The objective of this study was to evaluate wool (Dorset and Rambouillet) and hair (Dorper, Katahdin, and White Dorper) breeds for their ability to complement Romanov germplasm in an annual fall lambing system by estimating direct maternal grandsire and sire breed effects on economically important lamb and ewe traits. After 3 yr of evaluation under spring lambing, ewes of the five F1 types were transitioned to spring mating, exposed to composite terminal sires, and evaluated under a barn lambing system at 4, 5, and 6 yr of age. A total of 527 first generation crossbred (F1) ewes produced 1,151 litters and 2,248 lambs from 1,378 May exposures. After accounting for differences in dam age, birth type, and sex, lamb survival to weaning was unaffected by maternal grandsire breed (P = 0.30). However, lambs born to 50% Dorset (16.8 ± 0.21 kg) or 50% White Dorper ewes (16.8 ± 0.28 kg) were heavier at weaning than those born to 50% Katahdin dams (13.8 ± 0.32 kg; P < 0.001). Additionally, lambs born to 50% Dorset ewes were heavier than those born to 50% Rambouillet (16.0 ± 0.22 kg) and 50% Dorper ewes (15.7 ± 0.33; P ≤ 0.03), but no other pairwise maternal grandsire breed differences were observed (P ≥ 0.06). Ewe body weight (n = 3,629) was recorded prior to each of six possible mating seasons and, across ages, was greatest for Dorset- and Rambouillet-sired ewes (56.7 ± 0.44 and 56.5 ± 0.45 kg, respectively), intermediate for Dorper- and White Dorper-sired ewes (54.7 ± 0.78 and 54.1 ± 0.64 kg, respectively), and least for Katahdin-sired ewes (51.5 ± 0.45 kg). Fertility after spring mating (0.80 ± 0.03 to 0.87 ± 0.02), litter size at birth (1.46 ± 0.09 to 1.71 ± 0.07), and litter size at weaning (1.25 ± 0.06 to 1.46 ± 0.06) were not impacted by sire breed (P ≥ 0.16). Ewe longevity, assessed as the probability of being present after 6 production years, was also not affected by sire breed (0.39 ± 0.03 to 0.47 ± 0.03; P = 0.44). Rambouillet-sired ewes weaned more total weight of lamb (21.5 ± 0.94 kg) than Katahdin-sired ewes (17.8 ± 0.94 kg; P = 0.05), but no other sire breed differences were detected (P ≥ 0.07). Results demonstrated that incorporating the Romanov into a crossbreeding system is a practical means of improving out-of-season ewe productivity.