We explored migration patterns in Great bustards (Otis tarda), a species that shows strong sexual selection and the most extreme sexual size dimorphism among birds. The aim was to explain differential migration, examining whether Great bustards fulfil the main predictions of bird migration theory hypotheses and sexual segregation theory hypotheses. We radio-tracked the seasonal movements of 65 males and 68 females in central Spain. We found four main sexual differences. First, the proportion of migratory males was higher than that of migratory females (86% vs. 51%). Second, males abandoned the leks immediately after the mating season (late May to early Jun.), whereas females remained there for another 3–7 mo. Third, 54% of the migratory males used two different post-breeding areas, the first located northwards at 82 km from the breeding sites in summer, and the second southwards at 50 km in autumn–winter. Migratory females used only one area in autumn–winter which coincided geographically with that of males. And fourth, males returned to the breeding areas earlier (between Sep. and Mar.) than females (between Jan. and Apr.). These results show that the Great bustard is a differential migrant by sex in central Spain and support the weather sensitivity hypothesis (males were less tolerant to summer heat) and the specialization hypothesis (exclusive maternal care of the brood by females). Sexual differences in migratory behaviour are probably ultimately determined by the strong sexual selection in this species.