Was Gondwanan breakup the cause of the intercontinental distribution of Osteoglossiformes? A time-calibrated phylogenetic test combining molecular, morphological, and paleontological evidence.

The intercontinental distribution of living freshwater osteoglossiform fishes (Osteoglossomorpha, Teleostei) was hypothesized to predominantly represent the vicariant result of the fragmentation of Gondwana based on physiological and phylogenetic evidence. The fossil record, however, challenges this hypothesis by making transoceanic dispersal plausible because it provides post-fragmentation minimum ages of intercontinental clades and it includes several marine forms. The aim of this study was to test whether the divergence of Osteoglossiformes was compatible with the breakup of Gondwana using newly reconstructed time-calibrated phylogenetic trees based on a large dataset combining extant and fossil taxa and molecular and morphological characters. Bayesian tip-dating and node-dating approaches with different Teleostei age calibrations were employed. The results of the divergence tests are largely dependent on the a priori calibrated age of crown-group Teleostei, with two of the three specific vicariance hypotheses tested in this study not being rejected only when the age of the Teleostei was constrained to be as old as the early Permian. Molecules consistently push the age of crown-group Teleostei back to the Paleozoic, while the fossil record, which is considered informative, does not support such an ancient origin. Reconciling molecular and paleontological estimates of the age of crown-group Teleostei is central to determining the role of Gondwanan breakup in the intercontinental distribution of freshwater teleosts, including Osteoglossiformes.