How to Build a Dinosaur

Dinosaurs - you've probably seen hundreds of them.

You might think you know what they look like,

but almost every dinosaur you've ever seen is a work of fiction.

LOW GROWL You turn on the television,

it almost feels that we know everything about them,

and that's not really the case.

But now, a groundbreaking new exhibition is working

with the world's leading dinosaur scientists

to revolutionise the way we see these animals.

We've found, using computer models, that a human sprinter

would probably be pretty well matched for a muscular tyrannosaurus.

Scientists are pushing the frontiers of our knowledge

in new and surprising ways.

We can say these dark stripes were not red,

-black or whatever - they were ginger.

-That's just amazing.

But we've never even found a complete skeleton

of a Tyrannosaurus rex, the most famous dinosaur.

So, how on earth have we worked out so much about animals that lived millions of years ago?

How do we get from an incomplete pile of broken bones to this.

ROARING

How do you build a dinosaur?

I'm Alice Roberts. I'm an anatomist used to working with human bodies.

It's not hard to put a human skeleton together.

You only need to look in the mirror

to get a pretty good idea of where the bones go.

But what do you do when the bones belong to animals that went extinct

millions of years ago?

We all think that we know what dinosaurs looked like.

We've seen plenty of them - pictures, in films and animations,

even in toy shops.

But given that the last of the dinosaurs died out

about 65 million years ago,

none of us has ever actually seen a living dinosaur.

So, how do we know what they looked like

and how can we be sure that we're getting it right?

Here in Crystal Palace, in south London,

you can still see the first dinosaur exhibition that was ever built

anywhere in the world.

The sculptures were unveiled in 1854.

It was the start of an obsession that we've never got over.

But it wasn't long before the science behind these reconstructions

had lost credibility.

Even by the end of the 19th century,

our ideas about dinosaurs had changed so much

that these models were already looked upon with scorn.

This megalosaurus, for instance, is shown walking on all four legs,

but we now know he would have been bipedal -

he would have stood on just his hind legs

and his forelegs would have been quite small

and lifted right up off the ground.

When the first iguanodon was discovered,

only one thumb bone was found,

so palaeontologists thought it must have been a horn.

But iguanodon didn't have a horn.

It's very easy to walk amongst these massive models

and to laugh at the 19th-century idea of what a dinosaur was like.

We now know so much more.

We've worked out a phenomenal amount about the dinosaurs.

But how have we done that?

How do you start to get close to animals

that lived hundreds of millions of years ago?

From 19th-century London, to 21st-century Los Angeles.

150 years after the first ever dinosaur exhibition,

I want to know how we can be sure that we're now getting it right.

So, I've come to LA's Museum of Natural History.

The museum is undergoing major redevelopment at the moment,

and at the centre of it all

is a multimillion dollar new dinosaur exhibit.

'Luis Chiappe is director of the museum's Dinosaur Institute

'and curator of the new exhibition.'

-Hello, Luis. Hello.

-How are you?

-I'm very well. Nice to meet you.

-Likewise.

'He'll be packing it with everything we know about dinosaurs,

'from the biggest to the smallest, with the latest science

'on how they looked, moved and interacted.

Beyond the fact that the exhibition is about dinosaurs,

what's the idea behind it?

It's really how do we know what we know about dinosaurs?

You're not just presenting facts, you're showing how you got to that knowledge?

Yes, how do we translate the evidence that we find in the field

into scientific knowledge.

-So, can I get a sneak preview?

-Sure, of course.

-Yeah?

-Yes.

'Our knowledge of dinosaurs has been transformed over recent years,

'and that means that when it opens, Luis's exhibition

'will aim to be the most scientifically accurate representation of dinosaurs ever.

'The science will be brought to life

'by a wide and varied cast of dinosaurs, but right now,

'the exhibition hall is a building site.'

We are approaching the centrepiece of the exhibit, a large platform

that will support three Tyrannosaurus rex,

what we call a growth series of Tyrannosaurus rex.

Because a complete T rex skeleton has never been found,

Luis's team will have to reconstruct the missing bones.

Then he'll have to choose poses that reflect the latest scientific thinking

on how these animals stood and moved,

and with three T rexs on a single platform,

he'll even be considering how they interacted.

All this for animals that went extinct

65 million years ago.

But dinosaurs weren't all big and scary.

We're still learning more about some of T rex's relatives,

and Luis will also be reconstructing

a tiny chicken-sized dinosaur called fruitadens.

As you come in to the other gallery, there's going to be a platform

with a very large dinosaur, a long neck, called mamenchisaurus,

and a tiny little one, the tiny fruitadens,

the smallest dinosaur in North America.

They have to build fruitadens

from little more than these fossil remains.

It's never been reconstructed before,

so working out what it looked like is a huge challenge.

And Luis's team will be doing much more than just piecing bones back together.

They'll be creating a lifelike model of the animal,

which means adding muscles and skin.

I think when most of us go to an exhibition like this,

we don't think about all of the work that's gone into it,

and an exhibition on this scale

requires hundreds of people to be working together,

from scientists, to engineers, to artists, and designers.

But absolutely none of it would be possible

without the starting point of the hard evidence,

the fossils themselves,

because if we'd never found their bones,

we wouldn't ever have known that these ancient animals ever existed.

Luis has come to the southeastern corner of Utah.

Today, this is Wild West country,

a stopoff on the way to the Grand Canyon,

and its past is equally epic.

All the rocks you can see around here are mostly of Jurassic age,

so this is prime dinosaur country.

At the time of the Jurassic, the dinosaurs were in their prime

and this was their home.

But it was a very different world.

Back then, this area was awash with streams and flood plains.

It was the perfect habitat

for the largest land animals that have ever lived -

the sauropods,

long-necked, plant-eating dinosaurs.

It's just a phenomenal place.

It's beautiful and it's filled with clues about...

the ancient life.

In a vast desert, most of us wouldn't have a hope

of finding those clues.

But if you know what you're looking for,

the hint of a different colour on the ground is all it takes.

Let me take a closer look.

You can see the bones

right here, and here, and here.

It's very difficult to see what exactly they may be.

They're very thin.

It would probably be worth coming back and cleaning this a little bit

and taking a closer look at what they may be.

Amazingly, less than 100 metres away,

there are more clues to the past.

Luis's colleague has found the remains of a sauropod.

There's a piece of rib here that's going into the ground,

about this angle,

and then there's a piece of the... a pubis, the hipbone, right here,

and it's almost complete,

save for the very back end, which is already starting to weather off.

Luis has to decide what to do with these finds.

Starting a new dig is a huge undertaking,

requiring time and money,

and he has limited resources.

We already have two very good sites with long-necked dinosaurs.

I'm reluctant to open another excavation.

Just half a mile away is one of those sites.

Luis's team began work on it a year ago.

Most of the bones are still embedded in the rock

and must be painstakingly excavated.

Luis knows from the layer of rock they're digging

that this dinosaur died 150 million years ago,

but he doesn't know what species it is,

and it's potentially a dinosaur that has never been seen before.

We are actually collecting in an area

that has not been sampled, no-one has really worked here before.

The possibility of having a new species is very, very, exciting.

A fossil dig is like a murder scene -

every piece of evidence about what happened 150 million years ago

has to be salvaged.

The layout of the entire site will be mapped and the precise location

of every bone fragment recorded, to help piece together the remains.

The more complete the skeleton, the easier it will be to identify

and the greater the likelihood that this dinosaur

will be turned into an exhibit.

We have hind limbs, we have forelimbs, we have a lot of the tail,

we have ribs, we have many parts of the skeleton,

and now we're starting to uncover the neck.

I would anticipate

that we're going to have to keep opening the quarry

to uncover many other neck vertebra and, hopefully, the skull.

Working out what species this is

won't be possible until the bones are back in LA.

But fossils are fragile and moving them is a risky business.

Ready? One, two, three, move.

It has to be 400 pounds at least, right? If not more.

Go slowly.

The team begin the precarious task of shifting a femur,

the single heaviest bone in the dinosaur's body.

Try to keep in a line, because if we go on this side,

it's just going to be really difficult.

Doug, why don't you go that way?

Because the fossil is so delicate, it's been cased in plaster and reinforced with steel bars.

'When you're handling bones that are heavy and fragile,

'that is definitely not an easy process.'

Down.

'If, you know, you don't have the right people, the bones can break.'

It will take many more months of work to excavate the entire skeleton

and get it back to LA for analysis.

Good, good.

But to build an exhibition,

you don't have to spend months in the desert digging up bones.

There are other places to find fossils.

There are plenty of palaeontologists working out in the field

and excavating new fossils, naming new species every year,

but there are also scientists

who are combing through existing collections in dusty store rooms,

hoping to make new discoveries

from bones that were found decades, if not centuries ago.

I've come to the Natural History Museum in Oxford,

and I'm here to meet Darren Naish.

He's a palaeontologist who looks for new dinosaurs

in the back rooms of museums.

There are always a huge number of specimens behind the scenes...

..either because they're incomplete, unglamorous, or unidentified.

Darren, I do love these museum collections, when you come behind the scenes

and you suddenly feel that you're surrounded by treasures.

It's amazing to think that there are new discoveries to be made in here as well.

In a way there are almost too many specimens for the number of experts out there.

There's new stuff to find in collections.

You don't have to go out in the field.

You can rummage through museum drawers. You WILL find something new.

Recently, Darren and a colleague did exactly that.

They came across a bone

that had been lying on a museum shelf since Victorian times.

It may look unremarkable, but with several unique features,

it didn't fit with anything that had been found before,

and it was enough for them to describe a new species.

It must have been really exciting to name a whole new species of dinosaur.

Yeah. We realised straight away that, wow,

this is something completely new.

Naming a new species, not such a big deal.

-It's quite easy to do.

-Really?

-But finding...

-There are...

-For me, you'd think that would be a kind of once in a lifetime,

wow, I've named a new species of dinosaur, but no?

No. There's huge swathes of the tree of life,

there's very little work been done.

It's quite easy to find new species.

We're in a golden age of dinosaur discovery.

-There's about 50 new species of dinosaurs named every year.

-Really?

About 90% of all named dinosaurs have been named since about 1990.

If you were to generate a discovery curve of dinosaurs over time,

you'd have a curve that's shaped like this,

and we're currently on the steep upward curve of the graph.

Why do you think there's such a craze for naming new dinosaurs at the moment?

Regions of the world are being explored more that haven't been really looked at much beforehand.

So, places like southern South America, much of central Asia,

parts of Africa and Australia,

more people are going out to those places,

finding new dinosaurs and bringing them back.

'And the more we find,

'the more complete our understanding of the world of the dinosaurs becomes.'

It makes you realise just what a vast body of knowledge

we've now amassed about these extinct animals,

so that a palaeontologist can come along,

look at a single bone and say, "This must be a whole new species".

And it also makes you wonder how many other dusty, unloved specimens

are sitting there on store shelves, just waiting to be recognised.

'Back in Los Angeles, Luis's team are working on the bones that were dug up in Utah.

'The next step in turning them into an exhibition is to work out exactly what they are.'

Well, this is where the fossil bones end up,

and here the preparators continue the process of excavation,

this time using delicate tools

and cleaning away the last of the hard sediment,

revealing the bone itself.

It's here in the dino lab

that the dinosaurs really start to come back to life.

So, Luis, is this one of the specimens from Utah?

Yes, it is.

It looks like it's taking ages

-to extract this from the stony matrix that has built up around it.

-Yeah, yeah.

Well, Erica's been working on this bone for several weeks.

It will definitely take years

for the entire skeleton to be prepared, to be cleaned up.

Do you have an idea at the moment

what species of dinosaur this might be from?

Not entirely, in terms of the species.

But we know it's a camarasaurid.

'Camarasaurids were a family of long-necked dinosaurs.

'We currently know of four different species of them,

'but Luis is hopeful that he might have found a fifth.'

So, what features will you be looking at,

as the bones are cleaned up, to help you refine your identification?

Well, you'll be looking at the shape of the centrum here,

the configuration of the different processes, the struts,

the spines of the vertebra that are, in general,

very diagnostic, they're very telling.

You must have to be an amazing anatomist, and you must have to know

the anatomy of so many different dinosaurs,

to be able to work out what it is you're looking at?

Yes, but sometimes it's difficult.

For example, here...

we have two bones of one dinosaur.

Can you figure out what they are?

Well, mm... I'm a human anatomist

so this is stretching my expertise somewhat,

asking me to identify dinosaur bones!

-And these ones are not very well preserved, I'm sorry.

-Brilliant!

You know, they're fairly kind of flat pieces of bone,

so I would think maybe this is part of the skull or the jaw.

-Am I anywhere near?

-Yes, you're absolutely right.

So, what you have here are two lower jaws of a duck-billed dinosaur.

So, they'd come together in the mid-line somehow?

Yes, they actually come together right here.

-SHE CHUCKLES

-The other way round!

Brilliant.

I know how difficult it can be to piece together an ancient skeleton from fragments,

but I've only ever worked with one species, humans,

so I'm really impressed by palaeontologists,

who have to understand the anatomy of hundreds of different dinosaur species.

Identifying a dinosaur is just the starting point

for unlocking its secrets and getting it ready for display.

It will be years before this dinosaur is ready for the public.

Instead, the centrepiece of Luis's exhibition

will be three T rex skeletons that have already been excavated

and are now ready to be mounted.

They're being put together in a workshop in New Jersey.

Resurrecting these awe-inspiring creatures will require mounting the bones in a way

that reflects the latest scientific understanding

about posture, movement and behaviour.

But the fossil remains of each of these animals are desperately incomplete.

Paul Zawisha is in charge of turning the partial, distorted skeletons,

into the most up-to-date reflection of scientific knowledge.

Okey-doke...

We've got another several weeks and I'm trying to figure out where everyone's at. Tommy?

Right now we're about...

50 to 60% finished.

Everything is articulated. We have to get the new bases built.

Did you get those hands straightened out?

-Yeah, I think we finally figured it out!

-That's good!

Two days later.

Working closely with Luis, Paul and his team will turn

a miniature model of the three T rexs into a finished exhibit.

The science will come alive through a combination of art and engineering.

Luis came out here several months ago,

he pretty much shifted things around to the scenario

that's going on here.

But again, we have a little liberty, because we want to make

these things come to life, otherwise they just...

they just don't move and they don't look real.

Fossilised bones are essentially solid lumps of rock,

which means that mounting them into a skeleton is an enormous challenge.

Most of the bones are real, which makes them extremely heavy.

We're estimating the total weight of the bones

is a little over a ton.

The femur's probably...

a good 200, 250 pounds apiece,

and we have to set those in place with special rigging devices.

Heaven forbid one of them falls

because it would take quite a bit of time to get those back together.

The entire skeleton will be held together using a custom-made steel frame,

which needs to be strong enough

to support the enormous weight of the fossils.

This will fit in, this will get attached,

to this other section over here,

and I'll take one of these ribs here,

and I'm not sure exactly which one goes where at this point.

This is number five, so it would...

lay down right in there.

That will actually get screwed in at the bottom

and just settle itself right... right in here.

Like many T rexs, this one has been given a nickname - Thomas.

He's one of the best T rex specimens ever discovered

but is still only 70% complete.

The missing bones have all been made by Paul's team,

based on over 30 partial Tyrannosaurus rexs

that have been found so far.

This particular rib, you could see where the real rib

goes together with the artificial rib,

and this is a section that we had modelled

and you can see how it blends in with the real rib,

how it's glued, and it's also pinned on the inside

so it doesn't break.

And these ribs will break like icicles.

If you pick them up the wrong way, they'll just crack,

break right apart.

It's not just about hanging the skeletons safely.

The steel frame will be a work of art in itself, millimetre perfect

and subtle enough not to draw attention away from the dinosaur.

Hon Chin is filing down part of the rib armature.

Again, this is specifically made, like a piece of jewellery.

It has to hold a specific piece in a special way.

He's at the point where he's starting to clean up the welds

and it's going to be gorgeous by the time he's finished, so...!

The pose in which the dinosaur is hung, while being true to science,

will also involve a degree of artistic interpretation,

to really bring the exhibit to life.

-A little bit more of a sine wave in it.

-OK.

-It's a little too flat

and it's not moving well, so...

Myself and Kevin have been working on the tail

and I don't like the way it looks,

and now we're going to be taking that down next week

and putting a slight bend in that, to give it a bit more life.

But it's just a visual movement. For instance,

we might change the toes just a little bit

to give this thing a sneaking feeling, or a pausing feeling.

But it's very, very, very subtle.

You might move one toe just one inch, in one direction,

and that changes how you visualise this whole thing.

But putting dinosaurs back together

is about more than just reconstructing skeletons.

We need to work out how they stood, how they moved,

and even understand the details of their physiology,

and that's not something that's easy to get right.

For example, we used to think that T rex held its head high,

with its tail dragging along the ground.

We saw it as a cold-blooded, lizard-like creature.

It wasn't until recently that T rex became a forward-thrusting aggressor,

so fast, it could apparently outrun a car.

So, how did a T rex stand, and was it really that quick?

LOW GROWL

Palaeontologists now have access to an incredible set of clues

that can help us understand the posture and movement of dinosaurs.

It's a set of clues that can tell us what they might have looked like in the flesh,

a set of clues that can even shed light

on how quickly they might have run,

and a set of clues that we all see every day.

Birds are the living descendents of a dinosaur

because dinosaurs have living descendants.

Dinosaurs are not extinct

they did not become extinct at the end of the Mesozoic era.

'It's an incredible idea

'but most experts now believe that today's birds

'are the direct descendents of ancient dinosaurs.'

-So, does that mean birds actually ARE dinosaurs?

-Yes, absolutely.

How can you be sure about that?

You have evidence from the skeletal anatomy,

you have evidence from the shape of the eggs

and the microstructure of the eggshell,

a discovery of a wealth of feathered dinosaurs,

animals that are unquestionably dinosaurs

and yet have feathers that look just like the feathers of modern birds.

'It's a discovery that revolutionises the way we see dinosaurs.'

Even some tyrannosaurs were feathered,

but the relationship between birds and dinosaurs

can tell us much more than simply what they may have looked like.

So, does this mean that we can use living birds

to help us understand dinosaurs?

Absolutely.

You know, you have 10,000 living species of birds

that are providing you an enormous amount of information

that you can use to understand the biology of the ancient dinosaurs.

It's quite amazing, but it also makes a certain degree of sense

when you really look at them.

If we want to learn about how the ancient dinosaurs moved,

and even how quickly they ran,

few animals can tell us more than ostriches.

They evolved on an early branch of the avian family tree,

and like the dinosaurs they're related to,

they're large, bipedal and flightless.

-We have some living dinosaurs here to take a look at.

-Yeah, a whole field of them.

Hello, ladies.

-They're all ladies, are they?

-Yes.

Yes, they're a bit more manageable when they're females.

'Dr John Hutchinson is based at the Royal Veterinary College just outside London.

'He's one of the world's leading experts on dinosaur movement

'and Luis has been consulting him

'to make sure his T rexs reflect the latest theories.'

-Can I touch them?

-Yeah, yeah, yeah.

-Yeah? Will they peck me?

-They'll peck at your rings.

-Will they?

-Don't peck at my rings.

-They'll try to take them off.

-But they're not very strong at pecking.

-No.

SHE LAUGHS

I want to feel your feathers.

Now, this might be what a dinosaur felt like to touch.

-That's really soft and lovely.

-Yeah, just like a cuddly toy.

Aww!

-I'm stroking dinosaurs.

-Dinosaur.

-Yeah.

Get off me.

They do look like dinosaurs, especially when you know some dinosaurs were feathered.

They do, and those feathers are quite primitive in their structure,

a lot like some of the fossil feathers we find.

'The similarities aren't just on the surface.

'We can get a much better understanding of ancient dinosaurs

'by looking at the anatomy of their modern relatives in depth.

'And a local farm has recently had to put down one of its ostriches.

'As an anatomist, I'm very used to dissecting cadavers.'

Now, I don't usually wear Wellington boots when I'm dissecting, I have to say.

'But this will be the first time that I've ever dissected a bird

'or, for that matter, the descendant of a dinosaur.'

So, John,

talk me through the anatomy that we can see on the surface.

That's our heel, the ankle joint,

but birds walk with that clear of the ground,

just like their dinosaurian ancestors did.

And really just two toes, and one main one.

The middle toe is their dominant toe, just like in a dinosaur,

the third toe is the major toe of the foot.

'And there are other similarities to their ancient relatives.'

I don't know if you can see this, but here's the tip of the wing right here

-and there's a...

-Oh, there's a claw.

-..lovely little claw coming off it.

-Yeah.

-So...

-That's at the end of one of the digits

-on their arms, on their wings?

-Yep.

-And it's just there as a relic of their ancestors.

-Mm, yeah.

'The real clues about dinosaurs

'come from seeing what the relationship is

'between a bird's muscles and it's bones.'

Right away we can see some of the thigh muscles here.

-You can see this lovely...

-Yeah, I can see these.

-..red colour,

beautiful beefy muscle.

So, based on dissections like this, how accurately do you think

you can reconstruct the musculature of extinct dinosaurs?

You can look at any bone and tell something about the soft tissue anatomy of the animal,

from the scars,

the muscle scars and ligament and tendon scars on the bones,

that are attachment points for all these things

that we see here as soft tissue.

Actually, if I bring a bone over, we can superimpose these two.

It's got one big muscle attachment right here,

-and dinosaurs have a muscle scar just like this.

-Do they?

-It appears in the first bipedal dinosaurs, this scar on the outside of the fibula...

-Yeah.

..and is not present in earlier animals.

So, this is another link between dinosaurs and birds.

-So, the next time I see a bipedal dinosaur, I must look for this lump.

-T rex will have a huge one of those.

-Yeah.

-It's just a massive scar, like this big.

-Yeah.

By estimating the muscle sizes of extinct animals

and inputting them into computer models,

John is able to get an incredible new insight

into how dinosaurs actually moved.

It's basically running a simulation.

The computer's figuring out what is the best way to use these muscles,

given what we've put in, to raise the body up.

We're not animating it, we're not saying, "Do it this way".

We're just giving it some basic rules of biology,

this is what kinds of things you should be trying to do overall,

-and then it finds the best solution.

-Yeah, yeah.

So, John, you've actually done work

trying to reconstruct how T rex would have looked,

how his muscles would have worked, how he would have run.

What kind of results have you got from that?

Yeah, we've found, using our computer models, that a human sprinter

which can do 25 miles an hour or a little faster

would probably be pretty well matched for a muscular tyrannosaurus,

or an average human who can run about 15 miles per hour

would probably be a pretty good match for a skinnier version of a T rex.

John, I've heard some theories

where T rex has been put forward as running very fast,

probably faster than that.

So, has your work basically disproved that?

Yeah, I think it's put a lot of doubt in that idea,

that T rex could run as fast as a racehorse, or even faster,

so 40 miles an hour, something like that.

I don't think you'd need an automobile to outrun a T rex.

We'd have a chance of outrunning them?

-Running away?

-Maybe, but it's never going to happen, thankfully.

The work of scientists like John has allowed us

to not only refine our ideas about these extinct animals,

but has actually transformed our image of them.

If you think about Tyrannosaurus rex as an example,

we used to think of him as standing upright like Godzilla,

but now we know that he couldn't have worked like that.

If you treat him like an engineering problem,

inform that using comparative anatomy of living animals,

and now we know that his body was much more horizontal,

with his tail held up in the air,

and our reconstructions are much more robust.

We're getting as close as we possibly can

to what this long-dead animal would have looked like.

But even working out exactly what an adult T rex would have looked like

only gives you a snapshot of a moment in time.

To really understand this animal,

we need to know how it changed over the course of its entire life,

and that's why Luis's team

are attempting the first ever reconstruction of a baby T rex.

There are some small, very tiny segments of the baby,

but some of them are so small that we can't match anything up.

Nothing like this has ever been found before.

It's much harder to recreate a baby than an adult.

Only a few tiny fragments of a skeleton have ever been found.

Paul's colleague Tommy is trying to piece together the remains

from little more than dinosaur dust.

There's not a lot of pieces and it's only for the skull.

See, I mean, I've gotten several little pieces put together.

All these bones had similar colour,

the texture on the surface was pretty close,

and a lot of times I'll look at the edge of the bone.

You'll see this one has a little white and a little black.

A lot of times it's just trying the piece, seeing if it will fit.

A lot of people find it boring.

I don't know, it calms me!

Although useful for scientists,

these fossil remains are far too limited

to bring a baby T rex to life for an audience.

And that's why the entire baby skeleton will be a model,

its bones made not from fossils but from foam and resin.

This is where the artists come in.

They will produce creatures from their imaginations,

but they have to be guided by the science

which provides them with a range of possibilities.

Ultimately, the animal that they draw or sculpt

will be a blend of science and art.

The baby T rex will be sculpted by Doyle, one of Luis's artists.

When you're doing something that's brand new, that there is no precedent for,

it can be a little nerve-racking and it can be a lot of fun.

For my baby T rex, there's no reference for that,

so there's a lot of interpretation there.

'With his miniature model of an adult T rex for reference,

'along with the growing patterns of close relatives of tyrannosaurs,

'it's possible to work out the likely proportions of the baby.

'The starting point for the sculpture is a simple illustration.'

So, I'm going to start off.

T rex, usually, an adult skull is a great way to measure,

because it's so big.

But in babies,

the skull is going to be thinner,

and the rule is always that the orbit is going to be larger.

And also when you look at human babies,

I've noticed that they are about three heads tall,

versus an adult human, which is anywhere from seven to nine,

depending on how tall they are.

Do you find yourself at all looking at other people's reconstructions

and thinking, "They've got that wrong"?

Ah... Yes. SHE CHUCKLES

There are a bunch of people who are out there

who are coming from maybe film or special effects

or something like that.

They're doing this kind of work from a less informed background.

So, I'm very privileged to work with a scientist,

and that's definitely an asset that I don't dare forget.

-He's looking nice, this T rex, this little two-year-old.

-Yeah.

'But with limited fossil remains,

'the reconstruction has room for creative licence.'

So, can you draw me another baby T rex...

-Sure.

-..based on the same evidence,

but taking it off in a different direction?

Let's do the same thing.

We have our head.

There's a lot of evidence

that some of them had feathers,

and that maybe some of them,

when they were young, would have had some sort of downy covering

that would have left in adulthood,

so that it would have been shedded before they were fully grown.

This little baby's looking extraordinarily bird-like

-and has really long legs.

-Yeah.

Is this a reasonable interpretation?

-There's nothing that says that it can't be this way.

-Right.

Fantastic. It's the same creature, but they're very different.

The length of the legs is quite extraordinary in this one.

And I love the feathers.

That immediately makes it look like

a completely different creature.

It shows you there's quite a bit of room for artistic manoeuvre

-in these reconstructions.

-Yes. Definitely, definitely.

'The questions about Luis's baby T rex

'run even deeper than its appearance.

'With such limited fossils, some scientists have actually questioned whether the bones

'might belong to a different species of dinosaur entirely -

'something like a T rex, but much smaller.'

You're presenting a mounted skeleton of this baby T rex,

and this is the first baby T rex

that's been found and has been put on display.

How can you be sure that it is indeed a T rex, if it's a baby,

because bones change as juveniles turn into adults.

You can read the characteristics of the bone tissue

and that can tell you if the animal is a full-grown individual

or if it's a baby or a very young individual.

So, we know that perhaps in future, discoveries may prove

that there was another species of tyrannosaur

that essentially lived together with T rex

and that maybe this is a baby of that particular species.

But at the moment, with information that we have,

it seems that the most reasonable hypothesis is to say

that this one represents a baby of a Tyrannosaurus rex.

I think that's quite brave

to put something like a baby T rex in this exhibition as a mounted skeleton,

because there's nothing to compare it with.

It is our responsibility to make sure that people understand

that things are not written in stone

and our scientific conclusions change as we gather more evidence.

Back in New Jersey,

the T rexs are nearly complete, and Luis has come to inspect them.

This is phenomenal, you know.

-You like it?

-It looks awesome.

It's just fantastic. Really fantastic.

Everything you thought it would be?

-Better, better, better.

-Good.

It's hard to describe

but I feel that it's very dynamic, you know.

Well, we brought the right-hand foot over the centre line quite a bit...

-Yeah, I can see that.

-..with a turning, and...

I can see that.

-It gives a little... quite a bit of movement.

-Yeah.

I'm glad that you like it, Luis.

I think it's phenomenal.

But it's not completely finished.

Paul and his team need Luis's advice on a couple of issues.

There are several unknowns, and a complete tail has never been found.

So, on the older drawings that we have, there's maybe 53 tail vertebra.

The newer thinking is, there's close to 43.

Palaeontology, mostly it's a soft science,

so theories change with new evidence that is found.

One of the big questions about T rex

is what it's surprisingly short arms were used for.

They might have been used to hold on to prey,

or to push the body up from a sitting position - no-one knows.

And that's partly because each arm is anchored to the body

by the shoulder blade or scapula,

and there's no easy way of telling exactly where that sat.

With the scapula I've seen

they've gone up closer to the vertebra on the backbone.

I've also seen where they're lowered almost to where the belly is.

There's parts of the front end of the scapula, the coracoids.

Some people think they go together this much, some think this much.

But that all has to do with how everything hangs on the front end of this,

and also how the hands were used.

Those arms are just about the same size as a human arm.

The difficulty in placing the scapula on Thomas

is compounded by the fact that the bones were distorted

over the millions of years that they spent buried underground.

They're flattened

and they don't really have the curvature that they may have had

when the animal was alive, before.

It's really difficult to fit them on the sides of the ribcage.

I guess that that's the nature of the beast.

We're going to have to find a compromise and we'll live with it.

Back in LA, there are two months to go before the exhibition opens.

The three T rexs are now installed.

Oh, this is a bit different. There are dinosaurs here.

-Now, these guys I recognise.

-Yes.

So, this is your famous Thomas.

-Can we get up here?

-Sure.

-Yeah?

-You can... Absolutely. Feel free.

Face to face with a baby T rex.

'With three T rexs of different ages on one platform,

'it's possible for the first time ever to get an understanding

'of the entire life cycle of this legend of the dinosaur kingdom.'

-Having a series of juvenile skeletons gives you insights into the way dinosaurs grew?

-Absolutely.

The dinosaurs had growth spurts, so this animal

-is estimated to have died at the age of two.

-Right.

And this one here is estimated to have died at the age of 13.

There's, you know, there's a size discrepancy here,

but they're also 11 years apart.

-Mm.

-Yet this animal is only four years...

-Yeah.

-..older than this one, yet is enormously bigger than this one.

What this is telling you is that between 13 and 17

they were able to add about 1,500 pounds -

that's, what, 750 kilograms a year.

Wow. And when you see the two skeletons close to each other like that,

you really get a kind of physical impression of that.

'Although Thomas towers over the younger T rexs,

'even he wasn't fully grown.

'But at about 17 years old

'he was already 11 metres long and over three tonnes in weight.'

-So, this is a juvenile? This enormous skeleton?

-Indeed, indeed.

-This is an animal that probably died at the age of 17.

-Right, OK.

-So, rather young.

-So, still a teenager?

-And you can tell that it's a juvenile,

not only based on the histology on the bone tissue,

for which we have studies of it,

but also because there are many bones that would fuse

when the animal was a full-grown...

-Yeah.

-..that have not yet been fused.

One of them is here, the calcaneum and the astragalus are completely unfused,

and both with the tibia.

'And it's not just the phenomenal speed at which they grew

'that Luis is shedding light on.

'The final addition to this platform

'will be the carcass of another dinosaur - the T rex's dinner.

'It will give us an insight

'into how the three T rexs may have interacted.'

So, how realistic do you think it is

to show three tyrannosaurs coming together like this?

We have evidence suggesting that these animals lived in groups.

It's very reasonable to imagine a scene like this, in which you have a juvenile

eating a carcass of a duck-billed dinosaur,

and other individuals coming and being attracted by the carcass.

If there's going to be a skeleton here representing an edmontosaurus,

a duck-billed dinosaur, being eaten by the T rexs,

is there actually evidence that they ate this type of dinosaur?

You have evidence in the shape of bones of duck-bills,

like edmontosaurus,

that have tooth marks, essentially,

and those marks, those scratches on the bone, coincide well

with the shape of the crowns of the teeth of Tyrannosaurus rex.

That's quite forensic.

-So, you've actually got gnaw marks on a duck-billed dinosaur.

-Yes.

-Fantastic.

But the exhibition isn't only about T rex.

In amongst the 20 major mounts will be fruitadens,

the smallest dinosaur ever to be found in North America.

Working from his own illustration, Doyle has created five fruitadens.

It's the first time that this dinosaur has ever been reconstructed.

This is full-grown, to scale.

It's a very small dinosaur and one of the smallest in the world.

Because the specimen

is so fragile and sparse,

the information that we can gather,

a lot of it is inferred,

or we're guessing that it fits with a group of animals,

based on what information we do have.

We don't have a full skeleton.

By comparing the size of a forelimb to a thighbone,

it was clear that fruitadens was bipedal.

And by studying close relatives, it's possible to get a good idea

of what a complete skeleton would have looked like.

The real challenge was to turn that skeleton into a fleshed-out animal.

Musculature can be inferred from the bones. You can see muscle attachments.

Every animal has some sort of muscle

that pulls the leg back

and also something that supports the leg in front,

a calf muscle, gastrocnemius,

or any sort of tendon that would go down to the feet.

That's something that exists on every animal that walks on land.

With large teeth for mashing plants

and sharper teeth for eating insects and worms,

we can even tell that fruitadens was an omnivore.

The final piece of the puzzle in recreating this animal

is its colour, and that's something we can't be sure about.

If you push things too far, you go with polka dots and purple and pink,

your audience simply won't believe it.

But if you draw upon the examples of our living animals,

we can actually gain a lot just by looking at crocodile skin

and the colouration and maybe some lizards and fish, even,

and it will remain believable.

Like everything in the exhibition, the finished work will have to be approved by Luis.

So, one thing we need to keep in mind

is that although we want to have some variation in pattern,

or in colour, they obviously all need to look the same species.

You going to give me some freedom

-to experiment with colours, maybe in the face or the throat?

-Yeah...

I still think that overall we want to stick to standard grey, green, brown.

I think that it will be nice to be subtle,

but something that can be...

can be viewed when you're looking at it from, you know, six feet away.

Although the colour of fruitadens is unknown,

new scientific breakthroughs

are allowing palaeontologists to see some dinosaurs in a way

that's never been possible before.

We're still learning more about dinosaurs

as increasing numbers of specimens come to light,

but also as the techniques that we use to analyse them

become more and more sophisticated.

And I'm off to meet somebody now who's made great discoveries

in one particular aspect of dinosaur science

that many people thought would remain hidden for ever.

Here's another one we're going to look at. I'll just put it in.

It will take a minute or two to fire up the vacuum.

'Professor Mike Benton recently came across

'the remains of a dinosaur that was so exquisitely well preserved

'that feathers, as well as bones, had fossilised.

'Incredibly, those feathers can tell us the colour of a dinosaur

'that lived 125 million years ago.'

Going back, say, ten years ago, would you ever have imagined that you would have been able

to tell what colour any dinosaurs would have been?

No. I mean, I think at that time I, and everybody else,

would have said that is one of the things we'll never know.

And so if we just focus up, see what we've got here.

'Using a scanning electron microscope, Mike can find clues

'about the pigmentation of these ancient fossil feathers.'

If we have a look at this...

-We're at quite high magnification - that's 9,000 times.

-Right.

All of these sausage shapes, then, are melanosomes,

and in a living feather they would be full of the chemical melanin,

which, in fact, gives the colour.

And these sausage-shaped ones

are a sure indicator of a particular kind of melanin,

which is the one that gives a black or dark brown colour.

So, in some cases like this,

the field of view is completely packed with the sausage-shaped ones,

so we know this must have been intensely black.

If they were more loosely spaced,

we'd know it was a paler colour, maybe dark brown, or grey.

Right. So, is it just really the presence or absence

of the black pigments that you're able to ascertain?

The wonderful thing is that there's another form of melanin that gives a ginger colour.

And it is packaged in a different shape of melanosome,

not this kind of cigar-shaped, or sausage-shaped one,

but a spherical one, a little ball.

Close it up, get the vacuum going.

'A sample from a different fossil shows what the structures that carry this ginger pigment look like.'

Oh, that's entirely different.

This surface looks as though you've taken a melon baller

and scooped out lots of little spherical hollows.

So, what colour would these melanosomes have made?

This is definitely ginger.

If you look at a ginger hair from a mammal or a human being, that's what you'd see also.

So, is it relatively easy to compare your dinosaur feathers

with what's already known about the feathers of living birds,

to get that comparison, to know what colours you were looking at here?

We can put the specimens in one after the other.

There's the modern one, there's the fossil.

Spot the difference. No difference at all.

And who on earth would have thought a dinosaur is close to a bird?

But there we are, it's kind of proved in the skeletons

and now, if you like, proved in the anatomy of the feathers.

'For those few dinosaurs from whom fossilised feathers have been found,

'largely in China,

'we can now put the finishing touches to a reconstruction.'

Has this changed the way

that artists are painting their reconstructions, then?

We've got some dinosaurs

where you've got a very good idea exactly what they look like.

Yes, it is changing the way people view them.

If we have a look at these paintings of sinosauropteryx,

which is one of the lovely little dinosaurs,

this was probably done five or six years ago.

It looks a bit odd. They've got the texture of the feathers

and that's more or less what we would believe from the fossil,

but they've made it a strange sea green kind of colour.

A few years later, the same artists are able to produce

a picture like this, which shows the same dinosaur,

but with a very definite

ginger, white, ginger, white sort of barber's pole stripe on the tail.

-So, this is based on your analysis of colour in this particular dinosaur?

-Yes. Yes.

Of this particular dinosaur we took samples from the dark stripes

and we can say these dark stripes

were not red or black or whatever - they were ginger.

Right. That's just amazing.

So, this is more than just being able to put a bit of colour

on your illustrations -

it's actually telling you something quite important about dinosaurs?

Yes. It may say something about behaviour, which we wouldn't have thought we could ever get to.

If they are coloured, and if they are striped and patterned,

-there must be some visual purpose, signalling of some kind.

-Yeah, yeah.

Camouflage, or sexual display, or a warning thing -

you know, "I've got a flash of colour, don't mess with me", you know.

So, there's all sorts of reasons they may have had those colours.

These new discoveries really do bring dinosaurs

right out of the realm of the mythical and the fantastical.

They're not imagined creatures at all, they are real.

And with some of them, when we have all this information,

we can look at a reconstruction and know that

that is a lifelike representation of that animal,

from the size and shape of its body

to the way it holds itself, the way it moves,

down to its colour.

All of that is rooted in science.

Back in Los Angeles, last-minute preparations are under way to get the dinosaurs ready for the public.

It's only now that you get a sense of just how many people have been involved

in creating this exhibition, from the artists, to the designers,

to the teams that made the interactive media and the mounts for the dinosaurs,

all of it bringing to life the decades of research our current scientific understanding relies on.

CHEERING

The exhibition consists of over 300 specimens.

It's taken more than six years to complete

and cost tens of millions of dollars.

We've created an exhibit

that this part of the world has never seen.

And it's very rewarding for me to think about the millions of kids

and the millions of people that during the next 20 years

will visit this exhibit

and will remember this exhibit for the rest of their lives.

These animals look like something out of a comic book,

or a Hollywood studio, but they were real.

From a pile of dusty bones millions of years old,

we can put a skeleton back together,

flesh it out, tell what colour these creatures were,

and even say something about how they grew up.

I think this is a unique time to be a dinosaur palaeontologist.

We're finding so much, discovering new dinosaurs

and learning new things about them.

There are certainly still gaps in our knowledge

but I find it amazing just how much we do know

about these extinct animals that no-one has ever actually seen alive

and that lived so many millions of years ago.

The creatures themselves are utterly awe-inspiring

but I think so is the incredible amount of work

and the vast numbers of people involved in reconstructing them

so that we can come face to face with a dinosaur.

Subtitles by Red Bee Media Ltd

E-mail [email protected]