Studies were conducted from 1993 to 1995 in Southern Spain to determine the feasibility of controlling broomrape (Orobanche crenata Forsk.) in broad bean (Vicia faba L.) and lentil (Lens culinaris L.) by treating seeds with imazethapyr and imazapyr. In the broad bean, soaking for 5 min in 0·01–0·1% herbicide solutions or coating at 20–40 g ha−1 (seed sowing rate 160 kg ha−1) with imazethapyr (Pursuit-10) did not affect seed germination and crop growth, and resulted in 60–80% broomrape control. Furthermore, broad bean seeds treated with imazethapyr followed by an additional late post-emergence application of imazapyr (Arsenal-25) at 5 g ha−1 resulted in excellent broomrape control (>95%). Similarly, lentil seed treatments with imazapyr by coating seeds at rates equivalent to 5–10 g ha−1 or by soaking for 5 min in 0·25% solutions did not affect germination or crop growth, and controlled 85–95% of broomrape. As a result, with broomrape-efficient herbicide treatments, crop biomass/seed yield increased as compared to broomrape-infested, non-treated controls. Herbicide seed treatments with imazapyr in broad bean and with imazethapyr in lentil were less well tolerated and were less effective in controlling broomrape than treatments with imazethapyr and imazapyr, respectively.