When grown under optimum conditions, plant species from fertile, productive habitats tend to have inherently higher relative growth rates (RGR) than species from less favourable environments. Under these conditions, fast-growing species produce relatively more leaf area and less root mass, which greatly contributes to their larger carbon gain per unit plant weight. They have a higher rate of photosynthesis per unit leaf dry weight and per unit leaf nitrogen, but not necessarily per unit leaf area, due to their higher leaf area per unit leaf weight. Fast-growing species also have higher respiration rates per unit organ weight, due to demands of a higher RGR and higher rate of nutrient uptake. However, expressed as a fraction of the total amount of carbon fixed per day, they use less in respiration. Fast-growing species have a greater capacity to acquire nutrients, which is likely to be a consequence, rather than the cause, of their higher RGR. There is no evidence that slow-growing species have a special ability to acquire nutrients from dilute solutions, but they may have special mechanisms to release nutrients when these are sparingly soluble. We have analysed variation in morphological, physiological, chemical and allocation characteristics underlying variation in RGR, to arrive at an appraisal of its ecological significance. When grown under optimum conditions, fast-growing species contain higher concentrations of organic nitrogen and minerals. The lower specific leaf area (SLA) of slow-growing species is at least partly due to the relatively high concentration of cell-wall material and quantitative secondary compounds, which may protect against detrimental abiotic and biotic factors. As a consequence of a greater investment in protective compounds or structures, the rate of photosynthesis per unit leaf dry weight is less, but leaf longevity is increased. In short-term experiments with a limiting nutrient availability the RGR of all species is reduced, but potentially fast-growing species still grow faster than inherently slow-growing ones. Therefore, the absence of fast-growing species from infertile environments cannot be explained by their growth rate per se. The higher leaf longevity diminishes nutrient losses and is a factor contributing to the success in nutrient-limited habitats. We postulate that natural selection for traits which are advantageous under nutrient-limited conditions has led to the low growth potential of species from infertile and some other unfavourable habitats. Other examples indicating that selection for traits which allow successful performance under adverse conditions inevitably leads to a lower potential RGR are included. We conclude that it is likely that there are trade-offs between growth potential and performance under adverse conditions, but that current ecophysiological information explaining variation in RGR is too limited to support this contention quantitatively.