To evaluate the effectiveness of alternative options for biocontrol of whiteflies in greenhouse tomatoes, an experiment was carried out during the cropping season 2005-2006 in one of Sardinia's major horticultural districts (S. Margherita di Pula, Cagliari, Italy). Twelve long-cycle and 17 short-cycle tomato crops (8 autumn and 9 spring crops) were surveyed. All of them were treated for insect pest control at the beginning of the growing season, but in 19 out of 29 cases whitefly natural enemies were also released (BCA greenhouses), at least four weeks after the last treatment. The following release programmes were tested: on autumn crops, 1 Macrolophus caliginosus and 12 Eretmocerus mundus/m2; on long-cycle crops, 1 M. caliginosus (released in autumn or spring) and 24 Encarsia formosa/m2 or 48 E. formosa/m2; on spring crops, 1 M. caliginosus and 24 E. formosa/m2 or 48 E. formosa/m2. The cost of each option was fixed at approximately 0.25 Euros/m2. The remaining greenhouses were maintained as controls (no BCA greenhouses). While whitefly and mirid populations were monitored monthly, whitefly species composition and mortality of immature stages were estimated at least twice during the growing season. On short-cycle autumn crops, the release of M. caliginosus and E. mundus produced negligible results in terms of Bemisia tabaci control. On long-cycle and spring crops, even though in June mortality rates in BCA greenhouses were found to be 2- to 3-fold higher than in no-BCA greenhouses, Trialeurodes vaporariorum population growth was not significantly affected by natural enemies. Among the beneficials tested, E. formosa proved to be the most effective; E. mundus and M. caliginosus did not establish well, probably owing to the persistence of insecticide residues, scarce prey availability and intense plant de-leafing. The presence of indigenous natural enemies of whiteflies was observed in most sites, but in general they contributed little to biological control. The present experiment showed that in Sardinian tomato greenhouses the use of beneficial insects may result in inadequate biocontrol of whiteflies. In particular, the application of cultural practices which may disrupt the establishment and development of predator and parasitoid populations should be critically reexamined.