In a new paper just published in Proceedings of the Royal Society B, we solve a classic ecological conundrum posed 50 years ago: the small-island effect. MacArthur and Wilson’s classic theory of island biogeography predicts that the number of species on an island should be positively related to island area. This prediction is generally borne out by data, but it is violated for small islands, where there is no apparent relationship between island area and species richness. In the new paper, we explain this “small-island effect” as arising from a niche-structured regime that prevails when immigration is very low—a situation commonly found on small islands. We verify our theory by showing that a corresponding mathematical model gives excellent fits to 100 archipelago data sets from around the world, and by confirming key predictions about when the small-island effect should be more prominent.