Different plant species, ecotypes, varieties or cultivars can be selected, tailored and harnessed for the remediation of polluted soils, sites and brownfields, as well as for the phytotreatment of domestic or industrial wastewater. Several approaches are possible, but phytoremediation does always rely on in situ extraction, detoxification or immobilization of xenobiotic chemicals or trace elements. Phytoremediation has a strong potential as a natural, solar-energy driven approach for soils and sites moderately polluted over large surfaces, if the most appropriate plants have been carefully selected and the adequate agronomic methods are applied to manage the phytoavailability of contaminants. It can already offer a sustainable and less expensive alternative to soils or sites restoration, as compared to traditional physico-chemical techniques, even if the time required reaching the target end-points is often a severe drawback. Depending on the type of soil to be treated, as well as on the pollutants, their concentration and ageing, different approaches can be considered and will be presented with some examples of successful applications, but also highlighting their limitations: • Phytoextraction - uptake of trace elements into roots, then translocation into shoots, followed by the harvest and destruction of the contaminated plants, with possible recycling strategies to recover metals from biomass or ash. Two types of plants can be used: plants able to hyperaccumulate one or several trace elements, but producing low biomass; or plants accumulating moderate amounts of metals or metalloids, but producing high biomass. • Phytostabilisation - based on the immobilization of inorganic contaminants by adsorption to plant roots or soil particles, or precipitation in the root area, thus preventing their migration in soil, groundwater or air, and decreasing erosion, runoff and leaching. For large contaminated areas phytostabilisation is probably the only reasonable option to restore ecosystems. The most effective approach is the use of indigenous plant species, after the addition of appropriate soil amendments. • Phytodegradation - exploiting the huge potential and biodiversity of plant metabolism to detoxify, transform or degrade xenobiotic organic chemicals, like pesticides, explosives or dyes. • Phytostimulation - enhanced microbial metabolism by root exudates; plant/microbial interactions are important for such rhizospheric processes, especially for very hydrophobic compounds. In spite of a growing track record of commercial success, more large-scale trials are needed to definitively prove that phytoremediation is effective and applicable in many different situations, to rigorously assess its economics and to convince stakeholders and local authorities that it is a reality, not a dream. However, to be recognized as a mature and efficient technology, phytoremediation still needs significant progress, even if it is the focus of intensive research and development, from numerous laboratory basic studies and greenhouse experiments to successful demonstration sites. More fundamental research is still required to better understand and control the numerous metabolic processes occurring in the plant and its different organs, as well as the complex interactions between pollutants, soil, plant roots, and the rhizospheric and endophytic microorganisms. However, experimental conditions can modify plant physiology and biochemistry, and it is not obvious if results obtained with young and small plantlets under well-defined laboratory conditions can be extended to real full-scale fields. If the small-scale experimentation in laboratory or in greenhouse is highly valuable to better understand the basic plant molecular and biochemical mechanisms involved in the uptake, translocation, detoxification and sequestration of toxic trace elements and xenobiotic compounds, their direct transposition in the field is somewhat questionable. On the other hand, short-term experimental conditions in small pots used to cultivate plantlets do not allow the normal growth and development of roots and shoots, and thus impose stress which can affect the performance of plants and the efficiency of phytoremediation. In the field the following factors do play a major role, but they cannot be assessed under laboratory conditions: soil heterogeneity, meteorological uncertainty, total concentrations and spatial variability of contaminants, texture and depth of soil layers, nutrient and water availability, problems with the homogeneous input of amendments, differences in rooting depth and density, variability of soil moisture, impacts of pests, pathogens and herbivores. Molecular and biochemical mechanisms of plant adaptation to real stress conditions are still poorly understood and signalling pathways involved remain unclear. However, adaptation to environmental changes is crucial for plants and thus for the efficiency of phytoremediation. Most studies deal either with organic xenobiotics or with trace elements, whereas in the field, a mixed pollution is generally observed. Information on the probable antagonistic or possible synergistic effects of mixed pollutants on their mutual accumulation and detoxification remains however scarce and further studies are urgently needed to better understand the mechanisms involved and improve phytoremediation efficiency. On the other hand, the bioavailable fraction of soil contaminants, but not the total concentration, should be considered as the most important one from an ecological, toxicological and health standpoint, and is determined by the chemical properties of the pollutant, soil characteristics, ageing processes and biota behaviour. Ageing of pollutants usually decreases bioavailability, but root exudates, root-induced rhizosphere changes, mycorrhizal fungi and rhizospheric bacteria play a major role in the dynamics and the ability of pollutants to enter into plants; spreading of pollutants to the surroundings and groundwater can also be affected. The bioavailability of contaminants and their uptake by crops are also essential parameters for establishing risk-based regulatory guidelines and enhancing food safety. Finally, the economics of phytoremediation of organic pollutants is generally favourable, but cost is an acute problem for the treatment of trace elements. Ideally, plants should produce biomass with added value or should be used to recover valuable products. The fate of plant biomass after harvest should thus be considered in the context of biorefineries. Lignocellulosic feedstock can be used for fibers or biofuel production, after extraction of added-value products like oil for lubricants, fragrances or fine chemicals. Incineration of plant material should allow the selective recovery of metals from ash after energy production, thus providing an economically sound process, depending on the type and concentration of metal, as well as on the market price of the trace elements under consideration. With such approaches, marginal contaminated soils or brownfields can be cleaned-up, whereas owners are offered valuable sources of income instead of set-aside scenarios. In such a situation, time is no longer a limiting factor, and the term phytomanagement should be more appropriate than phytoremediation