Paleontological Data Reveals Unexpected Biogeographic Histories of Extant Organisms: Bonytongue Fishes (Teleostei: Osteoglossomorpha) as a Case Study

Abstract

Paleontological data are invaluable for reconstructing the biogeographic history of living organisms. Nonetheless, information from present-day species (neontological data) dominates biogeographic studies of extant clades, due to either incompleteness of the fossil record or challenges in integrating it into evolutionary inference. In this dissertation, I explore the paleontological record of the freshwater fish clade Osteoglossomorpha (bonytongues) to derive a deep-time perspective on the biogeographic history of this ancient and iconic group of fishes. The complex geographic distribution of extant bonytongues, coupled with their abundant fossil record when compared to other tropical freshwater fishes, makes this group an ideal target for biogeographic investigation through a paleontological lens. I first consider the temporal and geographic distribution of the fossil record of seven extant freshwater fish groups – including bonytongues – to derive confidence intervals on their times of origin and test the plausibility of vicariant scenarios in which continental break-ups shaped their modern distributions. I find that, even when fish groups are old enough to have been affected by continental fragmentation during the Mesozoic, successive dispersals and regional extinction tend to erase or confound vicariant patterns and shape the geographic distributions that we see today. The middle portion of my dissertation involves the description of two bonytongue fossil specimens from early Cenozoic marine deposits in Greenland and Morocco. The Greenland specimen extends the geographic range of the group to the Arctic and represents one of their earliest records in marine deposits, few million years after the Cretaceous–Paleogene mass extinction. The Moroccan specimen represents a new genus with cranial adaptations related to feeding ecology previously unknown in these fishes. I show how bonytongues reached a surprising ecomorphological diversity in marine settings during the early Cenozoic, and identify key anatomical features providing evidence for the phylogenetic affinities of fossil marine bonytongues with respect to modern species. Finally, I combine morphological, molecular, geographic, and environmental data in an integration of paleontological and neontological evidence to reconstruct the biogeographic history of bonytongue fishes under phylogenetic models of biogeographic evolution. I find strong support for a marine origin of osteoglossid bonytongues and for long-distance dispersals—followed by multiple marine-to-freshwater transitions—as the primary cause for the present-day widespread distribution of these fishes in tropical freshwaters. Moreover, I show how fossil data can completely overthrow biogeographic patterns that are apparent from the examination of extant distributions alone, highlighting the perils of ignoring paleontological evidence when inferring ancestral conditions for living organisms. This dissertation provides new insights into the evolutionary history of one of the major lineages of freshwater fishes, and establishes Osteoglossomorpha as a promising model system within vertebrates to explore the impact of paleontological data in evolutionary inference.