Large amounts of former cropland are being abandoned in developed regions. To formulate guidelines for land reclamation programmes, we explored the effects of artificial shading, irrigation, and removal of weed competi tion on the performance of Retama sphaerocarpa (L.) Boiss. seedlings in a factorial experiment located in an abandoned cropland in Central Spain. R. sphaerocarpa is of interest for revegetation because it is a drought tolerant leguminous shrub that is a major structural component of the native plant community. Seedling performance was evaluated in three ways: seedling survivorship, growth, and photochemical efficiency. We also measured soil moisture and weed biomass production and found that both increased under artificial shading conditions. Soil moisture increased very slightly where weeds were removed. Thus, increased transpiration from weeds outweighed reduced evaporation from soils due to shading by weeds. Artificial shading was the most effective treatment for seedling survivorship, followed by removal of competition by weeds. After summer, 34 % of the seedlings survived in the most favourable conditions (artificially shaded plots where weeds were removed), compared to ca. 1 % in full-light plots with no removal of weed competition. A positive effect of irrigation was found for growth of seedling cover and height in shaded plots. The analysis of photochemical efficiency pointed out the relevance of weed competition removal, and confirmed the usefulness of fast fluorescence transient techniques for the quantification of seedling performance. The data suggest that competition between seedlings and weeds was primarily for water rather than for light. We conclude that i) artificial shading improved seedling performance, but this is a little practical technique because of its cost; ii) as weeds compete with, rather than facilitate, planted seedlings, weed clipping around the seedlings is a feasible technique that would improve seedling sur vival; and iii) seedling performance could also considerably improve with a higher irrigation than was used in this experiment (751m^-2 per growth period), provided that weeds are removed.