Chromosome Evolution Model Reveals Hidden Variation in Karyotype-Driven Speciation in Ferns

Because karyotype change commonly generates reproductive isolation between diverging species, rapid karyotype evolution, like that seen in plants, may increase the total rate of diversification. However, few studies have investigated this predicted relationship. Tests of this prediction have identified a positive correlation between the rate of karyotype change (specifically whole genome duplications) and diversification, but tests have been restricted to relatively small and young clades. Fortunately, novel macroevolutionary models have recently become available to investigate patterns of karyotype evolution in large phylogenies, providing new opportunity to investigate variation in karyotype driven diversification. Ferns are one of the most karyotype rich lineages of land plants, but it is not well understood how karyotype changes contribute to diversification across the fern phylogeny. To investigate variation in the contribution of karyotype change to fern diversification, we extend the Chromosome Number Hidden State-dependent Speciation and Extinction model (ChromoHiSSE) to include polyploidy and demi-polyploidy. We use our extended RevBayes implementation of ChromoHiSSE to investigate the evolution of karyotype across 962 leptosporangiate ferns. We recover two distinct modes of chromosome number evolution in ferns. Our Diversification mode is characterized by low rates of karyotype change and net positive diversification, while the Karyotype-change mode is characterized by rapid cladogenetic and anagenetic state changes but net negative diversification. Our investigation is the first to simultaneously model anagenetic and cladogenetic chromosome change in ferns and provides the most complete reconstruction of fern chromosome number evolution. We reinforce the conclusion that modern fern diversity is shaped substantially by polyploid speciation but challenge the assertion that diversification rates are enhanced by karyotype-driven cladogenesis. In ferns, we find that lineages with high chromosome number lability are less likely to persist in the long-term.