Are mixed populations of plant species more productive than pure stands?

It has been conjectured that species mixtures ought to be more productive than corresponding pure stands, but experimental evidence for this has not been strong. Data extracted from 54 published experiments involving binary species associations are used to test the conjecture. Mixtures and pure stands are compared on the basis of relative land output: the ratio of yields generated by equivalent populations per species occupying the same total land area and growing under similar conditions. The comparison is made directly using observed biomass yields as well as with yields estimated from yield-density relationships. Mixtures prove to be significantly more productive than pure stands in 38 experiments, and are significantly loser in 8 experiments. On average, mixtures are 12% more productive than pure stands, based on 202 direct observations, or 13% more productive, based on 604 estimates using yield-density relationships.