(c) They worked only 8 1/2 hours per day, with Saturday a half-holiday,
while they had formerly worked 10 1/2 hours per day.

(d) An accurate comparison of the balls which were inspected under the
old system of day work with those done under piece work, with
over-inspection, showed that, in spite of the large increase in output
per girl, there were 58 per cent more defective balls left in the
product as sold under day work than under piece work. In other words,
the accuracy of inspection under piece work was one-third greater than
that under day work.

That thirty-five girls were able to do the work which formerly required
about one hundred and twenty is due, not only to the improvement in the
work of each girl, owing to better methods, but to the weeding out of
the lazy and unpromising candidates, and the substitution of more
ambitious individuals.

A more interesting illustration of the effect of the improved conditions
and treatment is shown in the following comparison. Records were kept of
the work of ten girls, all "old hands," and good inspectors, and the
improvement made by these skilled hands is undoubtedly entirely due to
better management. All of these girls throughout the period of
comparison were engaged on the same kind of work, viz.: inspecting
bicycle balls, three-sixteenths of an inch in diameter.

The work of organization began in March, and although the records for
the first three months were not entirely clear, the increased output due
to better day work amounted undoubtedly to about 33 per cent. The
increase per day from June on day work, to July on piece work, the hours