• Work performed
During the past two years, we collected a large amount of data from before (Jurassic and Cretaceous) and after (Paleocene and Eocene) the end-Cretaceous extinction. We scanned more than 30 Paleocene specimens but because of the quality of preservation, not all specimens could be analysed. For the brain endocasts, we segmented a total of 11 Paleocene and 17 Eocene specimens. The rest of the endocasts was gathered from the literature to build a dataset of 110 species including 25 from the Paleocene. We also obtained the body labyrinth (inner ear endocast) for 22 Paleocene and 7 Eocene species and with the addition of already published data, we constructed a database of 80 taxa.
We have studied in detailed the neurosensory anatomy (brain and inner ear) of one Paleocene mammal Chriacus, which was published this year in the Journal of Anatomy. This work made the cover page of the January issue of the journal. We quantified the size of the brain, neocortex, olfactory bulbs, and petrosal lobules (= paraflocculi) for all the specimens. We also obtained a series of quantitative measurements of the bony labyrinths. We gathered three variables for the 110 species: body mass, geological age and phylogenetic relationships before mapping the endocranial estimates onto a phylogeny tree. For the body mass, we used three ways to determine body mass based on dental, cranial and postcranial data. The age and phylogenetic relationships of all species were gathered from the literature.
We are currently working on the ancestral state reconstruction for the size of the brain, neocortex, olfactory bulbs, and petrosal lobules. We are analyzing the rate of evolutionary change occurring through time for each characteristic. These two types of analyses are allowing us to know the condition at each node of the phylogenetic tree and how fast a feature changed over time and especially across the end-Cretaceous extinction. The goal is to obtain the result in September 2020 and hopefully have a manuscript for October 2020. The same work will have to be done for the bony labyrinth dataset, although this step should be achievable in less time as we already have the age and body mass for the majority of the taxa and method that will be used.
In November 2019, we were still building the dataset and we did not have a phylogenetic tree, body mass or geological age yet for our sample. In order to start learning the technics that we would need for this project; Bertrand designed a study which included published data from her PhD. This project was fundamental in teaching her new tools in the statistical programming software R that could be applied to BEMADE while we were still gathering data. This study is currently under review for Biology Communications. One aspect that was not used in the above project corresponds to the rate of evolution for continuous characters because of the low sample size. This technic has never been used for reconstructing the evolutionary rate of neurosensory features and will provide unprecedent insight on how fast changes through time have occurred. For the past few months, Bertrand has learned to use new tools (e.g. MrBayes and BayesTraits) specifically designed to calibrate phylogenetic trees and study the evolutionary rate of a continuous trait. As of September 6th, 2020, we have some preliminary results and Bertrand will be presenting this work during the SVP meeting (Society of Vertebrate Paleontology) in October 2020.
• Preliminary results
We find that the Phylogenetic Encephalization Quotient (PEQ), a measure of relative brain size, increased in the Cretaceous along branches leading to Placentalia, but then decreased in Paleocene clades (taeniodonts, phenacodontids, pantodonts, periptychids, and arctocyonids). Later, during the Eocene, the PEQ increased independently in all crown groups (e.g. euarchontoglirans and laurasiatherians). The Paleocene decline in PEQ was driven by body mass increasing much more rapidly after the extinction than brain volume. The neocortex remained small, relative to the rest of the brain, in Paleocene taxa and expanded independently in Eocene crown groups. The absolute size of the olfactory bulbs, however, remained relatively stable over time after the extinction, except for a major decrease in Euarchontoglires and some Eocene artiodactyls, while the petrosal lobules (associated with eye movement coordination) decreased in size in Laurasiatheria but increased in Euarchontoglires and Multituberculata.
Our results indicate that an enlarged, modern-style brain was not instrumental to the survival of placental mammal ancestors at the end-Cretaceous, nor to their radiation in the Paleocene. Instead, opening of new ecological niches post-extinction promoted the diversification of larger body sizes, while brain and neocortex sizes lagged behind. The independent increase in PEQ in Eocene crown groups is related to the expansion of the neocortex, possibly a response to ecological specialization as environments changed, long after the extinction.
• Exploitation and dissemination of the results
Bertrand gave two talks at the University of Edinburgh, two talks at international conferences (Brussel, Belgium and Brisbane, Australia) and at invited talks (Institut des Sciences de l’Evolution, Montpellier, France; Department of Archaeology, York University, York, UK; University of Washington, Seattle, WA, USA). One article has been published, one is under review and others are in preparation.
