e (or the pecker, the effect would be the
same) with a device for increasing or diminishing its length, is also
unnecessary and unsound.

The wear on a well-designed gas valve operating mechanism is practically
nil; and even if there was wear, the effect would be to cause the valve
to open a trifle later and close sooner than it would otherwise, _i.e._,
it would remain open a shorter time during each charging stroke. This in
turn (other conditions remaining the same) would give us a weaker
mixture; and although too weak a mixture is preferable to a too rich
one, we should have to adopt some means of increasing the richness of
the mixture; otherwise the maximum power of the engine would soon be
seen to diminish.

To get the mixture normal again we must either enlarge the gas inlet or
cut down the air-supply somewhat, and so keep the proportions the same.
That is to say, the quality of the mixture is dependent upon the
relative dimension of the gas and air inlets. We know by actual trial
that if at the completion of the charging stroke the pressure in the
cylinder is approximately that of the atmosphere, better results are
obtained than when the pressure is considerably below that of the
atmosphere. Thus, the larger we make the inlet ports (but still
retaining correct relative dimensions) the more readily will the mixture
be drawn into the cylinder as the piston moves forward, tending to
create a vacuum. Of the two courses open to us to retain a good mixture
it is preferable to open out the gas-supply, for by cutting down the
air-supply, and sucking the gas in, due to the partial vacuum being
formed, we should be keeping the proportions correct at the expense of
reducing the total volume of the explosive mixture (more strictly
speaking, the density of the charge) admitted to the cylinder.

Under normal conditions it is not necessary to create a high vacuum to
suck the gas into the cylinder, but it is as well to understand what
results we would tend to produce, did we work on these lines. Of course,
with small high-speed engines fitted with suction air valve, the vacuum
is higher than it would be in slow-speed engines with mechanically
operated valves. If we take an extreme case as an example, where, to get
any gas to speak of into the cylinder the air-supply would have to be
cut down or throttled to an abnormal extent, we will realise at once
that such a small quantity of both air and gas would have been drawn in,
and