The mass ratio hypothesis provides the key link between plant functional traits (PFTs) and ecosystem processes. Despite its centrality to the field it has had few direct tests. A litter decomposition study using grassland species, singly and in mixtures, was set up to see whether simple leaf traits could be used to predict the decomposition of leaf mixtures. Leaf Dry Matter Content (LDMC) was the trait that performed best. Mass loss in single species bags was best predicted using an exponential model. LDMC explained 48% of the variance in mixture mass loss. There was no significant impact of mixture species richness on mass loss. This study generally confirmed the predictions of the mass ratio hypothesis, but adds some support to other studies that indicate it needs broadening to take into account non-linear and threshold relationships between PFTs and ecosystem processes.