The possible influence of geographic range size on speciation remains a controversial subject in evolutionary ecology, with theory and data supporting positive, negative and bell-shaped relationships between speciation probability and ancestor range size. In this study, a surrogate of the speciation-range size relationship of extant European tree species (22 genera, 11 families) is obtained by comparing the range-size distribution of candidate ancestors (i.e.species which are thought to have originated new species) with that of relatives, controlling phylogenetic inertia and macroecological sampling bias. In this comparison, species range size is measured qualitatively using six categories. The candidate ancestors included seem to have speciated mainly through allopatric speciation, with fewer instances of hybridogenesis by allopolyploidy. The results show that speciation is significantly facilitated for species with intermediate range size. In the European tree flora, this pattern could be the result of multiple causes, including intermediate dispersal ability at these ranges. Descendant tree species tend to have narrow geographic ranges, a trend which arguably comes from limited dispersal ability in this case. Low dispersal ability could be common in new species as a consequence of widespread adaptation to stable habitats isolated by geographic barriers (“island habitats”) during allopatric speciation. This mechanism could be widespread among regional biotas and would facilitate narrow ranges in new species. This would provide a simple explanation to the observed abundance of narrow ranges in most macroecological species-range size distributions.
The distribution of species range size: a stochastic process