The study of synchronization phenomena in ecology is important because it helps to explain interactions between population dynamics and extrinsic environmental variation. Grenfell et al. have examined synchronized fluctuations in the sizes of two populations of feral sheep which, although situated on close but isolated islands, were nevertheless strongly correlated (observed value of the population correlation, rp , 0.685). Using a nonlinear threshold model, they argue that this level of population correlation could only be explained if environmental stochasticity was correlated between the islands, with the environmental correlation, re , higher than 0.9 on average (Fig. 1a). This unusually high environmental correlation is far greater than would be predicted by the Moran effect, which states that the population correlation will equal the environmental correlation in a linear system. Grenfell et al. imply that a simple nonlinearity in population growth can mask or even destroy the Moran effect. Here we show that these surprising results are an artefact of the techniques used to measure noise correlations and synchronization.