This paper deals with the problem of selecting a subset containing the smallest parameter of k or 2 Poisson populations. The population parameters lambda sub i, i1.2,...,k are assumed unknown and there is no a priori information about the correct pairing of the ordered and unordered lambda sub is. Both unconditional and conditional selection rules are investigated. Tables are provided for approximate values of the constants necessary to carry out the procedures. Some other numerical computations have also been provided which shed light on the performance of the selection rule in terms of the probability of selecting a non-best population, the probability of a correct selection and the expected proportion in the selected subset.