irreparable. His next cylinder was cast and bored at
Carron, but it was so untrue that it proved next to useless. The
piston could not be kept steam tight, notwithstanding the various
expedients which were adopted of stuffing it with paper, cork, putty,
pasteboard, and old hat. Even after Watt had removed to Birmingham,
and he had the assistance of Boulton's best workmen, Smeaton expressed
the opinion, when he saw the engine at work, that notwithstanding the
excellence of the invention, it could never be brought into general use
because of the difficulty of getting its various parts manufactured
with sufficient precision. For a long time we find Watt, in his
letters, complaining to his partner of the failure of his engines
through "villainous bad workmanship." Sometimes the cylinders, when
cast, were found to be more than an eighth of an inch wider at one end
than the other; and under such circumstances it was impossible the
engine could act with precision. Yet better work could not be had.
First-rate workmen in machinery did not as yet exist; they were only in
process of education. Nearly everything had to be done by hand. The
tools used were of a very imperfect kind. A few ill-constructed
lathes, with some drills and boring-machines of a rude sort,
constituted the principal furniture of the workshop. Years after, when
Brunel invented his block-machines, considerable time elapsed before he
could find competent mechanics to construct them, and even after they
had been constructed he had equal difficulty in finding competent hands
to work them.[16]

Watt endeavoured to remedy the defect by keeping certain sets of
workmen to special classes of work, allowing them to do nothing else.
Fathers were induced to bring up their sons at the same bench with
themselves, and initiate them in the dexterity which they had acquired
by experience; and at Soho it was not unusual for the same precise line
of work to be followed by members of the same family for three
generations. In this way as great a degree of accuracy of a mechanical
kind was arrived at was practicable under the circumstances. But
notwithstanding all this care, accuracy of fitting could not be secured
so long as the manufacture of steam-engines was conducted mainly by
hand. There was usually a considerable waste of steam, which the
expedients of chewed paper and greased hat packed outside the piston
were insufficient to remedy; and it was not until the inve