RESUMO
BACKGROUND AND AIMS: Models of costs and benefits of dormancy (D) predict that the evolutionarily stable strategy in long-term stable environments is for non-dormancy (ND), but this prediction remains to be tested empirically. We reviewed seed traits of species in the climatically buffered, geologically stable and nutrient-impoverished campo rupestre grasslands in Brazil to test the hypothesis that ND is favoured over D. We examined the relative importance of life-history traits and phylogeny in driving the evolution of D and assessed seed viability at the community level. METHODS: Germination and viability data were retrieved from 67 publications and ND/D was determined for 168 species in 25 angiosperm families. We also obtained the percentage of embryoless, viable and dormant seeds for 74 species. Frequencies of species with dormant and non-dormant seeds were compared with global databases of dormancy distribution. KEY RESULTS: The majority of campo rupestre taxa (62·5 %) had non-dormant seeds, and the ND/D ratio was the highest for any vegetation type on Earth. Dormancy was unrelated to other species life-history traits, suggesting that contemporary factors are poor predictors of D. We found a significant phylogenetic structure in the dormancy categorical trait. Dormancy diversity was highly skewed towards the root of the phylogenetic tree and there was a strong phylogenetic signal in the data, suggesting a major role of phylogeny in determining the evolution of D versus ND and seed viability. Quantitative analysis of the data revealed that at least half of the seeds produced by 46 % of the surveyed populations were embryoless and/or otherwise non-viable. CONCLUSIONS: Our results support the view that long-term climatic and geological stability favour ND. Seed viability data show that campo rupestre species have a markedly low investment in regeneration from seeds, highlighting the need for specific in situ and ex situ conservation strategies to avoid loss of biodiversity.