The Skeleton in the Closet


After a few ups and downs, everything you always wanted to know about the effect of missing data on recovering topology using a Total Evidence approach is now available online (Open Access)!

This paper also treats many different questions that people might be interested in (Bayesian vs. ML; how to compare tree topologies; comparing entire distributions, not only their means and variance; and many more!) but I’ll leave it to you to discover it…

Back on track, more than one an a half CPU centuries of calculation ago, Natalie and myself wanted to build a Total Evidence tip-dated primates tree. The Total Evidence method is the method that allows you to combine both living and fossil species (or actually, read “both molecular and morphological data”) into the same phylogenies. The tip-dating method, is an additional method that uses the age of the tips rather than the age of the nodes for dating such a tree. But I’m not going to talk about that in this post.

At the start of the project, we were both confident about the idea behind it and that primates would be the ideal group for such work since they are so well studied. A study that I described in a former post also came out around the same time, encouraging us and comforting us in this project.

However, as you might guess, something went wrong, horribly wrong! For the Total Evidence method, we need molecular data for living species (check) morphological data for fossils species (check) and also for living species (che… No, wait)! After looking at the available data, we quickly found out that there was a crucial lack of living taxa with available morphological data (check our preprint to be submitted to Biology Letters putting the actual numbers on the problem). From that problem, rose the idea of actually testing how that would influence our analysis. And funnily enough, this problem become one of the two major parts of my PhD!

Running thorough (and loooooong) simulations, we assessed the impact of missing data on topology when using a Total Evidence method. We looked at three parameters where data would be missing:

  1. The first one, was obviously the one I introduced above: the number of living taxa with no available morphological data (at all!).
  2. The second one, was the amount of available data in the fossil record (because yes, fossils can be a bit patchy).
  3. And the third one, the overall amount of morphological characters.


We then compared the effect of different levels of available data for each parameter individually and and their combination on recovering the correct topology, using both Maximum Likelihood and Bayesian Inference. For the correct topology, we used the tree that had no missing data in our simulations. For each parameter combination, we measured the clades in common between the correct topology and the trees with missing data as well as the placement of wild-card taxa (typically fossils jumping everywhere).

Unsurprisingly, we found that the number of living taxa with no available morphological data was the most important parameter for recovering a good topology. In fact, once you go past 50% living taxa with no morphological data, the two other parameters have no effect at all, even if you have a perfect or a really bad fossil record or many or really few characters. This is kind of intuitive when you think about it because the only way to branch the fossils to living taxa is to use the morphological data. Therefore, if there are no morphological data for the living taxa, the fossils cannot branch with them regardless of the quality of the data. Therefore, in this paper, we argue that to improve our topologies in Total Evidence, we should visit more Natural History museums. And not only the exciting fossil collections but the well curated collections of living species as well!

All the code for this paper is available on GitHub.

Check out the latest presentation about both papers.

Paper 1: Guillerme & Cooper 2015 – Effects of missing data on topological inference using a Total Evidence approach – Molecular Phylogenetic and Evolution (doi:10.1016/j.ympev.2015.08.023).

Paper 2 (preprint):  Guillerme & Cooper 2015 – Assessment of cladistic data availability for living mammals – bioRxiv ().


Author: Thomas Guillerme, guillert[at], @TGuillerme

Photo credit: Thomas Guillerme (AMNH collections)

Looks can be deceiving

Small Madagascar Hedgehog Tenrec

We are all taught not to judge a book by its cover, it’s what inside that counts. Our new paper published in PeerJ shows that the same is true for tenrecs.

These cute Madagascar natives are often used as an example of a mammal family with high morphological diversity. It’s easy to see why: there are tenrecs which resemble shrews, moles, hedgehogs and even otters. These differences are even more remarkable when you consider that tenrecs are more closely related to elephants and aardvarks than they are to any of the small, “insectivore” mammals. One of only four native mammal clades in Madagascar, it appears that tenrecs have undergone an adaptive radiation to fill otherwise vacant, small mammal niches, evolving convergent similarities to other groups in the process.

Tenrecs are clearly very diverse in their appearance. However, prior to our study, no one had tested whether this apparently high diversity was more than skin deep. We tested whether tenrecs were more morphologically diverse than their closest relatives, the golden moles.

We measured the morphology of tenrec skulls and compared their diversity to the shape of golden mole skulls. This meant spending hours poking around the collections of natural history museums and many more hours placing landmarks on skull pictures for 2D geometric morphometrics analyses (Spotify was my friend!)

Pictures of an otter shrew tenrec (Potamogale velox) skull showing that landmarks (points) and semilandmarks (curves) that we used to summarise skull shape. See the paper for more information.
Pictures of an otter shrew tenrec (Potamogale velox) skull showing that landmarks (points) and semilandmarks (curves) that we used to summarise skull shape. See the paper for more information.

Tenrecs occupy a wider range of ecological niches (fossorial, arboreal, terrestrial and aquatic) than golden moles so we expected that tenrecs would be more morphologically diverse than their cousins. However, we found that tenrec skulls only have more diverse shapes than golden moles when we compared them in lateral (sideways) view but not dorsal or ventral views. These results show the importance of measuring morphology in many different views to gain a more complete understanding of overall morphological diversity.

The tenrec family includes species which convergently resemble many other, un-related small mammals. However, most tenrecs (19 out of the 31 tenrec species in our analysis) belong to the shrew-like Microgale tenrec genus. So, although many people tend to focus on the strange and unusual species (such as the otter shrew, hedgehog tenrec or the bizarre lowland streaked tenrec), most tenrecs are small, shrewy-type creatures that look very similar.

Microgale cowani
Microgale cowani

We tested whether the similarities among the Microgale tenrecs might be masking higher morphological diversity in the rest of the family. We repeated our analyses to compare the diversity of golden mole skulls to a sub-set of the tenrec family (including just 5 of the 19 Microgale species).  In this case, tenrec skulls were more morphologically diverse than golden mole skulls in all comparisons (skulls in dorsal, ventral and lateral view).

Overall, our results indicate that, while there are clear physical differences among tenrec family members, the majority of tenrecs are quite morphologically similar to each other so morphological diversity in the family as a whole is not as big as it first appears. Of course comparing skull diversity is just one aspect of overall morphology – analysing the shape of other traits such as limbs could yield different results – but our study represents the first step towards a greater understanding of the ecological and evolutionary diversity of tenrecs.

It’s also testimony to the fact that you should never judge a tenrec by its cover.

Author: Sive Finlay, @SiveFinlay

The moral of the story


Most of us have some inbuilt sense of right and wrong; don’t steal and don’t murder are as basic to us as our ability to breathe. But where does this moral sense come from? In general, people of a scientific bent don’t attribute it to God nor as some sort of free floating truth that can be grasped by the human intellect. If you hold a materialistic view, that is to say the idea that at its base the universe is composed of energy and matter, then it’s next to impossible to understand morality in those terms. Instead the scientific view proposes our morality is an evolved feature, something which gave group-living animals a selective advantage over their amoral competitors. A social group that tries to cooperate when it’s made up primarily of murderers, thieves and cheats won’t get very far. By contrast a crowd of goodies can gain the many benefits of cooperation.

There is a problem with this theory though. Irrespective of its truth, an evolved morality renders us with a situation where there is nothing objectively right or wrong about anything. Even an act of murder isn’t intrinsically immoral. One way to think of this is to compare it with our other adaptations. We don’t consider any other evolved traits ‘moral’, it’s not as if four legs good, two legs bad is something people really espouse. What we’re left with is a moral nihilism.

‘So what?’ you might ask.  We’re a smart species, we can decide for ourselves the best way to act such that our society can flourish. Why don’t we adopt some sort of utilitarianism, the moral system that promotes the greatest happiest for the most, and judge the rightness or wrongness of our acts that way? Indeed this is the way most secular societies establish what is permissible today. This idea can even allow for the expansion of our moral circle to include other beings who are capable of suffering.

Yet the modern understanding of our selves means even a created morality still can’t fairly punish or praise for a simple reason: humans have lost their soul. Modern neuroscience tells us there is no actor in our minds making decisions moral or otherwise. We are our brains, nothing more. There is no ‘I’, no ‘ghost in the machine’. The idea of a freely willed agent who can separate his or her self from their genetics and environment is anathema to anyone who takes materialism seriously.

Much follows from this. Most notably our justice system should be radically re-evaluated in light of this idea to become more biologically informed. Currently persons with certain mental disabilities are afforded more leniency when it comes to their sentencing because they are said not to be in full possession of rational thought processes. Something has affected their ability to have done otherwise. But as automata this is true for every person who has ever existed. This is not to say that we should open up the prisons and free every criminal the world over rather that we should focus much more on promoting environments that cause people to act in a way conducive to a functioning society.

We are all of us robots acting on inputs. Some people take these inputs and act like ‘goodies’ whereas others can take the same information and behave like ‘baddies’. Take your pick of a hero or tyrant from history. They don’t deserve your respect or your contempt. That is the price of a biological morality.

Author: Adam Kane, kanead[at], @P1zPalu

Photo credit:



Evolution is – surprise! – Darwinian!


I sometime come across papers that I missed during their publication time and that shed a new light on my current research (or strengthen the already present light). Today it was Cartmill’s 2012 Evolutionary Anthropology – not open access, apologies…

Cartmill raises an interesting question from an evolutionary point of view: “How long ago did the first [insert your favorite taxa here] live?”. This question is crucial for any macroevolutionary study (or/and for the sake of getting a chance to be published in Nature). If one is studying the “rise of the age of mammals” (just for example of course) the question of the exact timing is crucial to see whether placental mammals evolved after or before the extinction of avian dinosaurs.

Because Cartmill published in Evolutionary Anthropology let’s replace [insert your favorite taxa here] with humans. He proposes to answer to the question “How long ago did the first humans live?” by looking at the different ways people have addressed it through time. It all starts back with Simpson’s quantum evolution stating that clades share “adaptive shifts” or “adaptive trends”. For example, for humans, that will be bipedalism and an increase in brain size: “Everybody can sort humans out instantly from other sorts of things: […] they share a unique reliance on technology, a capacity for culture, and a gift for gab.”

This is an unfortunate classical view of evolution based on morphological data leading to a series of morphological discontinuities – the adaptive shifts (“human origins, primate origins, mammal origins, amniote origins, and so on” – I already discussed this gradualistic view about tetrapod origins). Cartmill uses a pertinent quote from Simpson to comment this trend: “Is this not, in fact, simply a recrudescence of the old naïve conception of a scala naturae[?]”. However, this raises a cladistic problem. Assuming we have the data on the oldest human. What defines his or her humanness? “Humanness [whatever that means] is not a coherent package. We have known since the 1960s that our terrestrial bipedality evolved more than two million years before the onset of what was long held to be the fundamental human characteristic, that is the great development of the brain.’’

Cartmill’s point is that, morphologically, there is no adaptive shift or trend that can define any group. Morphological evolution acts more like a succession of slow and discrete incremental changes rather than the Simpsonian quantum model: “there is only a long, geologically slow cascade of accumulating small apomorphies” and adaptive trends or shifts within clades “are fantasies, born ultimately of our wish to see ourselves as more decisively set off from other animals than we actually are.”

Will I have to rethink my current project looking at mammals morphological evolution? Well by using molecular data as well as morphological data we can accurately trace back the small incremental changes (the DNA mutations) as well as the actual changes in morphology through time. My PhD is not just a series of failures after all!


Thomas Guillerme: guillert[at], @TGuillerme

Photo credit

Do you speak Yamnaya?

Pieter_Bruegel_the_Elder_-_The_Tower_of_Babel_(Vienna)_-_Google_Art_Project_-_editedI bet you do!

One nice non-biological thing you can do with phylogenetics (unlike beers) is study the evolution of languages. If you aren’t familiar with evolutionary linguistics, it’s basically the same principles that we use to study the descent with modification of organisms but applied to words. Even though words do not evolve in a biological way, we can still apply similar phylogenetic principles by just adjusting the evolutionary models.

OK but let’s go back to my assumption (that you do speak Yamnaya). Since you are reading this blog post that I’m trying to write in English, you do speak English which is part of the linguistic family (or clade) called the Indo-European that consists of the vast majority of the European and Indian languages spoken by a good 3 billion people (as the name originally suggests- check this excellent visual phylogenetic summary). Even though it is not straightforward to see the similarities between Icelandic and Indi, evolutionary linguistics suggest that both languages have diverged from the same language based on words and grammar similarities. This language, generically called proto-Indo-European is estimated to have originated either around 9000 years ago in the Middle-East and spread across India and Europe along with agriculture (the ‘Anatolian hypothesis’). Or, a second theory postulates its origin around 5000 years ago on the northern shores of the Black Sea and its subsequent spread along with horse riding and wheeled transport (the ‘steppe hypothesis’).

Until last month, both hypotheses were lacking data to explain some crucial temporal problems: the proto-Indo-European language contains words related to wheeled vehicles which were not invented 9000 years ago therefore potentially falsifying the ‘Anatolian hypothesis’. However, DNA studies did support it with a common ancestral population to Indo-European speakers dated around 9000 years ago. Also on the DNA side, no clear evidence for population dispersion was available for supporting a later origin and faster spread of the proto-Indo-European (the ‘steppe hypothesis’).

But that was only until this month: a recent paper by Haak along with his 39 co-authors preprinted in BioRxiv provides evidence for a common ancestral population that originated in the Ukraine and spread at into northern and western Europe. This population links in space and time with the Yamnaya culture around 4000-5000 years ago suggesting that Yamnaya was close to the proto-Indo-European culture. Even though if the ‘Anatolian hypothesis’ cannot be excluded, this new paper strongly suggests that at least the European branch of the Indo-European language originated from the Yamnaya culture (see Extended Data Figure 5 p.32 and its legend p.27 of the preprint pdf for a nice visual summary).

Therefore it is likely enough that Yamnaya was the origin of most European languages and that it spread rapidly through northern and western Europe probably due to technical advancements in transport. I find evolutionary linguistic always amazing when you can state that you wrote/read a blog post in a derived Yamnaya language: English.

Author: Thomas Guillerme, guillert[at]

Photo credit: wikimedia commons

Ecology of religious beliefs


It is well known that your country of birth has a big influence on your religious outlook. That’s why Ireland is dominated by Christians whereas Iran has a mostly Muslim population. Your scientific outlook doesn’t escape from this either. For instance, it’s arguable that the idea of group selection is viewed much more favourably in the US than the UK. Turning back to religion, a group of authors have recently carried out a study on the ecology of religious belief. In their work they were able to predict the societies that believe in moralising high Gods by drawing on historical, social and ecological data.

As we’re in the beginning, we need a definition, so what exactly is a moralizing high God? These are “supernatural beings believed to have created or govern all reality, intervene in human affairs, and enforce or support human morality”. Supernatural belief has had a number of ecological correlates associated with it and this study points to environmental instability as one major driver. An environment with an unpredictable spatial or temporal distribution of resources lends itself to cultivating cooperation among the animals found within it. Among humans this results in a “reduction in cheating, increased fairness, and a tendency to cooperate”. And this is a fertile ground for the development of religious belief. One conclusion is that cultures in close proximity or those that share a common language exhibit similar religious beliefs.

However, the authors nuance this statement that this form of religion is being driven as a response to environmental harshness. In fact, they note that for societies living in a really harsh environment like the Inuits, variations tend to lead to more positive periods rather than negative ones and this seems to inversely correlate with the probability of believing in these type of gods.

One ironic point however (largely discussed in Jerry Coyne’s blog post), is that this paper is also influenced by the authors’ societal framework: a society traditionally believing in a god that they consider as moral and improving human lives. Taking that into account, some parts of the methodology heavily influence the results. The definition of morality by the authors and its benefits on humans as a species are highly depend on the societal framework where one is born. The “reduction in cheating, increased fairness, and a tendency to cooperate” is traditionally seen as a “good” thing for humanity in Abrahamic religions. Making it a universal or biological “right” behaviour is only the authors’ point of view (and probably one of their funding agencies).


Author: Adam Kane & Thomas Guillerme, @P1zPalu and @TGuillerme

Photo credit:

Say “the rise of the age of mammals” again, I double dare you!


In biology and among biologists, we like to use terms that we know are not correct but that still come in handy when you’re confident that your interlocutor understands them the way you do. I’m thinking of terms such as “key adaptations”, “living fossils”, etc… However, among them, there is one that particularly bugs me and makes me feel like Samuel L. Jackson in the iconic Pulp Fiction scene and that is: “the rise of the age of mammals”.


Recently, Barry Lovegrove and his students published a nice data driven paper in Proceedings of the Royal Society B on the hibernation of tenrecs. The team found that these amazing creatures (I refer you to Sive’s posts, our tenrec expert) do go into hibernation for 9 months straight even though they live in tropical latitudes. The paper first sparked my curiosity because of this new tenrec fact but also due to the spin that the authors put on the paper’s results. They create a broad significance for the implications of their research on hibernation in tenrecs by describing their potential applications for how we might biologically programme astronauts to hibernate on a journey to Mars.


But what struck me the most (and I’m coming to main point of this blog post) is that the authors place their paper in the context of our understanding of the K-T boundary event which,  they argue,  was a key event in the evolutionary radiation of placental mammals (according to work by  O’Leary and colleagues discussed here and here). And from there, the authors claim that tenrecs’  “predation-avoidance   hibernation may be an ancient plesiomorphic characteristic in mammals and is a legacy, perhaps, of the 163 Myr of ecological suppression by the dinosaurs. It enabled the ancestral placentals, as well as the marsupials and monotremes […], to endure the short- and long-term devastations of the K-Pg asteroid impact, a capacity which is possibly the sole explanation for the existence of mammals today.”

This suggestion is based solely on their findings about hibernation in tenrecs. Their rather crude extrapolations to what these results tell us about the origin of placental mammals are mainly based on two erroneous assumptions:

-(1) they “report a plesiomorphic (ancestral) capacity for long-term hibernation that exists in an extant, phylogenetically basal, tropical placental mammal, the common tenrec”

– (2) “The long ca. 160 Myr stint of the nocturnal, small, insectivorous mammal was over, and gave way to the age of the mammals, the Cenozoic” because of the “Ecological release from the vice grip which the dinosaurs held over Mesozoic mammals”

(1) Tenrecs are Afrotherians. They are a sister group of golden moles and nested somewhere within the elephant shrews and aardvarks clade that is sister to the elephants, hyrax and sirenians (so, even within Afrotherians, tenrecs are not particularly basal). Afrotherians are a sister group to either Xenarthrans or Boreoeutherians (depending on the genomic region) but all three together form the extant eutherians. It is mistaken to interpret tenrecs as a “basal” mammal clade.  The authors claim that their “hibernation data show some affinities to the ‘protoendothermy’ first noted in echidnas, suggesting retention of plesiomorphic characteristics of hibernation on Madagascar through phylogenetic inertia”. This implies that this hibernation characteristic has been lost in all other mammal groups .Using basic principles of parsimony, it would make more sense to attribute this hibernation characteristic to being yet another example of a  convergent trait in tenrecs, not an ancestral state which was lost in most other mammals.

(2) I grew up reading a steady diet of books about the history of life. These presented a nicely summarized picture: around 65.5 Mya (now updated to 66 Mya, a small detail that can easily be fixed), all dinosaurs went extinct because they were too big and too stupid and the clever small mammals survived without any problem, liberated from their domination by the big stupid dinosaurs. This vision was awesome as a child; it had all the elements for a really anthropocentric/biblical view of the story (think about the Exodus) and clearly explained why mammals are the dominant species today. It even explains the success of humans: we used our cooperation and intelligence to reach our dominant position as head of all the mammals.

However, this romantic vision of the history of life (driven by paleontological data prior to the amazing discoveries of new Jurassic and Cretaceous mammals in the 90s and 2000s and the advent of molecular phylogenies) has, thankfully, been updated to integrate the last two decades of excellent work. This lead to a picture that is less romantic and more complex. The dinosaurs didn’t really disappear and are actually still more numerous (species richness-wise) than mammals nowadays. Similarly, the placental mammals and their ancestor didn’t just “bloom” after the K-T boundary event: they had their origin back in the late Jurassic, roughly at the same time as the dominant tetrapods of today (the birds), and they radiated multiple times: mainly due to global climatic changes during the Paleogene such as the “Grande Coupure” or the PETM

I’m not sure why the authors chose to adopt an outdated vision of the evolution of mammals to introduce their work but it seems a pity to me that such a spin is necessary to present good work on unknown/understudied groups even if they’re as cool as tenrecs!

Author: Thomas Guillerme, guillert[at], @TGuillerme

Photo credit:

Night Life! Friday 26th Sept

Night Life no writing

This Friday, members of EcoEvo@TCD, as well as others from the Botany and Zoology departments and Trinity Centre for Biodiversity Research will present Night Life! in the Zoology building at Trinity College Dublin. The event is FREE to attend and we will be open from 6pm-10pm with the last entry at 9.30pm.

Night Life! is an opportunity to meet researchers and to find out the kinds of things we do. Prof. Yvonne Buckley will give you a taste of our research highlights, Kevin Healy will wow you with his research on snake venom (yes there will be snakes!), Sive Finlay will perplex you with the mysteries of tenrec evolution (if you don’t know what they are, come along and find out, they’re really cute!), Sean Kelly will explain how he discovers new bird species in Indonesia, Deirdre McClean will reveal the fascinating social lives of microbes, Thomas Guillerme will dazzle you with the lasers on his 3D scanner and the jaws of a shark, Claire Shea will amaze you by explaining why babies kick in the womb, Adam Kane will intrigue you with models of T.rex and maybe some vultures, and other students will be available to answer your burning questions about biology, evolution and ecology. So if you’re at a loose end on Friday night, come along and say hi!

Night Life! forms just one part of Discover Research Dublin, an annual event funded by the European Commission as part of European Researchers’ Night. The event is hosted by Trinity College Dublin, in partnership with the Royal College of Surgeons in Ireland. As well as Night Life! the evening will feature over 50 fun, interactive and free events and activities which will give you direct contact with researchers and allow for discovery, questions and participation. The event aims to challenge perceptions about researchers and show the creativity and innovation that exists in research across all disciplines. Activities are grouped under four broad themes – Body Parts, Creativity in Research, Meet the Researchers and Living Thought/Thinking Life.
We encourage you to visit, explore, discover and enjoy!

Author: Natalie Cooper, @nhcooper123
Image: Kevin Healy, @healyke

We’re back!


It’s that time of year again, the quiet before the storm of Fresher’s week and the start of a new academic year.

After our short break, EcoEvo@TCD is back and raring to go. You can expect lots more posts about our research, seminar series, outreach activitiesconferences and fieldwork as well as tips and tricks for surviving in academia.

We’ve already kicked off the year with our second annual NERD club AGM. It was a great opportunity to discuss what we covered throughout the year and to make plans for the months ahead. (For the uninitiated, NERD club is our networks in ecology/evolution research discussion group but feel free to think of it in the true sense of the word too).

Here’s our NERD club prize winners for 2013/14

Best session: Paul for his talk on carnivory in plants

Best blog: Adam for “Flatland” and the “Heat and Light of Science Communication

Best pun (aka the McMahon and Kane memorial punning prize): Adam and Thomas for “Gould Mine”

Contributor of the year: Kevin

Best PI called Andrew: What is Andrew Jackson? (No-one knows!)

We’ve had some excellent sessions about transferable skills, how to navigate the perils of academia and great discussions and collaborations on current research projects. We’ve got lots more interesting topics planned for the year ahead which will definitely make an appearance on EcoEvo@TCD.

It’s also been a very successful year for our blog. We had a winner at the ABSW science writers awards and two semi-finalists in another international blog competition. We’re also on the short list for the best science and technology blog in the Irish blog awards.

So whether you’re packing away your fieldwork gear after another season, dreading the darker evenings or sharpening your pencils for another academic year, rest assured that you can look forward to some more ecology and evolution- related musings from the EcoEvo@TCD team.

Author: Sive Finlay, @SiveFinlay

Image source


Dig for victory

In a previous post I showed what I think being a palaeontologist is all about, especially the point that palaeontologists are different from oryctologists. The first ones study changes of biodiversity through time, the second ones extract fossils (but again, both are far from exclusive).

Here is a short summary of  experience working at Upper Cretaceous excavation sites in the South of France (that’s around 80-65 million years old) namely in the Bellevue excavation site in Esperaza run by the Musée des Dinosaures.

First step is to find a place to dig.

Step 1.1: find something

Why along the road? It doesn’t have to be but it has two clear advantages: you can park your car next to it and it’s usually rich in fresh outcrops of rock (where you can find more fossils than in a crop field!).

Step 1.2: try again and again!

The second step, once you’ve decided that there might be something in the outcrop you’ve just explored, is to remove all the “annoying stuff”. To palaeontologists that obviously means all the wonderful fauna and flora and their associated environment (usually soil) that are growing above the potential fossiliferous site (how rude of them!).

Step 2: remove all the annoying stuff

Once you’ve removed the layer of living stuff, you can start the long and interesting part: hitting rocks with a hammer and a pike during the hottest days of summer.

Step 3: start hitting the rocks
Step 4: find something (hopefully!)

Finally, with a bit (a huge bit) of luck, you’ll find a fossil that was worth all this hassle.

Step 5.1: clean the fossil

Once you’ve found the fossil, the first step is to clean the surface facing you and start to build a trench around it in order to pour plaster over it and bring it to the lab. As you can see, paint brushes are useless here too: the hammer and the pike make ideal tools for the surrounding trench and an oyster knife and a smaller hammer do the cleaning jobs. Oh yeah, and a tube of glue. After around 80 million years, the bones get a bit fragile.

Step 5.2: clean the fossil… again!

The last step is to properly clean the fossil in the lab by removing it from all the surrounding rock. The best tools are mini pneumatic-drills and loads of patience. When all that is done, the palaeontologist can start to work on the fossil.

You can find more impressive pictures on the Musée des Dinosaures webpage.

Author: Thomas Guillerme, guillert[at], @TGuillerme

Images: Thomas Guillerme and Sébastien Enault (with the kind authorisation of Jean Le Loeuff). Feature image: