Skip to main content
European Commission logo print header

Avian brain evolution and higher level phylogenetics of modern birds

Final Report Summary - AVIAN EVOLUTION (Avian brain evolution and higher level phylogenetics of modern birds)

There is a growing consensus on the understanding of the basal divergences of modern birds (Neornithes). The phylogenetic affinities of many Cenozoic fossil taxa are still poorly known, however. Recent studies of the brain anatomy of early Palaeogene birds using high resolution X-ray computed tomography (CT) have provided an important source of phylogenetic information for avian systematics. Here, we used the state-of-the-art CT X-ray facility at the NHM (London) to obtain virtual brain endocasts for fossil and living species of birds. These provide new characters of the brain and endocranium to incorporate into a novel phylogenetic tree of modern birds in order to gain a better understanding of avian brain evolution across major avian clades, from the Cretaceous to the present.


Avian evolution aims:

1. to use the state-of-the-art micro-CT X-ray facility at the NHM (London) to obtain virtual endocranial casts of fossil and extant birds;
2. to define new phylogenetic characters of the brain and endocranium dealing with higher-level relationships of modern birds;
3. to provide a new, solid hypothesis for the early divergences of modern birds;
4. to gain a better understanding of avian brain evolution across major avian clades, from the Cretaceous to the present.


Virtual brain endocasts for a broad selection of extinct and extant birds were used to define new hypotheses of homology of the brain and endocranium. Anatomical characters used for phylogenetic analysis include:

(1) position / shape of different brain regions;
(2) form / path of cranial nerves, blood vessels and sinuses;
(3) size of various brain regions relative to overall brain size;
(4) complementary skull characters.

The resulting phylogenetic analysis is based on 40 species and 75 characters so far, and provides a clear picture of the evolution of specific regions of the avian brain through time.

Brain shape is connected with cognitive and sensory abilities, ecology and behaviour. Similar shape and sise of a given brain region can be found in unrelated birds with similar ecologies. However, our study has confirmed that brain shape is relatively consistent between clades and can be phylogenetically informative. More importantly, we show that certain features of the brain, cranial nerves and cranial blood vessels, as well as some complementary features of the skull, are evolutionarily conserved, and hence useful for defining major cladogenetic events in the history of birds.

We provide evidence that the central nervous system of modern birds (Neornithes) exhibits various evolutionary novelties that are not found in more primitive birds and non avian theropod dinosaurs. This confirms the hypothesis that forebrain expansion conferred modern birds with an evolutionary advantage over archaic lineages at the Cretaceous-Tertiary boundary (Milner and Walsh, 2009).

Most recent analyses support a division of modern birds into palaeognaths (ratites and tinamous) and neognaths, the latter clade being divided into Galloanserae (Galliformes - pheasants, megapodes and Anseriformes - ducks and screamers) and Neoaves, which encompass all remaining taxa. The phylogenetic affinities of many Cenozoic fossil taxa are still poorly known, however. Among these are the enigmatic pseudo-toothed birds or bony-toothed birds (Odontopterygiformes). Pseudo-toothed birds were highly specialised towards gliding flight and had an elongate and massive beak bearing hollow bony projections which were superficially tooth-like. These large seabirds existed throughout most of the Cenozoic, spanning the late Palaeocene-latest Pliocene. Pseudo-toothed birds have long been thought to be related to either pelicans and allies (Pelecaniformes) or tubenoses (Procellariiformes). Recent attempts to place pseudo-toothed birds into a phylogenetic framework proposed a sister group relationship between these and either Anseriformes (Bourdon, 2005 and 2011) or Galloanserae (Mayr, 2011).

The most important result of our brain study deals with the phylogenetic placement of the lower Eocene Dasornis, one of the oldest representatives of the pseudo-toothed birds. Dasornis shares with other neognaths the presence of a separate canal for the glossopharyngeal (ninenth) cranial nerve, which exits via a distinct foramen in the ventral part of the exoccipital bone. The brain of Dasornis is generalised in morphology, and the shape of the telencephalon and wulst shows that this early pseudo-toothed bird cannot fit into any of the neognathous taxa included in the study. We show that Dasornis is the sister taxon of Neoaves, which constitutes a very large clade of birds including pelecaniforms, tubenoses, as well as most aquatic birds and 'higher land birds' (Livezey and Zusi, 2007; Hackett et al., 2008). The new clade Dasornis and Neoaves is supported by 3 synapomorphies which are unique to this group: the rhombencephalon (hindbrain) is square in shape and strongly convex, by contrast with non avian theropod dinosaurs, Enaliornis, palaeognaths and Galloanserae. The abducens nerve (sixth cranial nerve) is widely separated from its counterpart and located near the rostral corner of the hindbrain in Dasornis and Neoaves. Outside this clade, the nerve is in more caudal and medial position. In Dasornis and Neoaves, the palatine bone broadly contacts (sheathes) the parasphenoid rostrum. In contrast, a palatorostral articulation is absent in outgroups and palaeognaths, and small in Galloanserae.

A well developed wulst delimited by a vallecula occurs in most palaeognaths, galloanserines as well as many neoavian taxa, and our phylogeny defines this feature as a synapomorphy of modern birds. In this context, the rudimentary wulst of Dasornis is interpreted here as a secondary reduction, rather than an early developmental stage (Milner and Walsh, 2009). Similarities between pseudo-toothed birds and either galloanserines or neoavian seabirds are due to convergence.

The avian fossil Vegavis from the upper Cretaceous of Antarctica is the first Mesozoic neornithine which has been confidently placed in a phylogenetic framework (Clarke et al., 2005). Its anseriform affinities set the minimal age of origin of Galloanserae and the minimal age of diversification of neognathous birds as upper Cretaceous, these cladogenetic events occurring earlier in time than Maastrichtian. Our brain study shows that pseudo-toothed birds have a more recent origin than previously thought. However, they originate earlier than all neoavian taxa, including most extant seabirds, and the early Eocene Prophaethon and Halcyornis. This indicates that stem-group representatives of the pseudo-toothed birds lacking some specialisations such as skeletal adaptations for gliding flight, presence of pseudo-teeth, and rudimentary wulst, must have been present in the Upper Cretaceous.


This project improves our knowledge of avian brain evolution and the diversification of birds through geological time, and will be a major contribution to the understanding of the history of life on earth. Additionally, this project will provide biologists with a solid working basis for better understanding how today's avian biodiversity might respond to environmental change and global warming. This will ultimately have an impact on conservation biology in the European area and around the world.

Contact details:

Dr Estelle Bourdon
Department of Earth Sciences
The Natural History Museum
Cromwell Road
London SW7 5BD
Tel: +44-(0)20-79425845
Personal webpage:

Scientist in charge:
Dr Angela C. Milner
Scientific Associate
Department of Earth Sciences
The Natural History Museum
Cromwell Road
London SW7 5BD
Tel: +44-020-79425028
Fax: +44-020-79425546