AIMING to understand rainforest trees? Assume they’re all the same. Snails,
shrimps and other sea-floor fauna? Set the distinctions aside. The birds of
Britain? Forget everything we’ve ever learned about ornithology and treat them
as a flock of flapping clones. The key to explaining biodiversity, says
ecologist Steve Hubbell, is to ignore it.
It’s a startling idea. If Hubbell is right, abundant species such as mallard
ducks haven’t “earned” their success by being somehow fitter or better than
rarer species such as white-headed ducks. Instead, species wander erratically
between dominance and extinction the way drunkards reel down a street. It’s not
survival of the fittest—it’s survival of the luckiest. Yet even in this
indeterminate and nihilistic world, Hubbell, a plant ecologist at the University
of Georgia, has found a sort of order—a way to predict how many species,
and in what numbers, any given habitat will support.
Harvard biologist Edward O. Wilson says ecologists will some day rank
Hubbell’s contribution among the most important of the past half century. In the
meantime, though, a lot of people are scratching their heads trying to figure
out how such an obviously wrong assumption can work so well. “I think a lot of
people would like to just call him a crackpot and ignore him,” says
palaeobiologist Doug Erwin of the National Museum of Natural History in
Washington DC. “But that will be hard.”
Advertisement
Hubbell’s work drives straight to the heart of the science of ecology. For
all the attention paid to biodiversity, ecologists still can’t answer the most
basic question about it. Take any plot of land or patch of sea. Why does it
contain this many plant and animal species, and not more or less? Why is this
species common and that species rare?
You only need step into the wild to grasp the size of the problem. With up to
300 tree species in a single hectare of tropical forest, and 100 types of coral
and 500 species of fish in one stretch of reef, biodiversity is almost
unfathomably complex. Or is it?
When Hubbell set out to understand tree diversity in a tropical forest in
Costa Rica back in the 1970s, he certainly expected it to be complex. A host of
factors could have determined the mix of species and the range of abundances he
saw: variable soil conditions, uneven distribution of the trees’ pollinators,
predators and other associates, and differences in the risk of infestation and
disease, to name a few.
But before wading into all that, Hubbell decided he’d better be clear about
his starting point: what would a forest look like if none of those factors
mattered? What if chance were all that counted, and species had merely drifted
to their present abundances, propelled by random fluctuations in the numbers
that died and reproduced each year?
It seemed like a trivial exercise. No climate, no chemistry, no biology. Just
a simple tally of deaths and births. Hubbell created a computer model of the
forest that used just a single rule—one that came to him as he stared up
at the canopy. “I realised there was no more room to fit trees in there,” he
says, so he decreed that each new tree could grow only in the vacancy left by
another’s death. Hubbell used the computer’s random number generator as the hand
of fate, dealing death and bestowing offspring to all individuals with equal
probability.
As the simulation rolled on and generations of cybertrees passed, Hubbell
tabulated each species’ share of the total tree population. He expected that the
result of the randomness would be erratic patterns that bore no resemblance to
the real world. He was in for a shock. “I got patterns that looked very similar
to the patterns I was seeing in the forest,” he says. In other words, it looked
as though randomness ruled in nature herself.
When Hubbell published his results in 1979, most biologists were dismissive.
Sure, they said, maybe identical trees could produce patterns like those in
nature, but real trees aren’t identical. “It’s kind of a fundamentally wrong
assumption,” says palaeobiologist John Pandolfi, a colleague of Erwin’s.
But Hubbell wasn’t arguing that species really were identical—just that
the differences between them might be “neutral”. In other words, he was
suggesting that the differences had nothing to do with how common a species is,
how competing species coexist, or why a habitat holds exactly as many species as
it does. And this is where Hubbell’s audience really baulked.
For decades, ecologists have viewed the characteristic features of a species
as “niche specialisation”—adaptation to a way of life that is uniquely its
own. In their thinking, having a niche of your own is crucial to
survival—and finding a bountiful niche is the key to success, because you
can exploit a bigger portion of the world’s resources. Faith in this is what
drives biologists to uncover what makes every species special. For many,
explaining biodiversity means finding, for instance, 100 different ways to be a
coral. It’s a task that has launched a thousand academic careers.
In the face of this entrenched belief, Hubbell spent the next two decades
refining and expanding his theory. Last year he finally published a book-length
treatise entitled The unified neutral theory of biodiversity and
biogeography—a title grandiose enough to have earned him some ribbing. But
then again, Hubbell delivers the goods. His theory ties together an
unprecedented range of biological processes, from the architecture of evolution
to the statistics of abundance and the basis of diversity.
Hubbell’s core achievement is a formula that for any given place and life
form—trees, grazing animals, fish or whatever—predicts how many
species will coexist and in what abundances. The formula itself is fiendishly
complicated. Even Hubbell himself can’t write it down for more than a handful of
species at a time, and for many realistically sized groups of
species—British breeding birds, for example—the calculations are too
much even for his PC. Despite this, Hubbell has been able to produce enough
predictions from his formula to go into the wild and see how it measures up.
The whole complex formula hangs on just four simple variables, each detailing
a real and, in principle, measurable aspect of the environment. The easiest way
to appreciate these factors is with an example—Hubbell’s favourite is
trees on Barro Colorado island in Panama. The first variable is the total number
of trees in the area under study. This number reflects the overall suitability
of that environment for trees—its soil, sunlight, climate, pests and so
on. The second is the total number of trees in the ecological
“universe”—in this case, the dry tropics of Central America or
thereabouts. This represents the pool from which immigrants will arrive and new
species be born.
The third variable is the rate at which individuals immigrate to the study
area from elsewhere in this universe. In this example, it’s the rate at which
wind and birds and bats carry seeds to Barro Colorado. The fourth and final
number is the rate at which new species emerge within the larger universe.
In balance
Together, these four factors describe a biodiversity equilibrium in which new
species arrive at the same rate as others become extinct. Diversity can’t go on
increasing forever because Hubbell includes the biologically realistic
assumption that the total number of individuals is constant and there’s no space
for more. This means the more tree species your habitat has, the fewer trees you
can have per species. But smaller populations fluctuate more erratically than
large ones, putting them at greater risk of extinction. So as biodiversity
climbs, the rate of extinction rises to meet it and you get an equilibrium.
Hubbell’s theory hasn’t come out of the blue. Wilson and his colleague Robert
MacArthur proposed something similar in 1963 when they predicted that an
equilibrium between immigration and extinction determines species diversity on
islands. Just as in Hubbell’s theory, niche adaptations and the idiosyncrasies
of species played no role. They supported their hypothesis with data showing
that real-life islands near the mainland—where immigration is more
likely—have more species than comparable islands far away from it.
But Hubbell goes far beyond MacArthur and Wilson’s island hypothesis. By
including a third factor alongside immigration and extinction—the
emergence of new species from old—he has shown that the equilibrium
unfolds on a continental scale and across evolutionary time. And because his
formula predicts not just the number of species but their exact relative
population sizes, it’s much easier to see the equilibrium in the real-world
data.
What’s especially remarkable about this equilibrium is that in a certain
sense it doesn’t look like one. The number of species and the pattern of
abundances appear roughly constant, but species would be constantly moving
around within this pattern—emerging, drifting up and down the abundance
rankings and going extinct. This is a radical departure from the traditional
niche perspective, which sees the composition of ecosystems as
stable.
While Hubbell’s ideas may offend many researchers’ sensibilities, nearly
everything about his theory is testable—at least in principle. Even the
dourest of sceptics can go into the wild, estimate Hubbell’s four crucial
numbers and then plug them into the formula to see if it predicts the true
pattern of diversity. Erwin expects it to provoke many into doing just
that—effectively forcing them to take the theory seriously.
No one has put the theory to a thorough test yet, but Hubbell has made a
start for a few groups of species, using hard numbers for the factors he can
measure in the field and whatever numbers work best for the other
variables—provided they are plausible. On Barro Colorado, for example, his
census of 20,541 trees includes many more rare species than you’d expect if abundances
fell on a bell curve. But Hubbell’s prediction hugs the data like a tailor-made suit
(see Diagram).
He has fashioned equally impressive fits for bees, bats, birds, fish and frogs,
as well as other trees in Malaysia, Belize and Kentucky.
In fact, Hubbell has probably put more biology under one mathematical roof
than anyone yet. In particular, he is the first to root modern patterns of
abundance and diversity in the process of evolution. A real test of this link
will take several years and require more detailed analyses of abundances in the
fossil record or the sampling of many more species’ DNA. Even so, the theory’s
evolutionary implications are nearly as provocative as the ecological ones.
The main lesson from Hubbell’s theory is that success begets success. Species
lucky enough to rise to prominence will persist longer and produce new species
more frequently and are more likely to sire a long line of descendant species.
That means the more abundant species should tend to be older as well, says
Hubbell—another testable prediction that is certain to attract
interest.
Some critics scoff that by tuning the four factors you can make Hubbell’s
model fit any pattern, random or not. But Jim Brown, an ecologist at the
University of New Mexico in Albuquerque, says many theories before Hubbell’s
have aspired to explain far less with much more flexible formulae. Even with
that leeway, none fits the data as well as Hubbell’s, says Brown. Moreover,
Hubbell’s key variables reflect real features such as the mobility of the
species under study, rather than arbitrary fudge factors. “I think it’s fair to
say his model is the best game in town,” Brown says.
Brown regards Hubbell’s theory as just a stepping stone toward a more
complicated theory, which he says will eventually have to take species’
differences into account. But others say Hubbell’s formula already captures
something true about natural systems and is ready for use. Even Pandolfi thinks
so— although he predicts the theory will have trouble explaining the
patterns found in most other real ecosystems, such as the coral reefs he
studies.
Pandolfi believes Hubbell’s theory plays a role analogous to the theory of
genetic drift, with which it shares a good deal of mathematics. Genetic drift
refers to changes in the prevalence of certain genetic variants as a result of
random differences among individuals in how many offspring they produce.
Scientists hunting for the footprints of natural selection know to subtract the
effects of drift, which can make a gene common in a population even when it
brings no advantage to individuals that inherit it.
Hubbell’s theory—which he sometimes describes as a “theory of
ecological drift”—might provide a similar detective tool. “Where we get
disparity from what the model predicts, it’s going to be extremely interesting,”
as Pandolfi puts it. “It’s actually going to provide us with a way of evaluating
the degree to which adaptation influences community structure.” Hubbell cites
the example of a patch of forest along the Manu River in Amazonian Peru, where a
few of the most common trees are about twice as abundant as his formula
predicts. He proposes that something biologically unusual about these species
makes them excel, yet his calculations show that their edge in this case would
have to be quite small. This means that even small oddities cast a long shadow
against the backdrop of the formula’s predictions, Hubbell says.
The biggest oddity so far, however, seems to be how little is left to
explain. How can a theory throw away everything biologists have ever learned
about species, yet explain what its competitors can’t? “We have to work that
out,” says Brown. Hubbell sees this question as a major challenge for the
future. In the meantime, ecologists may just have to embrace the theory first
and ask the questions later.
-
Further reading:
The unified neutral theory of biodiversity and biogeography,
by Stephen P. Hubbell, Princeton University Press (2001)