A hurricane created a rare scientific opportunity: Scientists turned seven empty Bahamian islands into a living laboratory to watch lizard evolution unfold from the beginning

A hurricane that devastated several small islands in the Bahamas gave scientists something nature rarely offers: the chance to watch evolution begin again.

In 2004, a powerful storm wiped out brown anole lizard populations on a group of tiny islands. The destruction left behind more than ecological damage. It created an unusual opportunity to investigate one of evolutionary biology’s longstanding questions: How much of a species’ future is determined by chance, and how much is shaped by natural selection?

A team of researchers led by biologist Jason Kolbe used the empty islands to conduct a carefully designed experiment. In 2005, they introduced one male and one female brown anole to each of seven islands, taking the lizards from the same large source island near Great Abaco.


The experiment, whose findings were published in Science in 2012, offered a rare glimpse into how new populations develop from a tiny number of founders. It also revealed that even when nature pushes populations in the same evolutionary direction, their starting point can leave a lasting mark.

A natural disaster opened a window into evolution

Evolution is often reconstructed from fossils, genetic records and comparisons between species living in different environments. But those methods have a limitation: they rarely allow scientists to observe the exact moment when a new population begins.

The brown anole experiment was different.

The hurricane had effectively cleared several islands of their lizard populations. Instead of waiting for the islands to be recolonised naturally, researchers used the opportunity to introduce carefully selected founding pairs and monitor what happened over time.

The species at the centre of the study, the brown anole (Anolis sagrei), is well suited to this kind of research. These lizards are found across the Caribbean and are known for adapting to different habitats, including forests, scrubland and other environments with distinct vegetation.

The researchers selected a large, densely forested island as their source. Its lizards generally had longer hindlimbs, a trait suited to moving among the wider branches and surfaces found in that habitat.

The seven smaller islands, by contrast, offered a different setting. Their vegetation consisted largely of narrow, scrubby plants and thinner branches. Such surroundings were expected to favour lizards with shorter hindlimbs, which could move more effectively through the available habitat.

That contrast created the central test: Would all seven populations gradually become alike because they faced similar environmental pressures, or would their origins continue to influence their evolution?

Fourteen lizards became the starting point for seven populations

In 2005, the researchers released exactly two brown anoles on each of the seven islands—one male and one female.

The total founding population was just 14 lizards.

Every pair came from the same source island, helping researchers reduce the possibility that differences between the new populations were caused by unrelated environmental or geographical factors. The islands were also similar enough in their broad ecological conditions to make comparisons meaningful.

The key scientific idea was the founder effect.

When a new population begins with only a few individuals, those founders carry just a small sample of the genetic variation present in the larger source population. Their particular combination of inherited traits may not represent the average characteristics of the original group.

That means a new population can begin with features that are unusual in the wider species—not necessarily because those features are better suited to the new environment, but simply because the founding individuals happened to carry them.

In this experiment, one pair might carry genes associated with relatively long hindlimbs, while another pair from the same source population might carry a different combination associated with shorter limbs.

The researchers could then ask whether those initial differences would disappear under natural selection or remain visible as the populations developed.

The first year revealed the power of chance

The early results were striking.

Within the first year, the seven island populations differed in hindlimb length. Some had relatively long-legged lizards, while others had shorter-legged individuals.

Those differences did not line up with the vegetation on the islands.

If the environment alone had determined the lizards’ traits from the outset, researchers would have expected populations living in similar habitats to show broadly similar characteristics. Instead, the early differences reflected the particular genetic combinations brought by the founding pairs.

This was precisely the kind of pattern predicted by the founder effect.

The result mattered because it demonstrated how a population’s evolutionary starting point could be shaped by an apparently random event. Two populations of the same species, beginning in comparable environments, could follow different paths simply because different individuals happened to establish them.

The hurricane had created the opportunity, but the founding pairs supplied the first evolutionary differences.

Natural selection changed the lizards—but did not erase their origins

The researchers continued tracking the seven populations for four years, from 2005 to 2009. They examined changes in hindlimb length and genetic markers, comparing the experimental populations with lizards on nearby source islands.

Over time, the island populations showed the expected movement towards shorter hindlimbs. The smaller branches and scrubby vegetation favoured traits suited to that environment, providing evidence of natural selection.

But the story did not end there.

The populations that began with relatively long-legged lizards still tended to have longer-legged lizards four years later. Those that started with shorter-legged individuals remained relatively short-legged.

In other words, natural selection was pushing the populations in a similar direction, but it was not making them identical.

The findings, discussed in a 2012 Nature Genetics summary, showed that both morphological adaptation and genetic variation could be observed in the experimental populations. Founder effects remained detectable throughout the study.

The experiment demonstrated a central principle of evolution: chance can determine where a population starts, while natural selection influences the direction in which it changes.

Neither process alone explained the final differences between the seven island populations.

Why this experiment changed the way scientists study evolution

The significance of the brown anole study extended beyond lizards.

For decades, evolutionary biologists had recognised the founder effect as an important process in population genetics. However, many real-world examples involved populations that had been established thousands or millions of years earlier.

By the time scientists discovered them, the original founders were long gone. There was no way to know exactly which individuals had started the population or what genetic variation they had carried.

That made it difficult to separate the influence of chance from the effects of later natural selection.

The Bahamas experiment addressed that problem in a remarkably direct way. Researchers knew the founding individuals, the source population and the approximate starting conditions. They could observe the populations at the beginning and return to study them over subsequent years.

It was not simply a comparison of two existing lizard populations. It was an opportunity to follow newly established populations from their earliest stage.

The work also highlighted why evolution does not always produce the same outcome in similar environments. Natural selection may favour similar traits across different locations, but the genetic variation available to each population can influence what happens next.

A population cannot evolve traits from genetic variation it does not possess. The founders help determine which possibilities are present at the start.

The wider lesson: Evolution is shaped by both history and environment

The brown anole experiment offers a useful way to understand why species living in similar habitats can still differ.

Consider two populations facing the same environmental challenge. Both may benefit from shorter limbs, better camouflage or improved tolerance to local conditions. Yet if one population begins with a different genetic makeup, its response may not follow exactly the same path as the other.

This is known as historical contingency: the idea that evolutionary outcomes can depend partly on what happened earlier.

The seven islands showed this principle in action. Their environments helped shape the direction of change, but the traits inherited from the founding pairs influenced the populations’ starting points and continued to matter years later.

The findings do not suggest that evolution is random in every respect. Natural selection remains a powerful force, and advantageous traits can become more common over generations. Instead, the experiment showed that selection works with the variation available in a population—and that variation can be determined by chance.

The hurricane that destroyed the original lizard populations therefore became the beginning of an unusual scientific story.

What appeared to be an ecological disaster created a rare chance to observe a process that usually remains hidden in the distant past. Fourteen lizards established seven new populations, and those populations became a living record of how random beginnings and environmental pressures can work together to shape evolution.

Frequently Asked Questions

1. What was the founder effect in the Bahamas lizard experiment?

The founder effect occurred when each new island population was established by only one male and one female brown anole. Because those two lizards carried only a small sample of the genetic variation in the larger source population, their offspring could inherit traits that were not representative of the original group.

2. How many lizards were released on the seven islands?

Scientists released 14 brown anoles in total—one male and one female on each of the seven islands—in 2005.

3. What did the lizard experiment reveal about natural selection?

The study showed that natural selection gradually favoured shorter hindlimbs in the island populations, which suited the narrower vegetation. However, populations that began with longer or shorter legs often retained those differences years later, demonstrating that founder effects could persist alongside adaptation.

4. Why was this experiment important for evolutionary biology?

It allowed scientists to observe the development of new populations from known founding individuals. This made it possible to study the founder effect and natural selection together in a way that is rarely possible with populations established long before scientific observation began.