It is becoming widely accepted that plant community structure is determined not only by local scale factors, but that regional factors may play considerable role. The research studying the associated processes in different environments with different species assemblages is still limited. We conducted a two-year seed sowing experiment to test whether a plant community in a low-productive mountain snowbed is limited by seed or microsite availability and how these variables depend on natural grazing. In a factorial design, half of the plots received a mixture of seeds of fourteen species naturally occurring at the study site and above ground biomass was removed from half of the plots. These treatments were applied to plots with long term grazer exclosures and to plots accessible to grazers. Both sowing and biomass removal increased the number of seedlings, the species richness of seedlings and total species richness. The number of seedlings was higher in open plots than in exclosures in the second year. Both seedling richness and total species richness were higher in open plots. Seedling recruitment was negatively related to the amount of above ground biomass and positively to the initial species richness. These results suggest that even fairly low-productive environments can be both seed and microsite limited and that these depend on grazing pressure. Natural grazing by mammal herbivores (e.g. lemmings and reindeer) favours species colonization and seedling emergence. Low-productive mountain snowbeds are prone to colonization from the local species pool and even high species richness may not constrain ingression of new species.