e chess
category, and approximates it to the likeness of active service. Most of
a general's decisions, once a fight has begun, must be made in such
brief intervals of time. (But we leave unlimited time at the outset for
the planning.)

As to our time-keeping, we catch a visitor with a stop-watch if we can,
and if we cannot, we use a fair-sized clock with a second-hand: the
player not moving says "Go," and warns at the last two minutes, last
minute, and last thirty seconds. But I think it would not be difficult
to procure a cheap clock--because, of course, no one wants a very
accurate agreement with Greenwich as to the length of a second--that
would have minutes instead of hours and seconds instead of minutes, and
that would ping at the end of every minute and discharge an alarm note
at the end of the move. That would abolish the rather boring strain of
time-keeping. One could just watch the fighting.

Moreover, in our desire to bring the game to a climax, we decided that
instead of a fight to a finish we would fight to some determined point,
and we found very good sport in supposing that the arrival of three men
of one force upon the back line of the opponent's side of the country
was of such strategic importance as to determine the battle. But this
form of battle we have since largely abandoned in favour of the old
fight to a finish again. We found it led to one type of battle only,
a massed rush at the antagonist's line, and that our arrangements
of time-limits and capture and so forth had eliminated most of the
concluding drag upon the game.

Our game was now very much in its present form. We considered at various
times the possibility of introducing some complication due to the
bringing up of ammunition or supplies generally, and we decided that it
would add little to the interest or reality of the game. Our battles are
little brisk fights in which one may suppose that all the ammunition and
food needed are carried by the men themselves.

But our latest development has been in the direction of killing hand to
hand or taking prisoners. We found it necessary to distinguish between
an isolated force and a force that was merely a projecting part of a
larger force. We made a definition of isolation. After a considerable
amount of trials we decided that a man or a detachment shall be
considered to be isolated when there is less than half its number of its
own side within a move of it. Now, in actual civilised warfare