and upon some all. Some, even, did not run till next day.

We were able to start the day trains out by 10 or 11 o'clock A.M., and
put them through in fair time. Of course, no freights were run that
day, and the next day was used in getting the cars which had been
changed out of the parks and into line. So our freight traffic over
the entire South was suspended practically three days.

The work of changing was to commence at 3:30 A.M., but many of the men
were in position at an earlier hour, and did commence work as soon as
the last train was over, or an hour or so before the fixed time.
Half-past three A.M., however, can be set down as the general hour of
commencement.

For five or six hours in the cool morning the work went on briskly,
the men working with much more than ordinary enthusiasm. But the day
was warm, and after 9 or 10 A.M. it began to lag. All was done,
however, before the day was over, and safe, so that trains could pass
at full speed.

The men all received $1.50 for the work, whether it was finished early
or late in the day, and were paid that afternoon as soon as the work
was done. Tickets were given the men, which the nearest agent paid,
remitting as cash to the treasurer.

On some lines it was deemed best to offer prizes to those who got
through first.

Reports showed some very early finishes. But the facts seem to have
been that under such encouragement the men were apt to pull _too many_
spikes before the change and put _too few_ in while changing. They
were thus reported through early, but their work was not done, and
they took great chances.

It was by most considered unwise to offer such prizes, preferring to
have a little more time taken and be sure that all was safe. Such
lines seemed to get their trains in motion with as much promptness as
others. This, with freedom from accident, was the end sought.

It was found after the work had been done that there had been little
inaccuracies in driving the gauge spike, to which the rail was thrown,
probably from various causes. The rail to be moved may not always have
been exactly in its proper place, and then the template in the hurry
may not have been accurately placed, or the spike may have turned or
twisted.

Whatever was the cause, it was found that frequently the line on the
moved side was not perfect, and, of course, many spikes had to be
drawn and the rail lined up and respiked. The more careful the work
had been done, the less of th