at will endure in its original state all your life." (!)

I may observe that this gentleman had not the slightest commercial
interest in steel bows.

I also came in contact once with an example of the opposite class.
This gentleman had a little son who was in the habit of borrowing his
father's violin bow surreptitiously for the purpose of perfecting
himself in the useful art of single stick practice. The inevitable
happened, and when I saw the bow it was proudly exhibited to me as an
example of what could be done with a little ingenuity. The two halves
of the broken bow had been well glued together, two steel pen nibs
had been placed so as to form a sort of metal tube to protect the
fracture, and the whole was bound securely with strong silk. In its
owner's estimation it was "as good as ever, sir, as good as ever."

I propose to state here briefly what can be done and what is
advisable to have done in the way of bow-repairing.

If a bow is broken in the upper part of the stick it is just possible
to splice on a new head and throat, but it is not worth doing, for
the _cambre_ and balance of the original can never be reproduced. In
the first place there is a different piece of wood which, however
well matched, is bound to be sufficiently strange to disturb such a
delicate instrument. And then the _cambre_ of the new piece has to be
set before it is joined on to the old stick and thus it becomes
impossible to make a satisfactory curve throughout.

To re-adjust the original head is not feasible, as the only joint
that will stand the strain to which a bow is subjected is a long
diagonal one extending for several inches.

[Illustration: FIG. 38.]

Splicing a new "handle" (Fig. 38_d_) is, however, frequently done,
and is often advisable. It occasionally happens that a valuable bow
becomes so worn by the pressure of the fingers or thumb, or by the
friction of the nut and screw, as to be beyond the reach of the more
usual repairs. It then becomes necessary to substitute a new handle,
and this can be done by skilful repairers as to make absolutely no
difference to the balance of the stick. The joint is in this case
also a diagonal one extending usually from near the upper extremity
of the "lapping" downwards for some four or five inches. It should be
seen that the surfaces brought in contact in such a joint are so
placed as to be perpendicular to the plane of the hair. Otherwise it
cannot endure for any length of time.