This paper is intended primarily to provide a relatively non- theoretical discussion corresponding to the material covered in RB-62-26. This paper contains a numerical example and attempts to serve the needs of the practitioner rather than of the theoretician. Like the earlier paper, this one presents a technique for using the Wilcoxon test to test a broader type null hypothesis, such as is encountered in the Behren's Fisher problem, where it is required to test the equality of the means of two normal populations having (possibly) unequal variances. (JGL)
