Abstract
Recurrence relationships among the distribution functions of order statistics of independent, but not identically distributed, random quantities are derived. These results extend known theory and provide computationally practicable algorithms for a variety of problems.