Aim: Lungfishes are one of the two surviving clades from the once diverse grade of lobe-finned fishes leading to tetrapods. This classic living fossil lineage, which is the living sister to four-limbed terrestrial vertebrates, appeared approximately 425 million years ago and rapidly diversified. However, the evolution of lungfishes after their initial radiation is poorly understood, and whether their present distribution tracks ancient geographic change is a classic problem in biogeography.

Location: Global

Taxon: Lungfishes (Dipnoi)

Methods: Here, we combine mitogenomic, nuclear gene, and fossil data to reconstruct the timing of lungfish diversification, and integrate fossil species in a Bayesian tip-dating approach to quantitatively test hypotheses of lungfish historical biogeography and divergence times. We sample all major living and extinct lungfish lineages, including three of the four species of African lungfishes (Protopterus spp.), the Australian lungfish Neoceratodus forsteri, the South American lungfish Lepidosiren paradoxa, and 13 fossils representing extinct lineages from across the globe.

Results: Our results demonstrate that the divergences of the three major living lungfish clades closely recapitulate the stepwise fragmentation of the Gondwana during the Mesozoic. All of our model-based biogeographic reconstructions support a Gondwanan vicariance model for the origins of the present distribution of lungfish lineages.

Conclusions: In turn, lungfishes provide an excellent example of how the integration of fossil data may drastically change support for historical biogeographic hypotheses previously discounted by molecular data and are one of the few living animal lineages that record incredibly ancient geographic changes in their phylogeny.

DNA sequence data were downloaded from NCBI's Genbank and aligned by eye and using MUSCLE. Please see the files attached for the Genbank ID's. The morphological dataset is from Kemp et al. (2017). 

PartitionFinder v. 2.1.1. (Lanfear et al. 2017)

IQ-TREE(Nguyen et al. 2015)

BEAST 2.6.6. (Bouckaert et al., 2019)

Tracer 1.7.1(Rambaut et al. 2018)