David Quammen · 1996

The Song of the Dodo

Island Biogeography in an Age of Extinctions

Rough AI truth score

83/100

Quammen accurately explains why islands and habitat fragments illuminate evolution and extinction. Species-area logic remains indispensable, but simple island analogies and reserve-size formulas cannot by themselves predict every fragmented ecosystem or conservation outcome.

Based on three central claims · medium-high confidence

The three claims

01Supported

Isolation, area, colonization, and extinction make islands natural laboratories for evolution and biogeography.

Island radiations, endemism, turnover, and extinction repeatedly demonstrate how geography structures lineages and communities. The framework remains foundational even as contemporary models incorporate traits, history, matrix habitat, and changing climate.

The Song of the Dodo review Species-area relationships and extinction forecasts

02Mostly supported

Habitat fragments often function like ecological islands, increasing extinction risk as area shrinks and isolation grows.

Fragmentation commonly reduces populations, connectivity, and recolonization while amplifying edge effects. Matrix quality, configuration, mobility, time lags, and species differences mean real fragments do not behave exactly like oceanic islands.

Species-area relationships and extinction forecasts The Song of the Dodo review

03Mostly supported

Species-area relationships permit straightforward numerical predictions of how many species a given habitat loss will eliminate.

The relationship is robust as a broad empirical pattern and useful for comparative risk, but direct extinction estimates depend on baselines, debt, habitat specificity, sampling, and the form of the model. Counts are not automatic forecasts.

Species-area relationships and extinction forecasts The Song of the Dodo review

Other claims worth checking

Endemic island species are especially vulnerable to introduced predators and habitat change.
Conservation biology must preserve processes and connectivity, not only isolated specimens.

What this number means. It is an AI-generated first-pass judgment of three central factual or causal claims—not a rating, exhaustive fact-check, or human peer review. Claim credits are 100% for supported, 75% for mostly supported, 50% for mixed, and 25% for weak, then averaged and rounded. Lower confidence means the score should move more as better evidence arrives.

Method three-central-claims/0.1.0 · checked 2026-09-01 · 3/3 selected claims assessed · method and source audit