¿ìè¶ÌÊÓƵ

Five hundred million years ago, something leapt from the mud at the
bottom of the sea – and a new act began in evolution’s long drama. The protagonists
were colonies of tiny, filter-feeding animals. For millions of years, these
oddball creatures, known as graptolites, had been confined to playing a
bit part on the ocean floor. But this was their big break: some colonies
had evolved the ability to drift free of the mud, and this meant they could
begin to feast on the oceans’ vast and as yet untapped reserves of edible
microplankton. They never looked back. For the next hundred million years
the open oceans swarmed with plankton-eating graptolite colonies.

It was a long and eventful reign. When the graptolites first leapt into
the realms of the plankton, life was still in its infancy. The vast landmass
of Gondwana still smothered the South Pole, uninhabited by plants or animals,
while on the ocean floor molluscs, trilobites and other shelled invertebrates
vied for dominance. Yet by the time the graptolites died out, evolution
had moved on at a smart pace. The first fish had appeared, amphibians had
taken their first tentative steps onto land, and the earliest spore-bearing
plants were beginning to take root.

The graptolites themselves were no evolutionary slouches. From the thousands
of different species preserved in the fossil record, it is clear that their
leap into the open oceans sparked off a phase of rapid evolution. At any
one time tens or even hundreds of planktonic graptolite species would have
co-existed in the oceans, most lasting for one or two million years before
dying out.

Rock markings

At first glance the fossil remains of these species look like little
more than serrated pencil marks on rock surfaces (‘graphein’ is Greek for
‘to write’; ‘lithos’ means ‘rock’). But a closer inspection reveals the
remnants of intriguingly sophisticated colony structures – complex patterns
woven from precisely engineered tubes which were themselves made from secretions
of a collagen-like protein, akin to the substance in finger nails. In some
species, the solid walls of the tubes are transformed into delicate meshworks,
while others sport long spines tipped with curious float-like structures.

Until recently, researchers believed these rock markings were all that
remained of the planktonic graptolites. They thought they would never understand
the workings of these colonies. How could the colony members – simple multicellular
organisms known as ‘zooids’ – produce such complicated skeletons? What use
could the zooids have for these structures? The ecology of the graptolites
seemed destined to remain a closed book, as did the reasons for their mysterious
demise.

Then a few years ago there was a breakthrough. A team of French scientists
working off the coast of New Caledonia, an island in the southwestern Pacific
east of Australia, dredged up a strange creature from the muddy depths of
the ocean. No more than a centimetre wide, and made up of a tangled mass
of tubes and spines, it seemed to bear the hallmarks of a graptolite.
Researchers have since studied the creature in detail and are now confident
about its identity. It is tangible evidence, they say, that at least some
graptolites avoided extinction and are still eking out an existence on the
seafloor from which they escaped 500 million years ago.

The discovery sets the seal on a detective story that has been slowly,
fitfully unfolding for more than a century. For most of this time, palaeontologists
could only mull over the fossil remains. The lack of any preserved soft
tissue in these fossils made it difficult to answer even the simplest questions.
What kind of animals were the graptolites? Were they related to other colonial
organisms like corals? And where did the collagen-like protein needed to
make the tubes come from? One possibility was that the graptolite zooids
themselves secreted it. Another was that the protein was produced by a layer
of specialised secretory tissue wrapped around the tubes, as is the case
in corals.

The question was hotly contested. An important clue came in the 1930s.
Digging in chert in the Holy Cross Mountains in Poland, a palaeontologist
called Roman Koztowski from the University of Warsaw discovered some unusually
well-preserved fossils of bottom-dwelling graptolites. Koztowski observed
that the tubular skeletons of these fossils closely resemble ones found
in an obscure group of living colonial marine animals called pterobranchs.
As far as we know, pterobranch colonies have always been rare, and today
only a handful of species, belonging to two genera, Rhabdopleura and Cephalodiscus,
survive.

Perhaps, thought Koztowski, the filter-feeding zooids of these living
pterobranchs held the key to understanding the extinct graptolites. Pterobranch
zooids are hardly sophisticated. Each comprises little more than a ring
of tentacles, a gut and a fleshy lobe that secretes collagen-like protein.
Yet, as Koztowski noted, these pterobranch zooids are quite capable of building
their own tubular skeletons, using their fleshy lobes to secrete rings
of collagen which can be linked to form a tube. And the autonomy of pterobranch
zooids is striking: if they need to, they can leave their tubes to repair
damaged tissue or construct spines which help in feeding.

DIY homes

Could graptolite zooids function in the same way? At first, many palaeontologists
thought not. The skeletons of graptolites and pterobranchs are broadly similar
– tubular structures made up of successive rings. But there are some big
differences. Whereas pterobranchs have tangled, irregular skeletons, graptolite
skeletons are as regular as they are intricate. Moreover, unlike the pterobranchs,
each graptolite species has its own particular colony architecture, based
on a spiral form, for instance, or a complex branch pattern. Not only would
each graptolite zooid have to build its own tube, but it would also have
to cooperate with its neighbours in an architectural feat as impressive
as the building of a beehive or wasps’ nest.

Most people thought this would be impossible for such primitive creatures,
and continued to believe that there must have been a layer of tissue around
the graptolite zooids to secrete the skeleton. The activities of the humble
zooids would be confined to feeding and breeding.

This orthodoxy was overturned about 15 years ago by a scanning electron
microscope study of graptolite skeletons. Peter Crowther and Barrie Rickards
at the University of Cambridge, and Dietmar Andres at the Free University
of Berlin, discovered that the skeletal surfaces were covered with irregular,
criss-crossing pairs of parallel lines. Closer inspection showed that these
parallel lines were in fact flat bundles of collagen fibres. The whole
skeleton was enveloped in these protein-based bandages like an Egyptian
mummy.

Crowther, Rickards and Andres suggested that these structures had been
formed by the secretory lobes of pterobranch-like zooids crawling over the
skeleton surface. But opponents of this idea raised awkward questions. How
could crawling zooids produce the long spines and float-like structures
seen in many graptolites? How could they produce the complex meshwork skeletons
found in some species?

Enter the French expedition. When Helmut Zibrowius, one of the zoologists
on the expedition, first saw the unusual creature that had been dredged
up from the deep, he identified it as a pterobranch, but one that had never
been seen before, with spines. He sent it to the world’s foremost authority
on this group, Noel Dilly, then working at St George’s Hospital Medical
School in London. When Dilly examined the creature more closely he realised
he was dealing with a major find. The clue lay in the spines, which were
made of small cylinders of collagen – just like those of graptolites.

These spines couldn’t possibly be formed by a single layer of tissue,
argued Dilly; rather, they were the handiwork of zooids. Dilly envisaged
each zooid crawling to the open end of its tube to feed and then secreting
a blob of collagen before crawling back inside. Next time the zooid crawled
out to feed, the blob would have hardened and the tiny animal would secrete
another collagen blob above it. So the spine would grow until it was many
times longer than a zooid body.

Most researchers now accept that the graptolite skeletons outlined so
vividly in the fossil record were built by teams of cooperative zooids like
those of the pterobranchs. The spines, floats and meshworks were all the
work of these simple animals with primitive nervous systems. Dilly went
even further. If, as seems to be the case, pterobranchs and graptolites
share the same kind of spine structure, perhaps pterobranchs are bottom-dwelling
graptolites – the last, reclusive survivors of a once-mighty group thought
to have been extinct for 400 million years. To make the point, Dilly called
the new species Cephalodiscus graptolitoides.

Working together

Not everyone agrees with Dilly that pterobranchs are living examples
of graptolites. Sceptics still believe that the fossilised skeletons of
graptolites are too complex to have been built by zooids, and are instead
the handiwork of an enveloping layer of tissue. The debate can probably
only be settled with further research into pterobranch zooids. A key goal
for Dilly and like-minded researchers must now be to show that these zooids
have the rudiments of cooperative behavioural programmes, of the kind that
would be needed to build meshwork structures.

But if Dilly is proved right, then the living examples of graptolites
will be a rich source of information about their extinct cousins. They
might throw light, for example, on why graptolites diversified so rapidly
during their 100-million-year reign. Palaeontologists often exploit the
many different kinds of graptolite fossils as markers to date rocks and
other fossils. Yet the reasons for their rapid evolution are unclear.

Research into living graptolites might also help scientists studying
the evolution of the oceans. Graptolites were widespread in many oceans
that disappeared hundreds of millions of years ago. Reconstructing the evolutionary
history of the graptolites could tell us how the temperatures and organic
productivities of these vanished oceans varied over time .

Were geological upheavals responsible for the decline of the graptolites?
Nobody knows. The free-swimming graptolites died out a little under 400
million years ago, although most bottom-living species hung on for another
100 million years. Perhaps the earliest fishes forced the graptolites into
decline by preying on them or stealing their food. But this doesn’t square
with the fact that most graptolites were already extinct before fish became
widespread. Alternatively, some as yet undiscovered soft-bodied animals
might have preyed on the graptolites, or outcompeted them in exploiting
plankton. In the end, maybe it was the graptolites’ slowness which let them
down. Perhaps, like the galleons or clippers of past centuries, the graptolites,
with their ornate structures, were eventually outperformed by more mundane
but effective vessels.

Jan Zalasiewicz and Sue Rigby are palaeontologists with the British
Geological Survey. Sue Rigby is also a lecturer at the University of Leicester.

* * *

Floating and feeding

During their 100-million-year reign of the oceans, graptolite colonies
adopted many unusual shapes. To find out why, I constructed some life-size
and scaled-up models of graptolites out of aluminium tubing and clingfilm
(see right).

The models – some simple, some highly complex – had one surprising feature
in common. When allowed to float in a large tank of water, they all rotated
slowly as they fell towards the bottom. This simple adaptation would have
been enough to fulfil the two key requirements for a successful graptolite
structure – stabilising the colony in seawater, and providing an efficient
feeding platform for the zooids. An unsolved problem, though, is how such
structures floated back upwards again. Did they rely on drift and buoyancy?
Or did they swim?

Feeding efficiency seems to have been a driving factor in graptolite
evolution. Experiments with models show that each colony would have fed
from microplankton in a column of water whose cross-sectional area was governed
by the colony’s shape. The intensity of feeding would also depend on shape,
since some shapes could pack more zooids into a given cross-sectional area
of water.

Using models, graptolites of different shapes and ages can be plotted
onto a feeding area and intensity graph. The result is illuminating. It
shows that although graptolites evolved new shapes very rapidly, each new
group was designed to fill the same feeding niche as the group it replaced.
Graptolites from the same habitat, but separated in time by 100 million
years, often look very different. But the models suggest that their shapes
enabled them all to filter water at comparable rates.

Access to food also governed how long the graptolites could live, or
so fossil studies suggest. The dark grey shales that outcrop along the
St Lawrence River near Quebec are teeming with graptolite remains. Here,
over the past three years, my colleagues and I have been measuring individual
fossils to glean information about their lifespans.

In unpredictable or harsh environments, animals are just as likely to
die young as old, and few fulfil their maximum potential lifespan. A graph
of survivorship against age will form a straight line. But in hospitable
environments with few predators, animals are more likely to survive to
old age and the graph is strongly convex.

Along the St Lawrence River the most common graptolite, Orthograptus
quadrimucronatus micracanthus, shows a convex survivorship curve. They range
in length from 1 millimetre to just over 3 centimetres, and comparison with
growth rates in their closest living relatives, the pterobranchs, suggest
that the largest colonies were about three years old.

Three years is a short lifespan for colonial organisms. Most live for
much longer because of their ability to reproduce asexually, producing new
members that carry the genome free of age-related defects. So what curtailed
the graptolites’ lifespan in waters off Quebec? The limiting factor was
probably food. Any colony growing to more than about 3 centimetres would
not have been adequately nourished by the microplankton in the water column.
Large colonies starved to death.