should be used in the construction of a boiler that does not improve
with punching, and further on I will show by the experiments made by
Hoopes & Townsend, of Philadelphia, that good material is improved by
punching; that is to say, with properly made punches and dies, by the
upsetting around the punched hole, the value of the plate is increased
instead of diminished, the flow of particles from the hole into the
surrounding parts causing stiffening and strengthening.

_Drilling Rivet Holes._--In the foregoing I have not referred to the
drilling of rivet holes in place of punching. The great objection to
drilling rivet holes is the expense, from the fact that it takes more
time, and when drilled of full rivet size we are met with the
difficulty of getting the rivet holes to correspond, as they are when
punched of full rivet diameter. When two plates are drilled in place
together, the drill will produce a _burr_ between the two plates--on
account of their uneven surfaces--which prevents them being brought
together, so as to be water and steam tight, unless the plates are
afterward separated and the burr removed, which, of course, adds
greatly to the expense.

The difference in strength between boiler plates punched or drilled of
full rivet size may be either greater or less than the difference in
strength between unperforated plates of equal areas of fracture
section. When the metal plates are very soft and ductile, the
operation of punching does no appreciable injury. Prof. Thurston says
he has sometimes found it actually productive of increased strength;
the flow of particles from the rivet hole into the surrounding parts
causing stiffening and strengthening. With most steel and hard iron
plates the effect of punching is often to produce serious weakening
and a tendency to crack, which in some cases has resulted seriously.