The fate and biological effects of Goltix®, a herbicide containing metamitron as the active ingredient, were studied in two artificial pond systems of different size. The smaller (outdoor microcosm) ponds (5-m3 stainless steel cylinders, 2 m diameter) contained a bottom layer of natural lake sediment. Benthic organisms, all trophic levels of planktonic organisms, and caged rainbow trout were examined. The larger (mesocosm) ponds (75 m2 water surface, maximum water depth 1.75 m, average depth 1.0 m) had a sediment layer on the bottom and side slopes. In these ponds, aquatic plants were also examined. The herbicide was applied twice to each pond with a two-week interval: one pond of each type was treated with 0.5 kg Al/ha had a second one with 5 kg Al/ha. Over a period of 5 months, the effects on the ecosystem (phytoplankton, zooplankton, macrobenthos, aquatic plants, and fish), and the distribution and fate of this herbicide, were examined. The results show concordance between both pond systems, as has been shown in earlier studies with other pesticides. With Goltix®, almost no biological effects were observed, although the laboratory results on the phytoplankton species Scenedesmus subspicatus indicate a potential hazard to these organisms at the very high, exaggerated application rates chosen for the study. The missing effects on phytoplankton in these micro- and mesocosm studies are explained by the very short half-life of metamitron in water under natural conditions.