If you could build a time machine and go back to Earth's distant past, you'd get a nasty surprise. You wouldn't be able to breathe the air. Unless you had some breathing apparatus, you would asphyxiate within minutes.
For the first half of our planet's history, there was no oxygen in the atmosphere. This life-giving gas only started to appear about 2.4 billion years ago.
This "Great Oxidation Event" was one of the most important things to ever happen on this planet. Without it, there could never have been any animals that breathe oxygen: no insects, no fish, and certainly no humans.
For decades, scientists have worked to understand how and why the first oxygen was pumped into the air. They have long suspected that life itself was responsible for creating the air that we breathe.
But not just any life. If the latest findings are to be believed, life itself was undergoing a tremendous transformation just before the Great Oxidation Event. This evolutionary leap forward may be the key to understanding what happened.
Earth was already 2 billion years old at the time of the Great Oxidation Event, having formed 4.5 billion years ago. It was inhabited, but only by single-celled organisms.
They evolved a way to take energy from sunlight
It's not clear exactly when life began, but the oldest known fossils of these microorganisms date back 3.5 billion years, so it must have been before that. That means life had been around for at least a billion years before the Great Oxidation Event.
Those simple life-forms are the prime suspects for the Great Oxidation Event. One group in particular stands out: cyanobacteria. Today, these microscopic organisms sometimes form bright blue-green layers on ponds and oceans.
Their ancestors invented a trick that has since spread like wildlife. They evolved a way to take energy from sunlight, and use it to make sugars out of water and carbon dioxide.
This is called photosynthesis, and today it's how all green plants get their food. That tree down your street is pretty much using the same chemical process that the first cyanobacteria used billions of years ago.
It was the cyanobacteria, pumping out unwanted oxygen, that transformed Earth's atmosphere
From the bacteria's point of view, photosynthesis has one irritating downside. It produces oxygen as a waste product. Oxygen is of no use to them, so they release it into the air.
So there's a simple explanation for the Great Oxidation Event. It was the cyanobacteria, pumping out unwanted oxygen, that transformed Earth's atmosphere.
But while this explains how it happened, it doesn't explain why, and it certainly doesn't explain when it happened.
The problem is that cyanobacteria seem to have been around long before the Great Oxidation Event. "They're probably among the first organisms we have on this planet," says Bettina Schirrmeister of the University of Bristol in the UK.
Maybe the cyanobacteria changed
We can be confident that there were cyanobacteria by 2.9 billion years ago, because there is evidence of isolated "oxygen oases" at that time. They might date as far back as 3.5 billion years, but it's hard to tell because the fossil record is so patchy.
That means the cyanobacteria were busy pumping out oxygen for at least half a billion years before oxygen started appearing in the air. That doesn't make a lot of sense.
One explanation is that there were a lot of chemicals around – perhaps volcanic gases – that reacted with the oxygen, effectively "mopping it up".
But there's another possibility, says Schirrmeister. Maybe the cyanobacteria changed. "Some evolutionary innovation in cyanobacteria helped them to become more successful and more important," she says.
Some modern cyanobacteria have done something that, by bacterial standards, is remarkable. While the vast majority of bacteria are single cells, they are multicellular.
Multicellularity could have been a game-changer for Earth's early cyanobacteria
The individual cyanobacterial cells have joined up into stringy filaments, like the carriages of a train. That in itself is unusual for bacteria, but some have gone further.
"Many cyanobacteria are able to produce specialised cells that lose their ability to divide," says Schirrmeister. "This is the first form of specialisation we see." It's a simple version of the many specialised cells that animals have, such as muscle, nerve and blood cells.
Schirrmeister thinks multicellularity could have been a game-changer for Earth's early cyanobacteria. It offers several possible advantages.
On the early Earth, single-celled organisms often lived together in flat layers of gunk called "mats". Within each mat there would have been many different species of cyanobacteria, and a host of other things to boot.
The Earth was being bombarded with harmful ultraviolet radiation from the Sun
A multicellular cyanobacterium would have one clear advantage compared to its single-celled rivals. It would find it easier to spread, because its larger surface area would mean it was better at attaching itself to slippery rocks. Such an organism would be "less likely to wash away in the current", says Schirrmeister.
Many modern multicellular cyanobacteria can move around within their mats. "They're not extremely fast but they can move," says Schirrmeister. That suggests the primordial ones could as well.
Moving could have helped them survive. At the time the Earth was being bombarded with harmful ultraviolet radiation from the Sun, and there was no ozone layer to keep it out.
"In modern mats, cyanobacteria will turn around and appear vertical instead of horizontal to protect themselves from excess sunlight," says Schirrmeister. "You have also movement between layers. It might be these multicellular cyanobacteria had the ability to position themselves optimally within the mat."
It's a neat idea. But for it to be true, cyanobacteria must have evolved multicellularity before the Great Oxidation Event.
Schirrmeister has spent the last few years trying to figure out when cyanobacteria first evolved multicellularity.
The clues lie in their genes. By examining genes that all cyanobacteria share, and identifying tiny differences between them, Schirrmeister could figure out how they are all related – essentially drawing up a family tree of cyanobacteria.
With that tree in place, Schirrmeister could then home in on the multicellular cyanobacteria, and estimate roughly when they first became multicellular.
Her first attempt, published in 2011, suggested that most modern cyanobacteria are descended from multicellular ancestors. That suggested multicellularity was ancient, but it was difficult to put a firm date on it.
Her family tree was only based on one gene
Schirrmeister refined her methods for a second paper, published in 2013. This suggested that multicellularity evolved not long before the Great Oxidation Event, at a time when cyanobacteria were diversifying rapidly.
But that didn't clinch the argument. Her family tree was only based on one gene, albeit a gene shared by every single species of cyanobacterium. That meant the tree was suspect.
So Schirrmeister has now gone one better.
"This time I worked with 756 genes," says Schirrmeister. "The genes I took are present in all cyanobacteria."
We have multicellularity evolving before the Great Oxidation Event
Her estimate of the origin of multicellularity is still rough, but it seems to be around 2.5 billion years ago – before the Great Oxidation Event.
There are several different ways to calculate these family trees, and they all gave the same answer. "No matter how we calibrate our phylogeny, it seems more likely we have multicellularity evolving before the Great Oxidation Event," says Schirrmeister.
The results are published in Palaeontology.
This may not be the end of the story. Even if Schirrmeister's results are confirmed, and cyanobacteria did become multicellular just before the Great Oxidation Event, there are two big questions.
It is one of the most important things to ever happen on this planet
The first is, did multicellularity really offer them the advantages she thinks it did? We don't know, but we could find out: by testing how modern single-celled and multicellular cyanobacteria cope with different situations.
The second question is harder: why did it take so long for cyanobacteria to become multicellular? If it is so advantageous, why did they not evolve it sooner, and trigger an earlier Great Oxidation Event?
"The next step is to find out which genes are responsible for multicellularity in cyanobacteria," says Schirrmeister. "Then I could say why did it take that long, why didn't it evolve earlier." If lots of new genes were required, it becomes understandable that it took the cyanobacteria a long time to evolve it.
Whatever caused the Great Oxidation Event, it's clear that it is one of the most important things to ever happen on this planet.
In the short term, it was probably rather bad news for life.
"Oxygen would have been lethal for many bacteria," says Schirrmeister. "It's hard to prove, because from the fossil record we don't have a lot of deposits from that time… [but] we can assume we had a lot of bacteria dying at that point."
Those first multicellular cyanobacteria triggered the evolution of complex life
But in the longer term, it allowed a whole new kind of life to evolve. Oxygen is a reactive gas – that's why it starts fires – so when some organisms figured out how to harness it, they suddenly had access to a major new source of energy.
By breathing oxygen, organisms could become much more active, and much larger. Moving beyond the simple multicellularity developed by cyanobacteria, some organisms became far more intricate. They became plants and animals, from sponges and worms to fish and, ultimately, humans.
If Schirrmeister is right, those first multicellular cyanobacteria triggered the evolution of complex life, including us, by producing oxygen on a global scale. "It made complex life possible," she says.
Not bad for a bunch of tiny blue-green bacteria.