Common lichens win the lottery: how leaf-dwellers colonized the Galápagos and Cocos Island

For more than half a century, ecologists have debated whether the species that make it onto a remote oceanic island arrive by pure luck or by merit. The neutral theory of island biogeography, formulated by Robert MacArthur and E. O. Wilson in 1967, holds that every species in a source community has an equal chance of colonizing an island, with species richness set simply by immigration and extinction rates as functions of time, island size, and distance. A new study of leaf-dwelling lichens on the Galápagos and on Cocos Island puts that assumption to a direct test — and the verdict is that island colonization is largely deterministic, not a lottery.

The research, published in Frontiers of Biogeography, is the first to predict the species composition of an oceanic island biota through simulation based on a continental source metacommunity. The team, led by lichenologist Robert Lücking of the Botanical Garden Munich and Frank Bungartz of the Charles Darwin Research Station, focused on foliicolous lichens — the tiny, often brilliantly colored symbiotic organisms that colonize the living leaves of tropical plants. These lichens are ideal for the question because their island endemism is remarkably low: roughly 20 percent in the Galápagos and just 5 percent on Cocos Island. In land plants, by contrast, endemism on oceanic islands reaches 40 to 90 percent, and in terrestrial vertebrates and arthropods 50 to 98 percent, making their island floras and faunas the products of complex evolutionary histories that no simulation from a source region could predict.

The study group comprised two of the only tropical oceanic archipelagos in the eastern Pacific. Cocos Island, a single island about 550 kilometers southeast of Costa Rica and 1.9 to 2.4 million years old, is blanketed in wet tropical forest. The Galápagos, an archipelago of 14 major islands lying about 925 kilometers west of Ecuador, are dominated by dry vegetation, with humid forest confined to isolated patches on the higher mountains of islands such as Santa Cruz and San Cristóbal. The researchers compiled occurrence data from collections made between 1964 and 2014 on eight Galápagos islands, identifying 113 species, of which 23 are presumed endemic. For Cocos Island they revised earlier collections, distinguishing 100 species, five endemic. Ninety Galápagos species and 95 Cocos species are shared with the American continent, and 38 species occur on both island groups.

To model assembly, the team defined a continental source region stretching from the southeastern United States to Chile and Argentina — the area where genuine foliicolous lichen communities occur in the Americas — and assembled a list of 632 species from published inventories. For each species they scored six ecogeographic parameters considered proxies for species-specific traits: preferred vegetation type, seasonality, microsite exposure, elevational range, continental distribution range, and commonness. These scores were multiplied together, normalized, and adjusted by the sixth root to yield a combined environmental E score between 0 and 1 for each species, representing the probability of successful dispersal and establishment on an oceanic island.

The simulation itself was elegantly simple. Each species was assigned a random number between 0 and 1, multiplied by its E score, and species exceeding a threshold were pooled into simulated island assemblages of the observed size — 90 species for the Galápagos, 95 for Cocos — repeated 100 times for each island. The same procedure was run without the E scores to generate entirely stochastic, neutral assemblies. The researchers then used non-metric multidimensional scaling, a standard ordination technique in community ecology, to visualize where the observed island biotas fell relative to the clouds of simulated communities, supported by Mann-Whitney U tests comparing distance values.

The results were striking. Under the neutral null model, predicted species composition was imprecise. But when assemblages were filtered through the ecogeographic parameters, the observed Galápagos lichen biota fell near the center of a tight cluster formed by the filtered simulated subsets — a highly significant result. The outcome for Cocos Island showed a relationship with the filtered prediction as well, though the observed composition fell slightly outside the tight cluster. In other words, the pool of species from which the island biotas are assembled is not random and can be predicted with considerable accuracy, even if the exact subset of species actually observed cannot.

Among the six parameters, commonness emerged as the single most important determinant of assembly for both island groups. This makes intuitive sense: the probability of successful dispersal to an oceanic island through vectors such as air currents relates directly to a species’ abundance in the source area, which also indicates the effectiveness of its dispersal mechanisms. Rare species are less likely to be carried across hundreds of kilometers of open ocean, and rarity itself may be a consequence of inefficient dispersal. Common species also typically have broader ecological amplitudes and are better equipped to colonize areas whose conditions do not exactly match those of the source region. Notably, a previous study of the Galápagos flora found stronger effects of environmental filters than dispersal traits, but did not include commonness as a parameter — an omission this new work suggests was significant.

The two archipelagos also revealed instructive differences. For the Galápagos, the strongest effects involved parameters related to both successful dispersal and establishment, whereas for Cocos Island the strongest effects concerned dispersal alone, with microsite exposure showing no significant effect. The explanation lies in habitat availability: in the Galápagos, leaf-dwelling lichens are largely restricted to the mountain and cloud forests of the humid highlands, while the dry lowland vegetation acts as a barrier to colonization. Cocos Island, though much smaller and ecologically less diverse, encompasses a wider range of habitats suitable for these lichens, including both lowland and montane forests, so habitat preferences are less critical there. Seasonality tolerance proved equally important for both groups — a surprise, since Cocos Island’s rain forests exhibit little seasonality — suggesting that slightly different suites of species are adapted to colonize each archipelago.

Perhaps the most intriguing finding concerns elevation. The Galápagos lichen biota shows a closer affinity to continental montane cloud forest than to lowland rain forest, even though the Scalesia forests and Miconia scrub of Santa Cruz’s humid zone occur at only 300 to 700 meters, while the most similar continental site, Guajalito Reserve in Ecuador, sits at 1,800 meters. The lowland Ecuadorian site at Jatun Sacha, at 400 to 500 meters — within the elevational range of the Galápagos lichen biota — displays a quite different species composition. The authors attribute this compression of elevational belts to a lower mass elevation effect on islands, augmented by oceanic currents. Empirical comparison with continental sites in Ecuador and Costa Rica further suggests a closer affinity of the Galápagos with South America, indicating that geographic distance plays some role in assembly as well, though a test of whether the Andes act as a dispersal barrier was not supported by the available data.

The study also documented the disharmonic nature of the island biotas, a known phenomenon in island biology. Among the 260 species with E scores above the critical threshold of 0.454, certain families such as Byssolomataceae were consistently over-represented in both archipelagos, while others — particularly certain Gomphillaceae, including species of the genus Asterothyrium — were conspicuously absent despite high predicted scores. The reasons remain unknown, but the authors suggest that phorophyte composition and leaf longevity, which determine how long a leaf surface remains available for colonization, may be factors not captured by the model. The stochastic component of assembly thus consists of a small number of unpredictable species and missing species resulting from this disharmony. The broader implication is profound: even for organisms whose island biotas develop literally from scratch, colonization follows deterministic assembly rules shaped by habitat filtering and dispersal efficiency as expressed through commonness. Oceanic islands, it turns out, are not lotteries — they are selective gates, and the species that pass through them are chosen by traits that ecologists are now learning to read.

Subject of Research: Stochastic versus deterministic assembly of oceanic island biota: leaf-dwelling lichens on the Galápagos and on Cocos Island

Article Title: Stochastic versus deterministic assembly of oceanic island biota: leaf-dwelling lichens on the Galápagos and on Cocos Island

Article References: Lücking, R., Moncada, B., & Bungartz, F. (2024). Stochastic versus deterministic assembly of oceanic island biota: leaf-dwelling lichens on the Galápagos and on Cocos Island. Frontiers of Biogeography, 17, Article e133819. https://doi.org/10.21425/fob.17.133819

Image Credits: AI Generated

DOI: 10.21425/fob.17.133819

Keywords: Stochastic, versus, deterministic, assembly, oceanic, island, biota, leaf-dwelling, lichens, pagos, Cocos, scientific research