Effects of river regulation on river-margin vegetation: a comparison of eight boreal rivers. Abstract. Regulation and fragmentation by dams belong to the most widespread deliberate impacts of humans on the world's rivers, especially in the Northern Hemisphere. We evaluated the effects of hydroelectric development by comparing the flora of vascular plants in 200-m-long reaches of river margin distributed along eight entire rivers in northern Sweden. Four of these rivers were free-flowing, and four were strongly regulated for hydroelectric purposes. First, we compared species diversity per site between entire freeflowing and regulated rivers. To reduce the effects of natural, between-river variation, we compared adjacent rivers. One regulated river had lower plant species richness and cover than two adjacent free-flowing ones, whereas two other parallel rivers, one regulated and another free-flowing, did not differ significantly. Second, river-margin vegetation responded differently to different types of regulated water-level regimes. Both along run-of-river impoundments, with small but daily water-level fluctuations, and along storage reservoirs, with large fluctuations between low water levels in spring and high levels in late summer and fall, the number of species and their cover per site were lower than along the freeflowing rivers. Regulated but unimpounded reaches were most similar to free-flowing rivers, having lower plant cover per site, but similar numbers of species. For reaches with reduced discharge, evidence was mixed; some variables were lower compared to free-flowing rivers whereas others were not. However, for the last two types of regulation, statistical power was low due to small sample sizes. Third, we classified all plant species according to their dispersal mechanisms and tested whether they respond differently to different types of regulated water-level regimes. Three out of four types of regulation had higher proportions of wind-dispersed species, and two out of four had lower proportions of species without specific mechanisms for dispersal, compared to free-flowing rivers, suggesting that dispersal ability is critical for persistence following regulation. Run-of-river impoundments had higher proportions of long-floating species and species with mechanisms for vegetative dispersal, suggesting that water dispersal may still be important despite fragmentation by dams. Fourth, plant species richness and cover varied with both local factors, such as water-level regime, and regional factors, such as length of the growing season. Presence of clay and silt in the river-margin soil, preregulation position of the contemporary river margin, nonreservoir sites, low altitudes, and long growing seasons were associated with high plant species richness and cover.