This paper investigates conditions under which the core of a TU cooperative game is stable. In particular the author extends the idea of extendability to find new conditions under which the core is stable. It is also shown that these new conditions are not necessary for core stability.