between two surfaces, both of which are convex, or one convex
and the other plain, as shown at _g_, it will collect at the point _n_,
at which the surfaces nearest approach one another. We now see very
clearly why the hole jewel is made convex on the side towards the
end-stone and concave on the side towards the pivot.

[Illustration: _Fig. 6._]

Particular pains should be taken to polish those portions of the pivots
which actually enter the jewel hole and to see tha