Skip to main content

The Dinosaurs of Crystal Palace: Among the Most Accurate Renditions of Prehistoric Life Ever Made

Often derided, and today somewhat neglected and forlorn, the famous prehistoric animal models of Crystal Palace in London have a lot to teach us...

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


Among the most famous depictions of prehistoric animals ever are those still on display at Crystal Palace in Sydenham, London. These life-sized constructions – today designated as Grade I listed buildings because no-one knows quite how to classify them – form part of a landscaped ‘prehistoric park’, properly the Dinosaur Court, created during the early 1850s to accompany the relocation of the Crystal Palace from Hyde Park to suburban Sydenham.

Jurassic and Cretaceous dinosaurs (at far left) and Jurassic marine reptiles (at right), humans for scale. Compare the sorry pile of rocks and boring edge of the watercourse to those shown in the scene below. Credit: Darren Naish

When opened in 1854, Crystal Palace was a dazzling public attraction featuring displays of artistry, engineering, craftwork and gardening, a zoo, a marine aquarium and ornamental, landscaped gardens with statues and fountains. Millions visited. Its success waned at the turn of the century as less effort was made to add new exhibits and displays and as the face of entertainment in the London area underwent general reform. By 1909 the company owning the palace was facing bankruptcy and it was eventually abandoned, sadly falling into disrepair over the following years. The Crystal Palace itself burnt down in November 1936 and the remaining metallic frame and outbuildings were demolished. However, the prehistoric park remained and is still in-situ today.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


My own introduction to Crystal Palace and its prehistoric animal models: a children's book that I got my mother to buy for me. I was probably 8 years old or so. Credit: PUFFIN BOOKS

Most of the things typically said of the Crystal Park prehistoric animals are familiar and have been said a million times before. Yeah yeah, they’re totally inaccurate in view of what we know today… blah blah, dinner party held inside the body cavity of one of the Iguanodon models, Iguanodon thumb spike, blah blah blah… yaaaawn. Fact is, there’s a much more interesting way of looking at the models and at what they represent. You have to see them as an outstanding example – perhaps the ultimate example – of an exciting, world-changing, public outreach campaign involving innovation, novelty, pluck and clever design on a scale not seen before, or indeed since (McCarthy & Gilbert 1994, Doyle 2008). And, by the way, people have been using the “of course, the models are completely inaccurate” line since at least the 1890s (Hutchinson 1892, Becker 1911). I’m not saying it’s untrue, but… read on.

The two Iguanodon and (at right) the shoulder-humped Megalosaurus as seen from afar. "Of course, we now know that these dinosaurs didn't look like this". Yes, I know, it's the only thing anyone ever says about these models. Credit: Darren Naish

Preamble; the cast, the setting, the 20th century ‘remodelling’. Exactly who first came up with the idea of a prehistoric park is still not entirely clear. Suggestions include Britain’s premiere palaeontologist and anatomist Richard Owen (later – in 1883 – to become Sir Richard Owen), Prince Albert (Albert and Queen Victoria, incidentally, were frequent visitors of Crystal Palace), and Joseph Paxton, a visionary designer already behind several other reconstructed, landscaped geological features. It was obviously Owen’s work on the fossil animals themselves that became so magnificently depicted in life-sized form.

Gideon Mantell – initial describer of a few of the animals and sometime competitor to Owen – was approached as a consultant in 1852 but his chronically ill health (he died later that year) meant that he was forced to decline (Cadbury 2001).

In its heydey, the Crystal Palace grounds and surrounds would have been truly spectacular. This is the famous George Baxter print of 1854, actually created before the park opened and thus featuring some hypothetical or only loosely accurate details. Credit: WELLCOME LIBRARY, London

The Crystal Palace models variously depict dinosaurs (just Iguanodon, Megalosaurus and Hylaeosaurus: the three founder members of Owen’s Dinosauria), pterosaurs large and small, Mesozoic marine reptiles (ichthyosaurs, plesiosaurs, crocodyliforms, mosasaurs), dicynodonts, temnospondyls and Cenozoic mammals old and young (palaeotheres, anoplotheres, deer, a sloth). There are also model plants (some of which were made of wrought iron), specially placed geological features (more on these below) and model geological sections (ditto).

Pleistocene megamammals at Crystal Palace: the sloth Megatherium (supported by a tree and restored with a short proboscis) and the deer Megaloceros (male, female and juvenile, restored as 'giant red deer' as was thought correct at the time, originally installed with actual fossil antlers). Credit: Darren Naish

The models are positioned around a series of islands situated in a lake. Today, it looks as if the amphibious animals (remember: ichthyosaurs and plesiosaurs were thought to be amphibious at the time, not fully aquatic) were always meant to be semi-submerged and at the water’s edge. However, they were originally positioned on a rocky beach adjacent to a promontory, and were thus fully exposed for at least part of the time. This isn’t at all obvious today as much of the original, carefully designed landscaping was destroyed and removed during a disastrous 1960s ‘renovation’ that aimed to remodel the watercourse.

The Jurassic 'beach' section and promontory as depicted in an old postcard. Compare and contrast with its horrendously different appearance today. Note also the clever positioning of the trees and the model pterosaurs. Credit: Public Domain

Indeed, we must lament the fact that much of the park was damaged beyond repair and even outright destroyed as part of a project of late 20th century modification. The entire section adjacent to the water’s edge was ruined, and a reconstructed limestone cliff associated with a tunnel and cave (featuring model stalactites, mineral veins and crystals) were destroyed (Doyle 2008). The cliffs have been restored in recent years. Another cliff features a model Carboniferous coal face with an unconformity and other stratigraphic features.

At top: the original, highly naturalistic, accurate coal measure cliff face, from an old photo. Below: as it appears today. It now looks much less realistic: I'm unsure if this is because we're now looking at a reconstruction. Credit: top: Doyle 2008, bottom: Darren Naish

The rocky promontory located next to the beach section was deliberately designed to illustrate the succession of (mostly Jurassic) rocks that had yielded many of the featured animals. In fact, education as goes geological processes, different rock types and the stratigraphic sequence of Great Britain was a major aim of the entire park. Not only were there those model cliffs and other geological features, the animal models were (and sometimes still are) positioned on or adjacent to rocks that actually represent the very same rocks their fossils come from. The animals and rocks are also arranged in geological sequence, there being an obvious break that demarcates the Mesozoic section from the Cenozoic one. By the way, visitors weren’t meant to figure all of this out for themselves – there was an official guidebook written by Owen himself.

The Crystal Palace animals are arranged in geological sequence and set within context on an actual geological column. This illustration is by B. W. Hawkins. Credit: CRYSTAL PALACE DINOSAURS

A huge, elaborate system of fountains – the largest of which sent water skywards some 76 m – were located in nearby man-made lakes and meant that the water level was designed to rise and fall such that those creatures positioned at the water’s edge appeared to appear and disappear. Unfortunately the extensive system of piping involved was prone to clogging and the entire lot was eventually dismantled and removed. Giant water towers held some of the water; they were demolished after the fire of 1936.

The Iguanodons have been given several different paintjobs over the years. I think they were even red at one point. By the way, the block of rock is genuine Wealden sandstone. Note also the addition of the cycad trunk. Credit: left: Casliber Wikimedia Public Domain, right: Jes Flickr(CC BY-SA 2.0)

I’ve visited the Crystal Palace models on numerous occasions and have consistently failed to photograph them as well as I should have. They’ve changed somewhat over time due to weathering and different paint schemes; in addition, they have at times been partially obscured due to adjacent plant growth and have also, sadly, suffered from vandalism.

Yup, vandalism. The park is permanently open to the public and the models have frequently been damaged; the smaller ones have even been stolen. One of the palaeotheres is currently (August 2016) headless due to the actions of the great British public (the head was pulled off and thrown into the lake), the megalosaur’s tail tip was broken off at some point, and certain of the other models have cracks and breaks due to mistreatment.

There are three palaeotheres at Crystal Palace, definitely designed with the tapir-like demeanour first depicted by Cuvier. One of the three was recently vandalised by someone who decided to wrench its head off. Credit: Loz Pycock Flickr (CC BY-SA 2.0)

The two smaller pterosaurs were destroyed during the middle decades of the 20th century, allegedly because they were used as target practice when the grounds were home to barracks. Based on photos of the originals, new versions made of glass fibre were reconstructed by John Wayne of Fredica Branks Sculpture and installed in 2002…. aaaand members of the public then broke them into fragments, and then stole the fragments. Consequently there are no small pterosaurs there as of right now. The good news is that these latest versions were relocated in 2014 and may eventually be installed again. I think the large pterosaurs are currently missing as well...

Incidentally, it was also the case that – until just a few decades ago – you could get onto the islands and walk right up to the models and even touch them and climb on them (should you wish). This is nicely illustrated by this old photo of Bernard Heuvelmans. It is, actually, still possible to get onto the islands, but you’d only do this if you were a colossal jerk.

Zoologist and author Bernard Heuvelmans in close proximity to one of the large pterosaur models. The photo dates to some time in the 1950s. Credit: Michell & Rickard 1982

The Crystal Palace models are super-accurate. For all those statements about the models being laughably anachronistic, a strong argument can be made that they’re among the most accurate reconstructions of extinct animals ever produced. How can I say this? Because they really do reflect, to remarkably detailed degree, what leading scientists of the time thought about the life appearance of the animals concerned. Furthermore, they were designed and made by a person with a world-class knowledge of animal anatomy and appearance. I’m talking of Benjamin Waterhouse Hawkins.

Hawkins is mostly synonymous today with Crystal Palace and you might get the impression that designing and building life-sized prehistoric animals is all he ever did (his ill-fated involvement in an American prehistoric park is similarly famous). In fact, he was a highly accomplished and skilled artist of modern zoological subjects long prior to being commissioned for this job, famously illustrating, for example, some of Darwin’s works on the voyage of The Beagle and those of prolific Victorian zoologist J. E. Gray. It was on the basis of his experience and expertise as goes illustrations of living fish, birds, mammals and other animals that he became a restorer of long-extinct ones – a valuable point relevant to some of my statements below.

Another very familiar image of the Crystal Palace models: a view of Hawkins' Sydenham workshop. The artist is unknown. Credit: Benjamin Waterhouse Hawkins and his New York City Paleozoic Museum Public Domain

It remains unclear exactly how much advice and assistance Hawkins received from Owen but we do know that Hawkins closely modelled the creatures on Owen’s views, examined numerous fossils firsthand, and constructed the models himself.

Yes, that’s right – he personally frikkin’ constructed (albeit with assistance) 30 ton models in his workshop. Smaller models were made in place, and all manner of different techniques and materials were employed in their construction; there was no one technique used consistently across all the pieces.

Here is some indication of what an expert Hawkins was on anatomy and how to portray it in art. He literally wrote books on it. Credit: ELIBRON CLASSICS

Compare what happened here with the way modern models of prehistoric animals are constructed and what the outcome is: those modern models typically do not reflect our collective palaeontological knowledge, they’re often created or designed by people who are not experts as goes their understanding of anatomy or the science and history of reconstructing extinct organisms, they’re very frequently given the OK by scientists who have scant interest in the life appearance of the animals concerned, and they’re typically constructed by teams of anonymous workers, not by expert designers or scientific artists. A consequence is that many (caveat: by no means all) modern depictions of prehistoric animals are redundant as soon as they appear. I contend that this stands in massive, unprecedented contrast to the Crystal Palace models. At the time of their creation and installation they were cutting edge and it would have been virtually impossible – perhaps, in fact, impossible – for a learned person, even a dedicated specialist, to find fault with them. Ok, certain of the models definitely involve more conjecture than others – Owen and Hawkins knew this and admitted it – but the results, while arguable, would have been acceptable and justifiable at the time. Science is a process; you work with the evidence you have at the time.

The Crystal Palace models accurately depict Hawkins' brilliant illustrations of animals (like this Megalosaurus), and these in turn accurately depict what Owen thought as goes their life appearance. Credit: Public Domain

Disclaimer: Owen’s so-called pachydermal view of dinosaurs replaced an earlier paradigm promoted by Gideon Mantell in which dinosaurs were assumed to be lizard-like and far longer-bodied and longer-tailed. There would therefore have been some experts who might have taken issue with the Crystal Palace dinosaurs specifically. However, this Mantellian view had mostly been abandoned (or usurped) by the 1850s. Owen’s motives and reasoning as goes his views on dinosaur life appearance may have been tied to his views on transmutation (Desmond 1979, 1984).

On the cutting-edge-ness of the models… imagine you’re a curious visitor, there at Crystal Palace shortly after its opening in 1854. The concept of dinosaurs is still new (Owen named Dinosauria in 1842), other spectacular fossil reptiles like ichthyosaurs and plesiosaurs are but three decades old, and the notion that pterosaurs (known as a group since the 1770s) might have exceeded modern giant birds in wingspan is brand-new (Martill 2010). Even Megaloceros the giant deer – while not technically a new discovery (it was named in 1799) – would likely seem new since the generic name emphasising its distinction from living deer species had only recently been published (in 1828).

I want you to imagine going to a modern theme park and seeing a full-sized, fully accurate model depicting an animal like this (as well as many others) - this is the giant hump-backed Cretaceous ornithomimosaur Deinocheirus. The thrill of attending Crystal Palace in the 1850s would have been many times greater than this because things would have been even more novel. Credit: FUNKMONK Wikimedia(CC BY-SA 3.0)

In other words, visiting the prehistoric park would be like visiting a 2016 attraction featuring the very latest in palaeontological discoveries and interpretations. You should imagine the Crystal Palace Dinosaur Court as if it were filled with scrupulously accurate models of a fully feathered, hump-backed Deinocheirus and a beautiful, fuzzed-up Yutyrannus, a group of striding and launching giant azhdarchids (long-necked and short-necked), a sleek mosasaur with a vertical tailfin, and frolicking marine reptiles like Malawania, Atopodentatus and Albertonectes. What I’m getting at is that it would all be new new new – people would be thinking “Wow, I hadn’t heard of that thing before!”, and “OMG, when did scientists discover that?!”.

It's true that several modern outdoor installations have featured memorable, charismatic, innovative renditions of prehistoric animals (this one - devoted to pterosaurs - was featured at the Summer Science Festival, London, in July 2010). Alas, none have been on the same scale as Crystal Palace. Credit: Darren Naish

The anatomy, the nuance, the experience. Something that’s not nearly stated enough is how remarkably detailed the Crystal Palace models are, and how good they are in terms of anatomical nuance. This is in keeping what with I said above about Hawkins and his skill and experience.

Remarkable and compelling anatomical detail (and note the replica cycad trunk used as a support for the animal's forelimb). Credit: Darren Naish

Let’s look at one of the Iguanodons as a case study. The level of detail here is extraordinary. Individual scales, fitting uniformly and neatly together as they would in a live animal, are arranged in transverse rows, those scales in some regions (like the top of the thigh and adjacent to the armpit) being far smaller than those elsewhere. Lines of larger scales fringe the upper surfaces of the fingers and toes, midline of the back and throat, and the edges of the jaws.

One of the Iguanodons (with Hylaeosaurus in the background) as seen from behind. Again, so much wonderfully portrayed anatomy. Credit: Darren Naish

Look at the contours of the body – how the skin flows from the top of the back towards and over the shoulder, the way the belly is convex and bulges outwards where it contacts the ground.

What we have here is a really compelling and convincing amount of real-world detail designed and constructed by people (or, a person) who knew a lot, and looked a lot, at real animals and at how their bodies work. Anatomical nuances, curves and contours, gravity and a sense of mass – it’s all here. Similar things can be said about many of the other models. I also have to mention the Teleosaurus. Look at how well done the scale anatomy is.

More remarkable scalation detail, this time on the Jurassic crocodyliform Teleosaurus. Scratching heron for scale. Credit: Darren Naish

Ichthyosaurs in x-ray. The majority of the Crystal Palace models do not depict dinosaurs, of course. There are in particular quite a few marine reptiles. Among the most familiar are the several ichthyosaurs, arranged at the edge of the channel close to the island with the dinosaurs. They depict Mary Anning’s Lower Jurassic ichthyosaurs of Dorset, in particular Ichthyosaurus and the far larger Temnodontosaurus.

Again, the story that mid-1800s science had it all wrong when it comes to the life appearance of ichthyosaurs is a very familiar one. Lacking excellent fossils that preserve traces of the body’s external outline (and thus the dorsal fin and the shark-like vertical tail fin), scientists of the time wrongly thought that Jurassic ichthyosaurs like Ichthyosaurus and Temnodontosaurus were crocodile-shaped, albeit equipped with paddles and a vertical expansion on the upper surface of the straight tail. These features are all accurately depicted – exactly how Owen and other scientists imagined them – in the Crystal Palace models. Again, they are super-accurate for the time.

Internal anatomical details depicted on the ichthyosaur Temnodontosaurus. Compare the forefin on the model at left to the anatomical details present in the hindfin shown at right (from Owen 1841). Credit: left: Darren Naish, right: Owen 1841

There are a few weird things though. Why do we see the sclerotic rings (those pineapple-ring-like structures, composed of interlocked sets of bony plates, embedded within the soft tissues of the vertebrate eyeball) set within the eye sockets, as if they were visible in the live animal, from the outside? And why is it that both the packed, mosaic-like bones that form the paddle’s skeleton and a peculiar series of plate-like elements are visible along the paddles’ leading edges?

I have to say that these features (the eyeball rings and the fin ossicles and plates) do somewhat ruin the illusion that we’re meant to be looking at live, accurately reconstructed animals here, and they somewhat contradict the assertion I made a moment ago about super-accuracy.

The Crystal Palace ichthyosaurs (like the Temnodontosaurus shown at bottom) are absolutely accurate when compared to other renditions of the time (like those of Hawkins and Louis Figuier shown at the top) Credit: Chris Sampson Flickr (CC BY 2.0)

However, what appears to be going on is that these features have deliberately been reconstructed as if depicted in x-ray. That is, we the viewers are being shown features of internal anatomy for educational reasons. This isn’t my idea: I learnt it from Crystal Palace authority Steven McCarthy. It’s supported by the fact that those plate-like structures along the leading edges of the forefins exactly match structures described in an unusually well-preserved hindfin from the Jurassic of Barrow-on-Soar, Leicestershire, described by Owen in 1841 (Owen 1841). UPDATE: this idea is likely completely wrong: Hawkins was, instead, actually thinking that the sclerotic ring and those paddle details would have been visible on the outside. See Ben Creisler’s comment below.

Why does the mosasaur lurk? Another marine reptile located in a different section of the park is less obvious: it’s the lurking head and thorax of the great sea-going Cretaceous lizard Mosasaurus, positioned such that it’s peering out of the water. Because the water level would rise and fall according to the activity of the fountains, the idea was that the mosasaur would appear and disappear in regular fashion, its unexpected appearance at the edge of the water perhaps being a frightening surprise to passers-by.

The Mosasaurus as it's meant to look (compare it to the image below). Note the block of Upper Cretaceous chalk beneath its jaw. Credit: Andrew Wilkinson Flickr (CC BY-SA 2.0)

The lurking nature of the model also reflects a pragmatic decision to avoid speculating too much about this animal’s overall appearance: when building the model, Hawkins (and other experts of the time) only really knew what the skull of Mosasaurus looked like. Sadly, a chunk of its lower jaw has recently broken off meaning that its jawline is somewhat ruined. Right now, it’s also somewhat obscured by surrounding vegetation.

We only see part of the Crystal Palace mosasaur, and mostly this is because Hawkins and others knew little of these animals at the time. The best preserved fossil they knew was the partial skull of the Maastricht Mosasaurus, shown at right in this famous 1812 illustration from Cuvier's study. Credit: Andrew Wilkinson Flickr (CC BY-SA 2.0) , Credit: right Cuvier 1812.

Finally, let’s look once again at the anatomical details. Look at how the scales have been arranged around the mosasaur’s face and the edges of the jaws. A sunken region in front of the eye is clearly meant to be an anatomical structure of some sort – I’m not entirely sure where the idea came from to depict it (there is presumably something relevant in Richard Owen’s writings on mosasaurs) but I wonder if it’s meant to be a fossa for a salt gland or something.

Yikes, here's how the mosasaur was looking as of July 2016. Mostly obscured and with a freshly broken jaw. Credit: Darren Naish

My apologies for not talking about the temnospondyls, dicynodonts or mammals at any length – that will have to wait to another time. Also worth saying is that several models were planned but never completed, including a mastodon, mammoth, Deinotherium, Glyptodon, the moa Dinornis, and a dodo.

Sorry I didn't talk about you, turtle-esque dicynodonts, maybe next time. Credit: Darren Naish

The most outstanding outreach campaign. I hope you enjoyed a view of the Crystal Palace models that is somewhat different from the typical stuff you might have heard before. I am, of course, far from the only person to have said that the models need to be appreciated within the context of the time, and celebrated both as an enormously successful outreach campaign and an outstanding good partnership between science, art and outdoor planning and design: my views have been heavily inspired by Peter Doyle, Steven McCarthy and others who have written about them.

But let me say again that it seems wrong to look at the models and simply scoff at how little was known at the time of their construction and at how much Owen, Hawkins and others got wrong. No, these models are scrupulously accurate based on the knowledge of the time, and they show things portrayed exactly how they should have been given the date of their construction. Again, this stands in massive contrast to the way things are done today, when models and other representations are known to be inaccurate even at the time of construction. And let us marvel again at a time and a collaboration whereby so much effort, so much brilliance, was invested in an outstanding project that brought knowledge of geology, palaeontology, animal diversity and anatomy to a fascinated public. Could such a thing ever happen again? That’s a good question.

Crystal Palace and some of the issues pertaining to palaeoart and historical depictions of dinosaurs have been mentioned at Tet Zoo a few times before. See...

Also, note that the upkeep and maintenance and restoration of the Crystal Palace models is carried out by a dedicated group of people who need your help and support. Check out the Friends of the Crystal Palace Dinosaurs blog and read about their work.

Refs - -

Becker, Hans. 1911. Alte und neue Rekonstruktionen ausgestorbener Tiere. Umschau 25, 1022-1026.

Cadbury, D. 2001. The Dinosaur Hunters. Fourth Estate, London.

Desmond, A. J. 1979. Designing the dinosaur: Richard Owen’s response to Robert Edmond Grant. Isis 70, 224-234.

- . 1984. Archetypes and Ancestors: Palaeontology in Victorian London 1850-1875. The University of Chicago Press, Chicago.

Doyle, P. 2008. A vision of ‘deep time’: the ‘Geological Illustrations’ of Crystal Palace Park, London. In Burek, C. V. & Prosser, C. D. (eds) The History of Geoconservation. Geological Society, London, Special Publications 300, pp. 193-201.

Hutchinson, H. N. 1892. Extinct Monsters. Chapman & Hall, London.

Martill, D. M. 2010. The early history of pterosaur discovery in Great Britain.In Moody, R. T. J., Buffetaut, E., Naish, D. & Martill, D. M. (eds) Dinosaurs and Other Extinct Saurians: A Historical Perspective. Geological Society, London, Special Publications 343, pp. 287-311.

McCarthy, S. & Gilbert, M. 1994. Crystal Palace Dinosaurs: The Story of the World’s First Prehistoric Sculptures. Crystal Palace Foundation, London.

Michell, J. & Rickard, R. J. M. 1982. Living Wonders: Mysteries and Curiosities of the Animal World. Thames and Hudson, London.

Owen, R. 1841. A description of some of the sort parts, with the integument, of the hind-fin of the Ichthyosaurus, indicating the shape of the fin as when recent. Transactions of the Geological Society of London 6, 199-201.

Darren Naish is a science writer, technical editor and palaeozoologist (affiliated with the University of Southampton, UK). He mostly works on Cretaceous dinosaurs and pterosaurs but has an avid interest in all things tetrapod. His publications can be downloaded at darrennaish.wordpress.com. He has been blogging at Tetrapod Zoology since 2006. Check out the Tet Zoo podcast at tetzoo.com!

More by Darren Naish