In 1990, a KARI/CIMMYT collaborative Striga control research project was established in western Kenya with funds provided by CIDA. Much of the project's initial research was directed toward adapting previously reported Striga control technologies. Although many of these management practices gave some Striga control, they were rarely cost effective or failed to give sufficient immediate control to make them attractive to farmers. In 1991, experiments directed toward developing longer term Striga control strategies were established. Factors investigated included hand-pulling before seed-set, soil fertility, stover management and trap cropping. Four seasons of hand-pulling Striga were required in order to diminish Striga seed banks to the point that Striga related yield losses were minimal. Over seasons, soil fertility and incorporation of maize stover positively reduced the level of Striga, particularly when combined with hand-pulling. There was little effect from 3 or 4 seasons of trap cropping with cowpeas or cotton compared to continuous maize on Striga seed numbers in the soil. A large reduction in the Striga seed bank that was not associated with any treatment effect was observed after 3 seasons of continuous cropping at 2 of the 3 locations. This observation needs further investigation. The primary points of a long term strategy are to maintain inputs of organic material and fertilizer, combined with hand-pulling before seed maturation.