This paper describes evolution strategy procedures for real-valued function optimization for the purpose of analyzing its asymptotic convergence properties. Two convergence theorems, which show that evolution strategy asymptotically converges to a global minmize point with probability one, are given.