Broadwood's, twenty years ago, had a strain of sixteen and one-half
tons; the strain has somewhat increased since then. The remarkable
improvement in wiredrawing which has been made in Birmingham, Vienna,
and Nuremberg, of late years, has rendered these high tensions of far
easier attainment than they would have been earlier in the century.

[Illustration: FIG. 3.--BROADWOOD.]

For me the great drawback to one unbroken casting is in the vibratory
ring inseparable from any metal system that has no resting places to
break the uniform reverberation proceeding from metal. We have already
seen how readily the strings take up vibrations which are only pure
when, as secondary vibrations, they arise by reversion from the
sound-board. If vibration arises from imperfectly elastic wood, we hear
a dull wooden thud; if it comes from metal, partials of the strings are
re-enforced that should be left undeveloped, which give a false ring to
the tone, and an after ring that blurs _legato_ playing, and nullifies
the _staccato_. I do not pose as the obstinate advocate of parallel
stringing, although I believe that, so far, it is the most logical and
the best; the best, because the left hand division of the instrument is
free from a preponderance of dissonant high partials, and we hear the
light and shade, as well as the cantabile of that part, better than by
any overstrung scale that I have yet met with. I will not, I say, offer
a final judgment, because there may come a possible improvement of the
overstrung or double diagonal scale, if that scale is persisted in, and
inventive power is brought to bear upon it, as valuable as that which
has carried the idea thus far.

[Illustration: FIG. 4.--BROADWOOD.]