The evolution of similar form and function in divergent lineages has long fascinated biologists — a classic example is the strikingly similar growth habit of desert plants (succulent cacti and euphorbs). The overarching question on such cases of convergent evolution is whether same or different physiological, developmental and molecular mechanisms are employed and, by extension, whether evolution is predictable. In this study, we investigate the evolution of the dwarf growth habit in willows (5-10 cm) that grow above the tree line in arctic and Alpine environments. Willows (Salix) are a large genus with 300-500 species. Dwarf willow species are not closely related to each other; instead, their closest relatives are shrub species, suggesting that dwarfism evolved multiple times (Wagner et al. 2018, Ecology and Evolution). We have sampled and re-sequenced 6 individuals of each of four dwarf-shrub species pairs from different lineages, as well as outgroup species. We further analyze published re-sequencing data from additional dwarf and shrub willows and from the genus Populus (trees) from the willow family (Salicaceae). For our analyses, we use a draft reference genome of the dwarf species Salix herbacea that was recently established in our group. In this project, we plan to employ population genomic and phylogenetic analyses to answer the following questions: (1) Are there variants in particular genes associated with dwarfism (molecular parallelism)? (2) Are there signs of selection in the same genetic pathways in the dwarf species as compared to the shrub species? (3) Are green revolution genes, that reduce plant stature, involved? (4) Do dwarfism associated genes exhibit signs of gene flow between species, indicating that dwarfism variants may be shared rather than derived (hybridization)?