In 1975, Ladner proved that computational complexty is dense in the sense that if C 0 is strictly contained in C 1 , for any two complexity classes satisfying standard closure properties, then there is a complexity class C 0.5 strictly between them. However, naturally occuring algorithmic problems tend to be complete for one of a tiny stable of important complexity classes. This is often observed and leads to the great practical usefulness of the study of algorithms and complexity.