The use of plant phenology for determining the timing of management practices is poorly understood in species-rich grasslands. The objectives were to assess the effect of management practices on the dates at which phenological stages occur and to compare different methods of calculating the growing degreedays in order to predict them. Dates at which phenological stages of plant species occurred were recorded and plant strategy for resource capture and use was assessed through measurements of the dry matter content of leaves in two experiments in regions with contrasting climatic conditions. In Experiment 1, conducted near Toulouse, France, a set of 31 species was sown in pure stands at two levels of N availability. In Experiment 2, a network of 18 permanent grasslands, located in the French Pyrenees, was studied. In both experiments, the dry matter content of leaves, and flowering and ripening times, were measured. In Experiment 1, the dates on which a given phenological stage occurred were correlated with one another, and the grass species showed a significant ranking of dry weight of leaves for these dates. In Experiment 2, the difference between average flowering times of plant communities was shown to be around 40 d and resulted more from the species composition of the plant community than from their sensitivity to management practices. Plant communities were significantly ranked by dry weight of leaves for their flowering time. The minimum and maximum base temperatures which minimized the growing degree-days between the two locations were 0 and 25°C, respectively, and the most appropriate date from which to start to accumulate temperatures was found to be 1 February.