91av

A FERN-LIKE impression appears in the rock face. Grasping at their rope, the boys very nearly miss it. It is April 1957 and three schoolboys are climbing in a quarry in Charnwood Forest in the English east midlands. Incredible as it would have seemed at the time, this serendipitous find was about to change our understanding of the history of life.

“Our reaction was, it’s a leaf,” says Roger Mason, who was 16 when he discovered the fossil and is now a professor at the China University of Geosciences in Wuhan. “We were surprised, but I didn’t know enough about geology to realise how startling a discovery it was.” A few days later he persuaded Trevor Ford of the nearby University of Leicester to have a look. Ford was sceptical. The Charnwood rocks were Precambrian, too old for fossils. Everyone knew that. But Ford was astonished at what he saw. The find was clearly a fossil resembling a fern frond about 20 centimetres long.

To understand the significance of finding a fossil like this in Precambrian rocks, you have to go back to the time of Charles Darwin. He was familiar with the mystery of the “Cambrian explosion” – the abrupt appearance in the fossil record of multicellular animals 542 million years ago. How could such diversity be seen in the Cambrian but not before? In 1859, Darwin wrote in On the Origin of Species that the Precambrian must have “swarmed with living creatures” that had not been preserved. The problem became known as Darwin’s dilemma and remained a puzzle for nearly a century.

The boys’ discovery set off a chain of research around the world, says Ford, who named the fossil Charnia masoni. “People started to look at Precambrian rocks again.” Australian palaeontologists soon realised that they had already found similar fossils but had misdated them as Cambrian. These finds, from the 1940s, came to be known as Ediacarans, after their location in the Ediacara Hills of South Australia. The name stuck, and as scientists re-examined similarly aged rocks, they realised that Ediacarans could be found worldwide.

After 50 years of study, we now know a lot about the Ediacarans. They were the first group of large-bodied life forms on Earth, appearing 575 million years ago and persisting for about 33 million. All of the 110 known species were soft-bodied and lived in the sea; most were large and immobile. The largest were over 4 metres long, although most types would have fitted comfortably into a shoebox.

For everything we know about the Ediacarans, however, there are still many mysteries. For 3 billion years, life on Earth had been microbial. What caused the sudden increase in size and complexity? What kind of creatures were the Ediacarans, and what led to their demise? Perhaps the biggest mystery of all is their biological affinity. Where do the Ediacarans fit on the tree of life, and how – if at all – were they related to the creatures of the Cambrian explosion?

In the past few years, palaeontologists have been edging towards consensus on all these points. Most now split the Ediacarans into two distinct groups. The first half of their reign was dominated by a type of organism utterly different from any life form on Earth today. From about 560 million years ago, however, new types evolved, and some of them went on to inherit the Earth.

When the Ediacarans were first described, palaeontologists thought they could slot them comfortably into known animal groups. They were interpreted as primitive ancestors of the creatures of the Cambrian explosion: worms, corals, arthropods, jellyfish and the like. However, as more fossils were found, more and more peculiarities came to light, and doubts mounted. It became clear that the majority of Ediacarans were oddballs of uncertain affinity. In the mid-1980s, Adolf Seilacher of the University of Tübingen in Germany tore up the rulebook. He argued that even though Ediacarans resembled known animals, they were, in fact, a hitherto unrecognised and extinct kingdom of life – a dead-end experiment in evolution that disappeared at the end of the Precambrian.

Earliest Ediacarans

For much of the past 20 years the debate has been polarised between those who believe that Seilacher was correct and those who maintain that the Ediacarans are the “long fuse” of the Cambrian explosion. It now turns out that both camps are, to some extent, right.

The breakthrough began in 2003 when Guy Narbonne of Queen’s University in Kingston, Ontario, Canada, and Jim Gehling of the South Australian Museum in Adelaide reported the discovery of the oldest known Ediacaran fossils. They had been scouring Mistaken Point on the Avalon Peninsula in Newfoundland, Canada, a spot already well known for Ediacaran fossils, when they found a cache of fossils that date back at least 575 million years, 10 million years older than the previous oldest (Geology, vol 31, p 27).

You might be forgiven for expecting the earliest complex life on Earth to be quite small. Narbonne and Gehling certainly did, but they were wrong. “Our search image was for things about the size and shape of our thumbnails,” says Narbonne. “Instead, we came up with 2-metre-long fronds.”

Those giant fronds looked similar to Charnia masoni, the original Ediacaran, but were about 10 times longer. They also discovered a solitary specimen of a 4-metre-long fossil that looked a bit like a champagne flute. It is the biggest Ediacaran ever found, but beyond that its identity is a mystery; it has yet to be formally described or named.

The two giants weren’t the only new species. There was Thectardis, which is the size and shape of an ice cream cone, and a small frond fossil. In addition there was a disc-like organism, Ivesheadia, which has the size and appearance of a pizza, complete with a circular crust and pepperoni.

Despite the novelty of these fossils it is clear that they are closely related to what comes later. The two frond fossils are species of Charnia, while Ivesheadia is known from the Charnwood Forest deposits – recently dated to 562 million years ago – and younger rocks at Mistaken Point. These younger rocks also contain similar, though more diverse, creatures: a host of other discs and fronds, the cabbage-shaped Bradgatia and several new and as yet unnamed creatures, including one like a spindle and another like a feather duster.

Together, these first Ediacarans are known as the Avalon assemblage and are the earliest known large, complex life forms on Earth. They dominated for 15 million years.

So what were the Avalon life forms like? Analysis of the sedimentary rocks they are found in suggests that they were exclusively deep-sea dwellers, living an estimated 500 metres to 2 kilometres down. They did not move, either lying on the seabed or hovering above it, anchored by a holdfast. The fronds resemble ferns or seaweed but they cannot have been either: water at that depth would have been too dark for photosynthesis.

In 2003, a team led by Narbonne studied more than 100 fossil beds at Mistaken Point to try to work out how the early Ediacarans made a living. They found that although these Ediacarans may not have looked like animals, they certainly acted like them. They were either bottom feeders or fed on organic matter in the water column, just as in modern marine environments (Paleobiology, vol 29, p 527). “Ecologically they behave more like animals than anything else,” says Narbonne.

In other respects, however, they are distinctly unlike animals. Kevin Peterson of Dartmouth College in Hanover, New Hampshire, has found that the Avalon creatures did not have any animal-like means of capturing or eating food: no appendages for grabbing, no mouth, no gut and no anus. Nor did there appear to be anything controlling their growth. Nearly all animals stop growing when they reach a certain size: not so with these organisms. In these important respects, they resemble fungi rather than animals. Scientists had to conclude that they could not place them with any living types of animal.

Then, in 2004, came a spectacular discovery. At another Newfoundland location called Spaniard’s Bay, Narbonne found a group of around 100 frond-shaped fossils in an exquisite state of preservation (Science, vol 305, p 1141). Ediacaran fossils are nearly always buried in coarse-grained sediment that does not show very much detail, but the new fossils were different. Details as fine as one-thirtieth of a millimetre were preserved, meaning that Narbonne could work out how these organisms were built.

He found that they were made in a completely different way from other animals: they were fractal. Each frond was built up from smaller, identical fronds, and each of these was composed of yet smaller fronds, and so on, down to the smallest scale visible. Narbonne called these fossils “rangeomorphs” after a similarly frondy fossil called Rangea that had been described years earlier.

There was more. When Narbonne re-examined the Mistaken Point fossils, he found that almost all the species shared the same fractal construction. “It’s like something out of science fiction,” says Narbonne. “Imagine inventing a completely different way of making animals. Yet this really did happen and it dominated life for 15 million years.”

Many researchers are coming round to the view that the rangeomorphs were a unique group unlike anything alive today. The most likely scenario, says Peterson, is that they were a primitive type of animal with fungus-like traits that left no living descendants. Narbonne and Gehling agree. “The rangeomorphs were a failed experiment in animal evolution,” says Narbonne. “They were not ancestral to anything in the present world.”

“The rangeomorphs are a failed experiment in animal evolution”

Why did they evolve when they did? For 3 billion years life had been microbial. What changed to suddenly make large, complex designs so desirable? The key may lie in the fact that, just 5 million years before the rangeomorphs appear, Earth emerged from the last major ice age of the Precambrian. The Gaskiers glaciation, which ended 580 million years ago, closed a period known as Snowball Earth, when much of the planet froze over. This year, a study has found that there was a massive increase in deep-sea oxygen levels after the Gaskiers glaciation (Science, vol 315, p 92). “Things are getting big, and the rise in oxygen has to be key to this,” says Gehling.

“Things are getting big, and the rise in oxygen has to be key to this”

Peterson argues further that the primeval ocean would have been bursting with dissolved organic matter freed up by melting glaciers, which the rangeomorphs may have absorbed through their body surface. “Dissolved organic matter is concentrated in deep water where Ediacarans are feeding,” says Gehling. Their fractal, modular design was an efficient way of mopping up this sudden burst of plenty.

The fractal construction may also have been what allowed them to evolve to get so big so quickly. The rangeomorphs were small units writ large through fractal repetition, and that process may have been quite straightforward. A simple genetic switch may have been all that was required for tiny frondlets to iterate themselves into 2-metre giants.

Conquering the shallows

The rangeomorph design remained a winning strategy for 15 million years but it appears to have had a major drawback: it only worked well in the deep sea. Extensive searches have been made for shallow-water Ediacarans that are more than 560 million years old, but none has been found. When shallow-water Ediacarans do begin to appear around 560 million years ago, no rangeomorphs have been found among them. It appears that the rangeomorphs were left behind in the deep.

The biological relationship between the rangeomorphs and the new, shallow-water Ediacarans is highly uncertain. What is clear is that the appearance of shallow-water Ediacarans marks the beginning of a golden age. For 10 million years, biodiversity was at a peak. “It was brimming with life,” says Mary Droser of the University of California, Riverside. “We have bedding planes that are absolutely covered – wall-to-wall fossils.”

The most spectacular shallow-water Ediacarans come from the White Sea region of Russia and the Ediacara Hills of Australia, and are collectively known as the White Sea assemblage. Among them are some well-known fossils such as Dickinsonia, a flatworm-like creature that reached up to a metre in length, and Tribrachidium, a 2.5-centimetre disc topped with three spiralling arms, one of the most enigmatic of all the Ediacarans. It has been described by some as a failed experiment in animal design, although others argue that it may be related to sponges. Another is Spriggina, regarded by some palaeontologists as either a type of worm or an early crustacean, though this has been hotly disputed by others. There’s also the mollusc-like Kimberella, which may have been among the first motile animals.

Even amid all this evolutionary innovation, the rangeomorphs clung on. There are some in an Australian group from 555 million years ago, though they are rare and are found only in the remains of deep-water canyons alongside rocks laid down in shallow waters. Their species composition is similar to the Avalon group. “There are almost two worlds in parallel,” says Gehling.

In one of those worlds, pivotal evolutionary events were unfolding. Most of the well-known shallow-water Ediacarans were large and immobile and have no clear link to the creatures of the Cambrian. However, recent work by Gehling and Droser hints at the beginnings of the Cambrian explosion in a hitherto neglected group of small Ediacarans, 1 to 10 centimetres in length. One of these, Parvancorina, bears a close resemblance to a recently discovered early Cambrian arthropod called Skania (Historical Biology, vol 18, p 33). Another, Arkarua, looks a lot like a Cambrian echinoderm, the group represented today by starfish and sea urchins. Like the mammals that coexisted with dinosaurs, it appears that some small Ediacarans went on to inherit the Earth. The Cambrian explosion had a long fuse after all.

While many scientists agree with this new interpretation, dissenters remain. “No Ediacaran is really diagnostic of any animal phylum,” says Martin Brasier of the University of Oxford. “Mostly, people have been trying to shoehorn them into existing groups.”

Unfortunately, a definitive verdict on the link between the Ediacarans and the Cambrian will be hard to reach. The fossil record tells us little about the final few million years of the Ediacaran period. Late Ediacarans are best known from Namibia, and range from 549 to 543 million years old. These are all shallow-water dwellers. The best known of this so-called Nama assemblage are frond-like forms such as Pteridinium and Swartpuntia, and the coral-like Cloudina; all are difficult to place on the tree of life.

Intriguingly, the Namibian group includes a shallow-water rangeomorph, Rangea. This conquest of the shallows was short-lived, though. The rangeomorphs disappear altogether about 545 million years ago, probably doomed by their own limitations. As long as organic nutrients were freely available, the rangeomorphs had no problem. But towards the end of the Precambrian, as this food supply became depleted by all the creatures now eating it, they ran into trouble. They were up against far more efficient feeders that drove them to extinction.

A handful of Ediacarans crossed over into the early Cambrian, including Swartpuntia (Journal of Paleontology, vol 74, p 731). The overwhelming majority did not make it, though; the few that did vanished within 5 million years. The first great experiment in complex, multicellular life was over. But we now know it laid the foundation for everything that followed.