A PROBLEM FOR BIODIVERSITY-PRODUCTIVITY STUDIES - HOW TO COMPARE THE PRODUCTIVITY OF MULTISPECIFIC PLANT MIXTURES TO THAT OF MONOCULTURES

The study of the relationship between species richness of a plant comm unity and its productivity has received much attention, recently renew ed by the concern on the loss of biological diversify at a global scal e. Here, we briefly review some indices widely used in agronomic and c ompetition experiments to compare monocultures and mixtures, and compa re them to other, more recently designed ones. These various indices a re then calculated for two experiments. In the first experiment, two g rass and two legume species were grown at six levels of nitrogen avail ability, either in monocultures or in mixtures of the four species in a substitutive design; in the second experiment, five grass species we re grown at 16 levels of total nutrient availability, either in monocu ltures or in mixtures of the live species in an additive design. These data clearly show that the conclusions drawn from the experiments dep end on the index used to compare the experimental communities. We argu e that a clear test of whether the productivity of communities increas es with species richness requires that: (1) all species present in the multispecies assemblages also be grown in monocultures under the same environmental conditions, and (2) the productivity of these assemblag es be compared to the most productive monoculture. We conclude that th ere are as yet very few cases where superior productivity of multispec ies assemblages as compared to monocultures has been clearly shown.