Using Artec Space Spider to discover the true evolutionary path of the extinct thylacine

Male and female thylacines in the National Zoo, Washington, D.C., by E.J.K. Baker, c. 1904. Image in public domain. Colorized by D.S. Rovinsky.

Despite being called the Tasmanian “wolf,” because of its striking physical similarities, the thylacine wasn’t one at all. Yet its resemblance is so strong that almost everyone, scientists and public alike, just naturally assumed that it had evolved along a very close trajectory to that of the wolf.

But one team of evolutionary biologists thought otherwise, and they circled the globe in their quest to reveal the evolutionary truth about this mysterious creature.

With their Artec 3D scanner in hand, Monash University researchers Dr. Douglass Rovinsky and Dr. Justin W. Adams visited museums and university collections around the globe, digitally capturing the hundreds of specimens of various species needed for the most comprehensive study of its kind ever conducted, one that would rigorously test whether the thylacine was in fact evolutionarily convergent with wolves.

Convergent evolution

In biology, convergent evolution simply refers to the process where unrelated organisms without a recent common ancestor independently evolve similar traits after having to adapt to similar environments or ecological conditions.

To put it another way, if two different species are convergent, they can appear genetically related, when in fact they’re not.

Examples of convergent species:

Turtles and snails: both have evolved hard shells to protect them from predators, yet while turtles are reptiles (like lizards and snakes), snails are mollusks (like shipworms, clams, and octopuses).

Squids and cockatoos: while one swims and the other flies, they’ve both separately evolved hard beaks for tearing apart their prey – whether that’s a squid feasting on a deep-sea crab, or a cockatoo chomping down on a peanut or slow-moving insect.

In the case of the thylacine, for more than a century, most researchers just conveniently assumed that the overall “dog-like” body form meant that it was ecologically similar to the wolf – particularly because it didn’t resemble other groups of marsupials that were being discovered and described.

And numerous studies began with this very assumption in mind, with few academics ever questioning the interpretation that thylacines were essentially just wolves with pouches.

From body mass to convergence

Facing such widespread assumptions about this remarkable creature that went extinct nearly a century ago, Rovinsky and Adams set out to methodically delineate an accurate and far more complete understanding of the thylacine and its now-lost way of life.

As Adams explained, “Unfortunately, back when the thylacine was still with us, no one ever recorded basic natural history data about the animal, this includes information about its body mass, dietary preferences, predation strategies, locomotor habits, and habitat preferences. Without these details, it’s impossible to even begin to understand what kind of creature the thylacine truly was.”

Researcher Douglass S. Rovinsky scanning a thylacine skull with Space Spider and Artec Studio software.

Building upon their prior study that established a significantly different average body mass of the thylacine than had been assumed or used across research studies, in their latest work, Rovinsky and Adams deeply investigated and analyzed patterns of similarity between the thylacine and other species in three distinct areas: cranial shape, diet, and relative prey size.

Taking everything into account, the present study makes clear that there is no basis for interpreting the shape of the thylacine skull as convergent with wolves, and as such, the thylacine would not have exhibited similar feeding habits or dietary ecology.

Instead, Rovinsky and Adams determined that the thylacine exhibited convergence with a diverse range of canids including African jackals (Lupulella adusta, Lupulella mesomelas) and certain South American foxes, including the Pampas fox (Lycalopex gymnocercus) and the Maned wolf (Chrysocyon brachyurus).

3D scanning for capturing cranial data

Skull overlays: The average skull shape of the gray wolf (blue), thylacine (pink), and the significantly convergent group of canids (green) overlaid in Artec Studio software so the differences in shape can be seen. The convergent group shape is the average skull shape of four different species of canid, the maned wolf, black-backed and side-striped jackals, and the Pampas fox.

In order to accurately measure and study the cranial shapes of this rare, recently-extinct species, most specimens of which are held in collections with limited and controlled access, a non-destructive method of capturing data down into the submillimeter range was called for.

Rovinsky commented, “When we’re given access to a certain specimen at a museum, the last thing the curator wants is to worry about any scratches or other damage from calipers or manual measurement devices, not to mention excessive handling for repositioning. And placing any markers or targets on these specimens would be unthinkable.”

The Artec Space Spider

This is what made their choice of 3D scanner for the project all the more pivotal. Relying upon their Artec Space Spider, a lightweight, handheld color 3D scanner that captures up to one million data points per second, with an accuracy of 0.05 mm (the diameter of a human hair), Rovinsky and Adams captured the skulls of specimens in mere minutes.

Commenting on how the Space Spider compares to the traditional use of calipers and rulers in paleontology, Adams said, “The thylacine, as well as other extinct species, and any living creatures are not two-dimensional (2D) entities.”

He continued, “So if we try to describe the shape of an organism using 2D measurements, it will never be able to capture the subtle ways biological form varies and adapts to specific functions. That’s why our Space Spider scanner was of the utmost importance for this study.”

Altogether, the crania of 223 animals from 57 faunivorous species were scanned, including thylacines, hyenas, civets, mongooses, quolls, fossa, dogs, raccoons, and more.

Phylogenetic tree: The representative skulls of the 57 species used in the study, arranged by phylogeny (evolutionary relatedness) and colored by family (e.g., felids are green, canids are light blue, etc.).

The large variety of species were brought into the workflow to ensure that the dataset and complete analysis would be adequately balanced.

From an evolutionary perspective, including other carnivorous marsupials besides the thylacine, such as quolls, was imperative, along with a selection of other small carnivores, in particular, weasels, civets, and mongooses.

The scans were then transformed into 3D models in Artec Studio software and exported to other software, including Geomagic Studio (the predecessor of Geomagic Wrap), for quantitative analyses of 3D shape across the skulls of the selected species.

From scanning to dataset analyses

Screenshot of Geomagic Studio software, showing mesh re-orientation and alignment prior to use in the 3D geometric morphometrics program.

Following the scanning, the subsequent 3D Geometric Morphometrics (shape) analyses required digitally placing 381 anatomical landmarks across the surface of each 3D cranium.

These landmarks were used to accurately identify and capture the cranium’s distinct shape characteristics, and in conjunction with quantitative analyses, allowed the researchers to address many issues relating to its shape and form.

Landmark template created on the approximately average-shape cranium of the dhole (Cuon alpinus), shown on the side (a), top (b), and bottom (c). The curve and surface semilandmarks allow data collection on otherwise “featureless” areas of the cranium.

Conclusions and new revelations

The results of the analyses and convergence tests complement prior questions about the dietary ecology of thylacines posed by researchers who had previously looked at bite strength in the species.

The thylacine simply wasn’t built for taking down larger animals. In order to do this, an animal needs a robust jaw, and the thylacine didn’t have this.

The respectable bite strength required for tearing into a large mammal, as wolves are known to do, simply wouldn’t have been possible with the thylacine’s more delicate cranial structure.

On the other hand, the data shows significant support for such convergence between the thylacine and African jackals as well as South American foxes, with less support for convergence with coyotes and red foxes.

But, as Rovinsky and Adams reinforce in their study, the thylacine may exhibit convergence with these canids that are adapted to consume small prey – but they aren’t identical to them either.

These research results, combined with results from prior studies of the anatomy of the species, continue to underscore that the thylacine was a truly unique animal in its own unique category.

Artistic rendition of the thylacine by Damir Martin

The thylacine was a creature with a distinct history stretching back to the Oligocene epoch, more than 23 million years ago, and was neither a “marsupial wolf” nor a “marsupial tiger,” and not a “marsupial jackal” for that matter.

In Rovinsky’s words, “After intentionally erasing this creature from the planet, the least we can do is give it the respect it finally deserves.”

Moving forward, exploring the past

Future plans are in place for a companion study focused on exploring the lower jaw (mandible) of the thylacine, to show an even tighter relationship to feeding behavior.

Additional studies include projects to examine the thylacine’s elbows and feet, to see how clearly we can begin to understand the animal’s movement patterns and ways of capturing prey.

Rovinsky added, “We still have much to add to the picture of the thylacine. I hope that future researchers will go on to explore other facets of convergence in regards to the physical characteristics of the thylacine, as this will help us further refine our understanding of this lost animal.”