Three theoretical non-point water pollution (NPS) control schemes were tested repeatedly in experimental studies tax-subsidy scheme (K. Segerson, 1988), collective fining (Xepapadeas, 1991) and random fining (Xepapadeas, 1991). Camacho and Requate (2004) summarized results reported by Spraggon (2002), Vossler et al (2002), Cochard et al (2002), and Alpizar et al (2004) and replicated their experiments. This paper discusses similarity and differences among all the reported results and in particular the following two. First, both collective fining and random fining induce abatement under the target, their performance deteriorates over time and is relatively consistent over the replications. Second, tax-subsidy scheme induced abatement over the target, its performance is consistent over periods, but not over the replications.