The Hummingbird-Sized Dinosaur Story

A tale of how science works

Oculudentavis, tiny dinosaur trapped in amber
Amber specimen with Oculudentavis preserved inside

“Every proposed evolutionary relationship is hypothetical (even the notion that humans are mammals!), and they get tested all the evidence may show that Oculudentavis isn't a lizard after all, but some kind of lizard-like and bird-like lineage of reptiles that lived in the Cretaceous.”   Dr. Luis Chiappe

The bird-like skull was minute, about the size of a hummingbird’s, and encased in amber. The row of tiny teeth stretching to just under the very large eye sockets inspired the name "Oculudentavis", and this new discovery was first published in the journal Nature.

The international team of scientists, including Dr. Luis Chiappe, NHMLAC’s Senior Vice President of Research and Collections and Director of the Dinosaur Institute, as well as research associates and longtime colleagues Drs. Jingmai O’Connor and Lars Schmitz, believed they had potentially uncovered a hummingbird-sized, Cretaceous bird, and because all birds are thought to be a branch of dinosaurs, ultimately a diminutive dinosaur. Nonetheless, a subsequent discovery upended their interpretation of Oculudentavis as a bird (and a dinosaur), leading scientists to conclude the specimen to be more akin to lizards. This week, Nature issued a retraction in light of this new evidence. 

Creating new knowledge is a process inherently full of churn: new discoveries overturn previous understandings, conclusions are revisited, ongoing research produces new information. This is a story of how science works. Please join us for a discussion with Dr. Chiappe about the nature of scientific discovery and the story of Oculudentavis below. 


What do you think the most useful lens for a lay person looking at this would be?

Well, NHMLAC is also a research institution. We have a large team of researchers that are making discoveries and creating new knowledge. They propose scientific hypotheses about the world around us and its past with the understanding that that knowledge changes as we gather new evidence. That's the nature of science. To be very honest, I think it's about more than science; it's the nature of how we live our lives--we are always taking evidence into account.

We learn things that we didn't think about before that inform the way we live our lives. Science is no exception. Scientists are making hypotheses based on the evidence they have all the time, and then when new evidence is discovered, those hypotheses are tested. Sometimes those hypotheses survive the test and continue to be viable, and sometimes they don't, they get rejected by the new evidence and replaced by new hypotheses.  

When the hypothesis claiming that Oculudentavis was a tiny dinosaur was proposed, the research team used the available evidence and modern methods to reach that conclusion.  The analysis the team performed, using only the available evidence, was sound, and as such it went through the rigorous peer-review process that is required by high-profile scientific journals prior to publication. What the team could not imagine back then, was the enormous amount of evolutionary convergence [when unrelated species evolved parallel adaptations, like wings in bats and birds, for example.] that was hidden, because the team only had a skull available (the only part of the body of the animal that was contained in the amber piece included in the study). 

Are there classic examples in dinosaur paleontology where our understanding has changed?

Yes, of course. For example, when T. rex was discovered, those early discoveries did not include the arms, so if you look at the early mounts from the first half of the 20th century, T. rex is mounted with three fingers, like most typical theropod dinosaurs. Those mounts reflected the evolutionary hypothesis that T. rex had the arms typical to its other theropod dinosaur relatives. But later, when the arms were recovered with more complete specimens, it was found that T. rex only had two fingers at the end of very short arms. Today, most people don’t realize that the tiny, two-fingered T. Rex arms that have become so iconic also tell an interesting story of scientific hypotheses being modified as a result of the discovery of new evidence. 

Do you want to briefly talk about clades, evolutionary relationships and convergent evolution?  

While many well established evolutionary relationships are accepted as facts, the reality is that they're still hypothetical in nature. We could say that the hypothesis that humans are mammals is immensely strong, but regardless of the strength that supports such a notion, it's still a hypothesis. The same way that when we say birds are dinosaurs, there's a whole body of evidence nowadays supporting the idea that birds are nested within dinosaurs, but it's still a hypothesis.

All evolutionary relationships are hypothetical, and they get tested all the time. And of course, some of them, like in the case of Oculudentavis, lose support in light of new evidence. This new evidence appears to indicate that Oculudentavis is in fact a bird-like lizard.  But that’s not the final word either. That's still a hypothesis, and new evidence may show that Oculudentavis isn't a lizard, but some kind of a lizard-like and bird-like lineage of reptiles that lived in the Cretaceous. 

Could I get a couple examples of convergent evolution?

The evolution of life is filled with examples of evolutionary convergence (when unrelated animals or plants evolve similar traits) at different levels of specificity. For example, birds, bats, and the extinct pterosaurs have wings that they use in flight. Yet, these are three unrelated groups of animals, and their wings evolved from wingless forelimbs independently. The finned and streamlined bodies of sharks and dolphins are another classic example of convergent evolution, both adaptations of proficient swimmers.  More specifically, the complex eye of vertebrate animals (backboned) and that of octopuses and squids share many similar anatomical details, which again, evolved from simpler eyes independently. Convergent evolution is everywhere in the history of life.

You're dealing with things from the far distant past that can kind of be upended relatively quickly. 

Yeah. It’s just not fun to think that you put a lot of work and time in studying something, analyzing something, getting to a conclusion, an exciting conclusion in this case, and then, realizing that your hypothesis is being demolished by new evidence (and in this case, so quickly!). But that’s the nature of science and the way we learn. If you are not ready for that, or if you are not open-minded enough to embrace new evidence even if it completely changes your previous interpretations, science is not for you!