and shutting the cocks alternately, he tied strings to the
handles and then connected to the working beam in such a manner that the
cocks were opened and closed exactly at the nick of time; this caused
the engine to work far more regularly and to do twice the work it had
done previously, the boy's name was Humphrey Potter.

40. _Question._--What did James Watt do in connection with the
atmospheric engine?

_Answer._--Watt being a mathematical instrument maker, was requested to
repair an old engine used by some students of Glasgow University; having
finished the repairs, and in working this model (the best type of the
atmospheric engine), he found and proved by many and various
experiments, that an enormous waste of fuel was absolutely necessary in
working the engine; he found great difficulty in keeping the air from
entering the cylinder, and the cylinder top was so exposed to the
atmosphere that the steam was much condensed when it entered the
cylinder, and he came to the conclusion to put a cover on the top of the
cylinder, and allow the piston-rod to play in a hole in the cover with a
gland and stuffing box, and _to press down the piston with steam instead
of the atmosphere_. This engine was no longer atmospheric, it was a real
steam engine, the first ever seen or constructed, for steam was used to
create the vacuum, and steam was used to work the piston; but this was
only the beginning of his great improvements. This engine though
suitable for the purpose of pumping water, was totally unsuitable for
continuous rotary motion, the steam acting only on the downward stroke
after the piston had been pulled up to the top of the cylinder by means
of the additional weight fixed on the pump end of the beam. He devised a
method to admit steam under the piston as well as above it, but the
flexible chains although suitable for the down stroke of the piston were
powerless in the up stroke, they would hang listless and useless. This
being so, he determined to get rid of the chains at both ends of the
beam, and also both arched ends, and substitute a ridged connection at
both ends of the beam. He put an iron connecting rod from the end of the
beam to the pump rod, and the other end of the beam was connected to the
piston rod by a crosshead; to this engine he attached that grand
appendage the "Parallel Motion" which is the pride of the beam engine up
to to-day. He devised the improvement of the separate condenser for the
exhaust