The point of ecology is to understand the distribution and abundance of species. Neutral models of species' abundances, which suppose that these fluctuate entirely at random, are currently enjoying considerable attention. To date, much of the credibility of such models has been based on their ability to mimic observed abundance distributions. A new paper by Brian McGill shows that, at least for some data, neutral theory performs worse than does an older, simpler null model. However, this is not the end for neutral theory: rather, it is the 'end of the beginning'.