Six images of bird skulls with coloured dots showing data points.

The team used computer modelling to track tiny changes in bird skulls. Image: Ryan Felice.

Read later


During Beta testing articles may only be saved for seven days.

What birds can teach us about evolution

A new study is helping to explain how evolution works.

Evolution is the process of living things changing over time. Over billions of years, modern humans can trace their development from single-celled organisms to vertebrates and much later, to apes and modern humans.

Similarly, all other living organisms have an evolutionary history stretching back millions or billions of years.

Scientists have been trying to decipher the intricacies of this process for centuries. A research team in the Museum, led by Prof Anjali Goswami, is working to find out more about how and why we evolved the way that we have.

Their latest paper, authored by Anjali and postdoctoral researchers Ryan Felice and Marcela Randau, is suggesting a new model for how we can expect evolution to occur.

When you look at the vast diversity of life on Earth, from bees to blue whales, it might feel like evolution can occur in limitless, random directions. But the team say that is not the case, because there are constraints on what evolution can do. When we think about how the animals of today might evolve in the future, it may be that nature only has a limited number of options available.

Anjali says, 'Very early on in a group's development, nearly anything is possible and there is usually a burst of diversification. But then as animals develop, limitations arrive and groups get locked into pathways of evolution which dictate what happens afterwards.

'Developmental programmes dictate how things can and can't evolve.'

Simply the best (skull)

Take for instance the thylacine (or Tasmanian tiger), an extinct carnivorous marsupial. Its skull looks like that of a dog.

But thylacines and domestic dogs are separated by 160 million years of evolution, and are not closely related. So how did they end up with virtually identical skulls?

It could be argued that this similarity is solely because both animals evolved to be the best possible predator.

However, Anjali and her team argue it is far more likely that rather than there being an optimum way to be a predatory mammal, there are actually relatively limited ways that predator skulls can evolve. Animals are a product of their very early development, so they only have certain evolutionary options open to them.

So for predatory mammals, this pathway usually ends up with this type of skull.

The team used a recently published dataset composed of surface geometric morphometric data. Image: Ryan Felice.


Scanning birds for answers

Even when evolution happens quickly, like after a mass extinction, we might see that the same forms are evolved over and over again, rather than many new types evolving. 

To demonstrate this, Anjali and her team studied bird skulls. They used surface and CT scanners to create 3D models of 352 bird species. Detailed scans allowed them to map different parts of the bird's skulls and examine how each part was constrained or facilitated in its evolution.

Ryan, who is leading the bird study, explains, 'The scans show the shape of the birds' skulls in a lot of detail. Having those allowed us to place digital markers all over the skulls, which captures the shape of the skull mathematically.

'Then we can identify changes to bird skulls, like the way the very tip of the snout has changed over time. This way, we can clearly see which parts are variable and which ones haven't changed much at all.'

The team found that the structure of the skull is divided up into parts, each with a different evolutionary history and future.

For example, bird beaks can evolve much more quickly than the part that holds their brain. Parts of the bird skull that have a tight constraint on them - like where the skull and neck join together - evolve slowly and with little variety in shape.

However, parts of the skull that are not constrained as much, like the beak, seem to be creating new shapes more often.

Various types of bird skull against an evolutionary tree, in which the colours represent the rate of evolution.

Various types of bird skull against an evolutionary tree, in which the colours represent the rate of evolution. Image: Ryan Felice.


As well as this, the parts of the skull that are evolving fastest tend to be made of specific types of tissue. Anything made of older, ancestral tissue types are evolving more slowly, whereas bones made up of a mix of tissues evolve the fastest. This suggests that diversity in developmental origin creates diversity in form.

With such a large dataset, the team can use their findings to make certain assumptions about how evolution might occur in birds and other vertebrates. They suggest that different parts of an animal can evolve at vastly different rates, but those parts which have constraints on them can only change in specific ways.

The scanning of the bird specimens for this study is part of a much larger project at the Museum, which is documenting the skulls of all terrestrial vertebrate families. More than 2,000 specimens have been scanned to date, and all of these scans are being made available once they are analysed.

In 2017, Anjali's team used these scans and their analyses to try to reconstruct the ancestor of living birds. This hypothesis of what the long-extinct ancestor might have looked like is also available as a 3D scan that can be downloaded and studied to improve future studies.