Aim: Mountains are often used to study how environmental factors influence biodiversity. However, we have limited understanding of the processes causing biodiversity variation in mountains and whether such processes vary across trophic levels and spatial scales. The aim of this study was to evaluate (i) whether community assembly processes varied along elevational gradients, (ii) whether there were differences in such variation between primary producers (vascular plants) and secondary consumers (spiders) and (iii) whether there were scale dependencies in any elevational variation in community assembly. Location: Fennoscandia, Northern Sweden. Taxon: Vascular plants, spiders. Methods: We used phenotypic and phylogenetic dispersion to quantify how elevation influenced community assembly of vascular plants and spiders and whether there were any scale dependencies in such influences. Our original data of plant and spider communities came from our own field surveys, phenotypic dispersion was calculated based on matrices of ecological traits, and phylogenetic dispersion was calculated from phylogenetic trees for each organism group. Trait matrices were based on a combination of literature values and our own measurements. The phylogeny for vascular plants was based on a published plant super-tree, whereas the phylogeny for spiders was created by ourselves based on the DNA sequences at the mitochondrial cytochrome c oxidase subunit 1 (COI). Results: Plants were environmentally filtered throughout all elevations and scales, but the importance of convergent evolution increased with elevation. For spiders, the importance of environmental filtering as well as niche conservatism increased with elevation. For both groups, communities at smaller scales were more influenced by biotic regulation and niche conservatism than at larger scales. Main Conclusions: Our study highlights both taxonomic differences and scale dependencies in how elevation influences community assembly. We argue that these results can have broad ramifications for our understanding of how spatial variation in biodiversity is generated and maintained. This may have particular relevance for our ability to predict the ecological consequences of climate change. Our results further highlight that high elevation specialists may suffer increased risks of climate driven extinctions due to a combination of increased competition and increased fragmentation of suitable habitats. Particularly for spiders, which had high elevation specialists clustered along specific lineages, such extinctions could lead to significant loss of phylogenetic variation.