Meeting Abstract
As the only living jawless vertebrates, cyclostomes (hagfish and lampreys) are a crucial taxon to infer conditions at the crown vertebrate node. The Hagfish Model accepts hagfish as nesting outside other vertebrates and interprets the morphology as primitive, whereas the Ammocoete Model compares ammocoete larvae of lampreys with amphioxus to derive a filter-feeding vertebrate ancestor.
The Hagfish Model has been challenged by molecular inferences that support cyclostomes as a clade. Aided by synchrotron radiation scanning of preserved soft tissues, I report a new fossil from the Cretaceous of the Middle East. This new taxon nests within the hagfish crown group. The exquisitely preserved soft-tissue anatomy elicited revision of morphological characters in hagfish that were considered primitive. A new phylogenetic analysis recovered cyclostomes as a clade. This topology falsifies a prediction of the Hagfish Model.
The Ammocoete Model is non-parsimonious under the current chordate phylogeny, with tunicates as an immediate vertebrate outgroup and hagfish as sister to lampreys. If the model is correct, the filter-feeding larval stage should have been present in stem lampreys. I test this prediction using stem lampreys from the Devonian southern polar region. In this taxon, skeletal correlates of filter feeding are lacking across all ontogenetic stages. Instead, the larvae have correlates of predatory habits. The absence of a filter-feeding larval stage in this stem lamprey suggests a secondary acquisition of a filter-feeding larval stage in the living lamprey lineage.
Therefore, the new fossil evidence rejects both the Hagfish and Ammocoete models to reconstruct the last common ancestor of all vertebrates. Cyclostomes form a clade under the new analysis, and a filter-feeding larval stage likely represents a secondary acquisition.
