sed for turning the staff down
in the rough; B for the conical pivots and square shoulders and C for the
under-cutting. The other tools and attachments needed will be described
as I come to them in use.

The balance staff should be made of the best steel, tempered to such a
degree as to give the longest service and yet not so hard as to endanger
the breakage of the pivots. Select a piece of Stubb's steel wire, say No.
46, or a little larger than the largest part of the finished staff is to
be, and center it in a split chuck of your lathe. Be careful in selecting
your chuck that you pick one that fits the wire fairly close. The chuck
holds the work truest that comes the nearest to fitting it. If you try to
use a chuck that is too large or too small for the work, you will only
ruin the chuck for truth. Turn the wire to the form of a rough staff, as
shown in Fig. 2, leaving on a small part of the original wire, as shown
at A. After the wire is roughed out to this general form, remove from the
chuck and get ready to harden and temper it. The hardening and tempering
may be effected in various ways, and I am scarcely prepared to say which
method is the best, as there are several which give about the same
general results. One method of hardening is to smear the blank with
common yellow soap, heat it to a cherry red, and drop endwise into
linseed oil. Petroleum is preferred by some to linseed oil, but, to tell
the truth, I can see no difference in the action of linseed, petroleum or
olive oil. Be sure and have enough oil to thoroughly cool the blank, and
a deep vessel, such as a large-mouthed vial, is preferable to a saucer.
The blank will now be found too hard to work easily with the graver, and
we must therefore draw the temper down to that of fine spring steel.
Before doing this the blank should be brightened, in order that we may
see to just what color we are drawing it. The main object in using the
soap in hardening is that it may form a scale upon the blank, and if the
heating is effected gradually the soap will melt and form a practically
air-tight case around the blank. This scale, if the hardening is
carefully and properly done, will generally chip and fall off when the
blank is plunged in the oil, particularly if the oil is cool, and if it
does not fall off of its own accord, it can easily be removed by rolling
the blank upon the bench. If it does not come out clean, or if soap is
not used, it may be brightened by agai