Heinz Ellenberg's historically important work on changes in the abundances of a community of grass species growing along experimental gradients of water table depth has played an important role in helping to identify the hydrological niches of plant species in wet meadows. We present a previously unpublished complete version of Ellenberg's dataset from the 1950s together with the results of a series of modern statistical analyses testing for hypothesized overyielding of aboveground net primary production as a consequence of resource-based niche differentiation. Interactions of species with water table depth and soil type in the results of our analyses are qualitatively consistent with earlier interpretations of evidence for differences in the fundamental and realized niches of species. Arrhenatherum elatius tended to dominate communities and this effect was generally positively related to increasing water table depth. There was little overyielding of aboveground net primary production during the two repeats of the experiment conducted in successive single growing seasons. Examination of how the effects of biodiversity on ecosystem processes vary across environmental gradients is an underutilized approach--particularly where the gradient is thought to be an axis of niche differentiation as is the case with water availability. Furthermore, advances in ecology and statistics during the 60 years since Ellenberg's classic experiment was performed suggest that it may be worth repeating over a longer duration and with modern experimental design and methodologies.