these two oblongs combined to represent the area of two countries
which are in alliance, and which are further so situated that B is the
weaker Power to the alliance both (1) in his military strength, and
(2) in his tenacity of purpose. Next grant that B is divided by the
dotted line, CD, into two halves--B not being one homogeneous State,
but two States, B1 and B2.

Next let it be granted that while B1 is more likely to remain attached
in its alliance to A, B2 is more separate from the alliance in moral
tendency, and is also materially the weaker half of B. Finally, let
the whole group, AB, be subject to the attack of enemies from the
right and from the left (from the right along the arrows XX, and from
the left along the arrows YY) by two groups of enemies represented by
the areas M and N respectively.

[Illustration: Sketch 8.]

It is obvious that in such a situation, if A is the chief object of
attack, and is the Power which has both provoked the conflict and made
itself the chief object of assault by M and N, A is by this
arrangement in a position _politically_ weak.

That is, the strategical position of A is gravely embarrassed by the
way in which his ally, B, separated into the two halves, B1 and B2,
stands with regard to himself. B2 is isolated and thrust outward. The
enemy, M, upon the right, attacking along the lines XX, may be able to
give B2 a very bad time before he gets into the area of B1, and long
before he gets into the area of the stronger Power, A. It is open to M
so to harass B2 that B2 is prepared to break with B1 and give up the
war; or, if the bond between B2 and B1 is strong enough, to persuade