Poaceae (the grasses) is arguably the most successful plant family, in terms of its global occurrence in (almost) all ecosystems with angiosperms, its ecological dominance in many ecosystems, and high species richness. We suggest that the success of grasses is best understood in context of their capacity to colonize, persist, and transform environments (the "Viking syndrome"). This results from combining effective long-distance dispersal, efficacious establishment biology, ecological flexibility, resilience to disturbance and the capacity to modify environments by changing the nature of fire and mammalian herbivory. We identify a diverse set of functional traits linked to dispersal, establishment and competitive abilities. Enhanced long-distance dispersal is determined by anemochory, epizoochory and endozoochory and is facilitated via the spikelet (and especially the awned lemma) which functions as the dispersal unit. Establishment success could be a consequence of the precocious embryo and large starch reserves, which may underpin the extremely short generation times in grasses. Post-establishment genetic bottlenecks may be mitigated by wind pollination and the widespread occurrence of polyploidy, in combination with gametic self-incompatibility. The ecological competitiveness of grasses is corroborated by their dominance across the range of environmental extremes tolerated by angiosperms, facilitated by both C3 and C4 photosynthesis, well-developed frost tolerance in several clades, and a sympodial growth form that enabled the evolution of both annual and long-lived life forms. Finally, absence of investment in wood (except in bamboos), and the presence of persistent buds at or below ground level, provides tolerance of repeated defoliation (whether by fire, frost, drought or herbivores). Biotic modification of environments via feedbacks with herbivory or fire reinforce grass dominance leading to open ecosystems. Grasses can be both palatable and productive, fostering high biomass and diversity of mammalian herbivores. Many grasses have a suite of architectural and functional traits that facilitate frequent fire, including a tufted growth form, and tannin-like substances in leaves which slow decomposition. We mapped these traits over the phylogeny of the Poales, spanning the grasses and their relatives, and demonstrated the accumulation of traits since monocots originated in the mid-Cretaceous. Although the sympodial growth form is a monocot trait, tillering resulting in the tufted growth form most likely evolved within the grasses. Similarly, although an ovary apparently constructed of a single carpel evolved in the most recent grass ancestor, spikelets and the awned lemma dispersal units evolved within the grasses. Frost tolerance and C4 photosynthesis evolved relatively late (late Palaeogene), and the last significant trait to evolve was probably the production of tannins, associated with pyrophytic savannas. This fits palaeobotanical data, suggesting several phases in the grass success story: from a late Cretaceous origin, to occasional tropical grassland patches in the later Palaeogene, to extensive C3 grassy woodlands in the early-middle Miocene, to the dramatic expansion of the tropical C4 grass savannas and grasslands in the Pliocene, and the C3 steppe grasslands during the Pleistocene glacial periods. Modern grasslands depend heavily on strongly seasonal climates, making them sensitive to climate change.
