form, the whole of the good effect may be secured without the evil. Suppose
he addresses the boys just before they are to go out at the next recess, as
follows:

"I think, boys, that twenty minutes is about the right length of time for
the recess, all told--that is, from the time you go out to the time when
you are _all_ back in your seats again, quiet and ready to resume your
studies. I found yesterday that it took five minutes for you all to come
in--that is, that it was five minutes from the time the bell was rung
before all were in their seats; and to-day I shall ring the bell after
_fifteen_ minutes, so as to give you time to come in. If I find to-day that
it takes ten minutes, then I will give you more time to come in to-morrow,
by ringing the bell after you have been out _ten_ minutes."

"I am sorry to have you lose so much of your recess, and if you can make
the time for coming in shorter, then, of course, your recess can be longer.
I should not wonder if, after a few trials, you should find that you could
all come in and get into your places in _one_ minute; and if so, I shall
be very glad, for then you can have an uninterrupted recess of _nineteen_
minutes, which will be a great gain."

Every one who has had any considerable experience in the management of boys
will readily understand how different the effect of this measure will be
from that of the other, while yet the penalty is in both cases precisely
the same--namely, the loss, for the boys, of five minutes of their play.

_The Little Runaway_.

In the same manner, where a child three or four years old was in the habit,
when allowed to go out by himself in the yard to play, of running off into