This paper introduces four new algorithms, the M-algorithm, the L-algorithm, the S-algorithm and the T-algorithm, which enable recognition of distributional functions without doing a fit of the data to the distribution. These algorithms workbest for large