More than 120 million years ago, in a hot, conifer-filled forest in what's now India, a small insect made a terrible mistake.

While searching for a tasty meal of pine pollen it wandered one step too far, only to find itself trapped in sticky, yellow resin.

Tired from its flight, this small weevil was quickly entombed in the fragrant yellow material, which eventually became the substance we know today as Amber.

Then, in 1993, scientists cracked open this very same piece of amber.

They took the body of the weevil, and they sampled its DNA.

Now, this is not a scene from the Jurassic Park franchise.

But this research IS from the 1990s, a decade when scientists were rushing to find the most ancient DNA.

And at the time, this weevil was the oldest thing ever to have its DNA sampled.

Or, at least, so we thought.

The fact is, we can indeed get the DNA of extinct organisms from some fossils.

It's fragmented, and it's imperfect, but it's possible.

It's just not possible for every type of fossil, and, most importantly, not from every time period.

It took another few decades of research, and a lot of take-backs, before scientists could figure out how we could truly unlock the genetic secrets of the past.

The first piece of ancient DNA ever replicated was of an animal called the Quagga, a subspecies of zebra that went extinct in the 19th century.

It was sampled in 1984, pretty much just to see if ancient DNA could be sampled at all.

But that research turned out to be extremely useful, and not just because it inspired Michael Crichton's famous novel.

The researchers used the size of differences in the DNA sequence to determine when the Quagga, which is now known to be a subspecies of Plains Zebra, diverged from another species, the Mountain Zebra.

That split happened about 3 to 4 million years ago, it turns out.

So even though these species of Zebras look very similar, we now know that they parted ways a long time ago, before the last Ice Age began.

Now, that DNA was tested from a sample of dried muscle taken from a museum specimen.

And for the next few years, the search for ancient DNA drew from similar sources - soft tissue, preserved in things like permafrost or ice, or mummified, or trapped in amber.

And in the search for the oldest material, amber seemed like the best place to look.

After all, amber traps organisms in a perfect medium for preservation.

It dehydrates the DNA, which makes it more stable, and tree resin has antimicrobial properties, which keeps the tissues from breaking down.

So, in addition to our friend the Jurassic Weevil, paleontologists sampled termites, bees, and other insects from their amber tombs.

Not mosquitos though.

Amber containing mosquitoes has not been sampled for DNA yet.

Still, these early efforts taught us a lot about ancient DNA, and the organisms that managed to hold on to it for us.

But there was a growing suspicion among scientists that the oldest DNA to be extracted -- including the stuff from that weevil -- wasn't what we thought it was.

Experts already knew that such ancient DNA wasn't perfect or pristine.

Because, DNA is degrading all the time!

Even in living things!

Including you!

The tiny components, or base pairs, that form its code are always being changed by different processes.

The most common of these is a process called depurination.

It's caused by water molecules in your cells that attach to some of the base pairs, which makes them more likely to come off.

Water is great for your cells, but over time, it causes damage too, including to your DNA.

But usually, damage like this isn't a big deal.

Your cells have countermeasures that straighten, fix, or discard DNA that's been altered by things like depurination.

However, all those repair services go out of business ... once you die.

But the degradation continues.

Now, back in the 1990s, scientists knew all this.

It was part of why getting DNA from a Jurassic Weevil seemed like a miracle to some, and an impossibility to others.

What scientists weren't sure about was how long it took DNA to degrade to the point where it was no longer readable.

Was it 100 years?

Or 100 million years?

Today we know that DNA has a half-life, kind of like radioactive elements do.

That half-life marks the amount of time until half of the DNA in a sample is degraded beyond use.

But it can vary a lot, depending to some degree on the organism, but to even greater degree on the quality of preservation.

For example, recent research has shown that, in cores of ocean sediments, the amount of DNA from single-celled algae known as diatoms drops in half about every 15,000 years.

So that's it's half-life.

But, by contrast, one study of the bones of the extinct, large, flightless bird called the moa, showed that its DNA had a half life of just 521 years.

Now, as DNA decays, it doesn't just disappear - it breaks apart into smaller, harder-to-read fragments.

But these half-lives do mean that there's an upper limit to how long DNA sticks around.

This is where preservation comes in.

Ideal environments for DNA preservation include colder temperatures with very limited fluctuations.

Closed environments are good, too.

DNA on the inside of bones is better preserved than DNA on the outside, because there's less interaction with the environment.

But even in freezing cold temperatures with best case preservation, there's a limit.

A study done in 2012 of 158 well-dated fossils concluded that, even in the best circumstances, DNA decays well beyond readability by 6.8 million years.

That's still slow enough that readable DNA from the chloroplasts in diatoms can be found in marine sediments that are up to 1.4 million years old.

Here at Eons, we researched this a lot, and to our knowledge, that's the oldest confirmed DNA that's ever been sequenced.

Yet, anyways.

But with new genetic techniques, scientists can read increasingly smaller chunks of DNA and put them together to make longer strands - like the full genome of a 700 thousand year old horse, which was sequenced in 2013 from many, many small chunks of DNA.

And it helps that shorter chunks of DNA, like the DNA found in your mitochondria or a diatom's chloroplast, are more stable and can last longer.

So if DNA becomes unreadable in less than 6.8 million years, how the heck do we have DNA from a weevil that's 120 million years old?

Well it turns out, that "ancient weevil" DNA wasn't actually from an ancient weevil.

And the problem was in the methodology.

In order to read a DNA molecule, you need a LOT of it to make sense of what you're reading.

This means you need to make many copies of it, in a process called amplification.

The easiest and most efficient way to amplify DNA is a process called PCR, or Polymerase Chain Reaction.

PCR can quickly make even small amounts of DNA into large, consistent samples that are easy to test.

And it's really sensitive: All you need is an itty, bitty bit of DNA to start with.

But because it's so sensitive, it can also accidentally replicate things you didn't want.

Like, if a single human skin cell should fall into the sample, it could be replicated so quickly and thoroughly that its genetic code would overwhelm the sample.

And that's exactly what happened with the sample from the weevil.

In the late 90s and 2000s, when lab conditions became better controlled, samples that were tested in the early '90s were re-tested.

And a lot them couldn't be reproduced successfully.

The DNA that we thought was from that Jurassic weevil actually turned out to be mostly from a modern fungus that had gotten into the sample.

And the rest of the DNA was from a modern weevil, probably because the scientists were comparing the old DNA to DNA from living species, and accidentally cross-contaminated.

Likewise, the termites and the bees preserved in Amber were all re-tested... and their DNA was found to be from humans, trees, fungi and other modern contaminants.

And when you're dealing with tiny snippets of DNA, it's actually not that hard to mistake one organism for another.

After all, we all share a lot of our DNA with other organisms, even ones that bear no resemblance to us.

So if these scientists happened to pick the wrong section of DNA to replicate, they could end up reproducing a section that's in a weevil, but is also in a tree, or a human.

So... does that mean there isn't DNA from fossils after all?

Nope!

We can get great DNA samples from some fossils, as long as they're more recent, and most importantly - if you're really careful about preventing contamination.

Nowadays, you have to wear a bodysuit and two pairs of latex gloves to keep your DNA from falling into the mix.

Labs have to be sealed off from outside air, and surfaces must be bathed frequently in UV light to kill any lingering genetic material.

And if you're comparing ancient DNA to modern DNA, you have to use two separate labs so they doesn't get mixed up.

But all these precautions are worth it, because when it's amplified properly, ancient DNA can reveal to us some wonderful things!

For example, DNA from fossil humans has shown us a lot about where different human populations came from.

It's demonstrated that humans, Neanderthals, and Denisovans were all probably interbreeding during the last 100 to 200 thousand years.

And in 2014, ancient DNA also showed us that the extinct flightless Elephant bird from Madagascar was most closely related to the Kiwi of New Zealand, and not Ostriches, like we once thought.

So even though it doesn't reach back to the days of the non-avian dinosaurs, some DNA that we've sequenced is still pretty darned old - like that 700,000 year old horse from the Yukon Territory.

In 2013, it helped to illuminate the story of horse evolution, and showed that bone DNA is better preserved in permafrost than we previously thought, possibly storing readable pieces for up to a million years.

And recent research has changed what we know about DNA decay rates, too.

In 2016, scientists studying diatom DNA found that even though it decays rapidly for the first hundred thousand years, the older stuff decays more slowly, and no longer follows the regular half-life pattern.

Likewise, an analysis in 2017 found that older bones of large mammals held more DNA than expected, given the half-life of DNA.

And other research has even shown that certain types of bone, like the dense bones of your inner ear, hold more DNA and more likely to preserve DNA for longer.

So, all of this suggests that bigger chunks of DNA could last longer than we thought.

We've learned a lot about the limitations we currently face when it comes to studying the DNA of long-gone organisms.

And all of this adds up to the knowledge that extracting DNA from a 75 million year old velociraptor is impossible.

At least, for now.

Remember: 25 years ago, it seemed impossible we'd ever have ancient DNA at all.

And right now, truly ancient DNA - dinosaur DNA - is out there, but it's dissolved into pieces.

It's impossible to read using our current technology.

It'd be like trying to piece together an entire book that's been chopped up into individual words... or letters.

But that doesn't mean it's always going to be impossible.

It just means that, at the moment, we can't make out what that book says.

Maybe one day, perhaps even in our lifetimes, we'll find a way to crack that code.

But for now, the mysteries of that weevil's genetic code remain a jigsaw puzzle of base pairs that we have yet to put together.