What is the smallest number of units that need be sampled from a lot for the probability to be at leastγ% that the lot will be rejected if it contains P % defective units or more? This article answers the preceding question and describes the resulting minimum size sampling plans. Such plans call for lot acceptance only if all sampled units are found to be non-defective. An important shortcoming of the plans is that the lot percentage defective must be appreciably lower than the specified percentage P for there to be a high probability of acceptance. Some“almost minimum size”plans are also described. Technical details are given briefly in a series of Appendices.