Molecular exploration of fossil eggshell uncovers hidden lineage of giant extinct bird

<p>Madagascar's extinct elephant birds are the least studied of the ratites, yet in recent years have proven to be critical for understanding the evolution of all birds. With large spatio-temporal gaps in the fossil record and poor biomolecular preservation of skeletal specimens, the systematics of these flightless giants remains controversial. Here, we overcome these limitations using a comprehensive molecular analysis of 1000-year-old fossil eggshells to provide the first description of elephant bird phylogeography and reveal new insights into the ecology of the world's largest birds. High-throughput sequencing of 21 mitochondrial genomes from across Madagascar reveals genetic variation that is correlated with eggshell morphology, stable isotope composition, and geographic distribution. The elephant bird crown is dated to ca. 30 Mya during the Eocene-Oligocene, a period of marked global cooling when Madagascar's local palaeoclimate is estimated to have become less arid as the island moved northward. High levels of between-clade genetic variation argue for the reclassification of Mullerornis into a new family. Low levels of within-clade genetic variation are unexpected and suggest that there were only two elephant bird genera, with one species per genus existing in southern Madagascar during the Holocene. However, we reveal a newly discovered eggshell collection from Madagascar's far north that is phylogenetically, morphologically, and ecologically distinct, representing a unique lineage of <em>Aepyornis</em>. Furthermore, molecular dating estimates that divergence within <em>Aepyornis</em> coincides with the aridification of Madagascar during the early Pleistocene ca. 1.5 Ma, and is consistent with the fragmentation of populations in the highlands driving diversification, as well as the evolution of extreme gigantism over short timescales. Molecular analysis thus provides the first insight into elephant bird speciation, and we advocate for a revision of their taxonomy that integrates palaeogenomic and palaeoecological perspectives.</p>