Skip to main content
Dryad

Data from: Phylogenomics and a new fossil synthesis illuminate the early evolution of palms (Arecaceae)

Bellot, Sidonie1; Condamine, Fabien2; Matsunaga, Kelly3; Morley, Robert J.4; Ramirez-Barahona, Santiago5; Cano, Ángela6; Couvreur, Thomas L. P.2; Cowan, Robyn1; Eiserhardt, Wolf L.7; Kuhnhäuser, Benedikt G.1; Maurin, Olivier18; Siros, Michelle1; Forest, Félix1; Leitch, Ilia J.1; Baker, William J.1

Published Apr 28, 2026 on Dryad. https://doi.org/10.5061/dryad.pzgmsbcwg

Data files

Apr 28, 2026 version files 66.23 MB

Click names to download individual files

Abstract

Tropical rainforests are home to almost half of plant diversity, yet a shortfall in phylogenetic hypotheses for tropical plants hinders our understanding of how rainforests have formed and adapted to past global changes. Phylogenetic and historical biogeographic evidence from key rainforest lineages, such as palms (Arecaceae), is required to illuminate the history of these ecosystems. However, our current understanding of the palm tree of life is based on uneven sampling of plastid and nuclear data. Moreover, numerous palm genera and palm fossils have been described or revised over the past decade, casting doubt on palm relationships, ages, and ancestral ranges inferred in early studies. Here, we infer the phylogenetic relationships of all 184 palm genera based on data from 1,033 nuclear genes generated using target sequence capture. Our palm phylogenomic tree is highly resolved and supported. Remaining areas of ambiguity reflect the complex dynamics of palm evolution, including rapid diversification events in subfamily Arecoideae and putative cases of ancient reticulation throughout the family. We undertake a comprehensive review of the palm fossil record and use a vetted selection of fossils to estimate divergence times with two Bayesian methods, the first based on calibration of five nodes using the age of fossils assigned to them, and the second based on co-estimation of divergence times and phylogenetic placements of 113 fossils with a Fossilized Birth-Death model. We then use the distribution ranges of extant and fossil taxa to infer ancestral ranges. We show that the palm family first diversified in the Early Cretaceous in regions corresponding to what is now North, Central, and South America and Oceania, that many tribes and subtribes had originated by the Late Cretaceous, and that two-thirds of the genera had diverged by the Oligocene. Fossil-informed analyses provide a more complex picture of the early biogeography of palms than analyses relying only on the ranges of extant taxa. Despite uncertainties regarding fossil placement, it is clear that palms dispersed dozens of times across oceanic gaps, and that dispersal and extirpation patterns are consistent with an ancient affinity of palms for megathermal climates. Our dated phylogenomic trees and curated fossil dataset provide a new foundation for evolutionary studies on palms, opening the door to deeper research on the rainforest biome in which they thrive.