Standardised procedures have been used to measure 67 traits in 43 common plants of the British flora. This paper provides an interpretation of the most consistent patterns in the resulting matrix by means of correlation, ordination and classification analyses. Only a weak coupling was observed between attributes of the regenerative and established phases of the life history. However, within each phase, attributes were strongly aggregated into sets and a high proportion of the variation between species coincided with a single axis. Attributes of the established phase displayed remarkably consistent trends, with a strong 'Axis I' being identified by three different multivariate methods. There was a marked correlation between foliar concentrations of N, P, K, Ca and Mg, high concentrations of which coincided with the capacity for rapid growth in productive conditions and an inability to sustain yield under limiting supplies of nutrients. A diverse array of other traits, less immediately involving mineral nutrients, were also entrained in Axis I; these included life history, root and shoot foraging, the morphology, longevity, tensile strength and palatability of leaves, and the decomposition rate of leaf litter. This pattern occurred in both monocotyledons and dicotyledons and appeared to reflect a tradeoff between attributes conferring an ability for high rates of resource acquisition in productive habitats and those responsible for retention of resource capital in unproductive conditions. The second axis of variation evident in the established phase was related to phylogeny and distinguished between monocotyledons and dicotyledons on the basis of a diverse set of traits including genome size, cell size, root and shoot foraging characteristics and vascular tissues. A third axis was detected in which ephemerals and perennials were separated by differences in attributes such as breeding system, leaf decomposition rate and a set of traits reflecting the small stature of many short-lived plants. In the regenerative phase, the leading axis was clearly related to the widely recognized tradeoff between seed size and seed number and was consistent with current understanding of seed banks, and with modern theories explaining species coexistence in terms of complementary responses to temporal and spatial variation in vegetation gap dynamics. The data provide strong evidence of functional integration between evolutionary specializations in root and shoot and support Donald's unified theory of competitive ability. The data are not consistent with theories of functional types based upon evolutionary tradeoffs in allocation between root and shoot. We suggest that the evidence assembled here and elsewhere in the current literature points to the existence of primary functional types, including those recognized by Ramenskii and Grime. These functional types can be reconciled with the individuality of plant ecologies in the field and provide an effective basis for interpretation and prediction at various scales from the plant community to regional floras. There are particular opportunities for prediction of successional trajectories, the role of herbivores in vegetation succession and the response of vegetation to eutrophication and extreme climatic events. It is also suggested that aspects of this investigation may provide a Darwinian underpinning for Odum's theory of ecosystem maturation.