Global diversification of the common moonwort ferns (Botrychium lunaria group, Ophioglossaceae) was mainly driven by Pleistocene climatic shifts.

Background and aims: The cosmopolitan Botrychium lunaria group belong to the most species rich genus of the family Ophioglossaceae and was considered to consist of two species until molecular studies in North America and northern Europe led to the recognition of multiple new taxa. Recently, additional genetic lineages were found scattered in Europe, emphasizing our poor understanding of the global diversity of the B. lunaria group, while the processes involved in the diversification of the group remain unexplored.

Methods: We conducted the first global phylogenetic study of the group including 533 ingroup accessions sequenced for four plastid loci. We compared results of Bayesian and Maximum Likelihood based methods. We used the phylogenetic relationship we recovered to estimate the timing of divergence with BEAST. We explored ecological segregation between species with climatic variables (CHELSA database) and soil pH measurements. The ploidy level and genome size were estimated with flow cytometry.

Key results: We recovered nine well-supported clades, although relationships between clades were inconsistent between Bayesian and Maximum Likelihood analyses. We treated each clade at the species level, except for one clade including two ploidy levels and one including two recognized diploid species one of which appeared as subclade (B. nordicum) of the other (B. lunaria), resulting in the recognition of 11 species, 4 of which are unnamed. In contrast to previous studies, we found species diversity to be equally distributed across the northern hemisphere, with 6-8 species per continent. We estimated the stem age of the B. lunaria group at 2.5-5.3 million years, with most species 1.5-2.6 million years old, and subclades 0.2-1.0 million years old. Diversification thus coincided with Pleistocene climatic fluctuations that strongly affected the areas inhabited by the group, suggesting that diversification was driven by climatically induced cycles of extinction, dispersal, and migration. Furthermore, ecological differentiation between species suggests these complex population dynamics were associated with adaptations to specific environmental conditions. We found limited evidence that speciation is driven by polyploidization and hybridization.

Conclusions: The B. lunaria group radiation was most likely driven by the Pleistocene climatic shifts. For the first time, we show that ecological drivers may have played a role in the diversification of this group, rather than polyploidization. Furthermore, the B. lunaria group has greater species level diversity than previously assumed and we suspect that further cryptic species may await discovery, especially in the B. neolunaria clade.