Muller-5 males of Drosophila melanogaster were irradiated in N 2 or O 2 and mated to excision repair deficient, post-replication repair deficient ( mei-9 a, mei-41 D5, mus101 D1, mus201 D1, mus302 D1, mus306 D1 and mus308 D2 ) or repair proficient females. The surviving fraction (dominant lethality) was estimated in the F 1 and used to reassess existing recessive lethal and translocation data. The surviving fraction was found to decrease if repair deficient females were used (maternal effect). The dose-effect curves are often biphasic with a steeper slope at low doses than at high ( ≧5 Gy ) doses of X-rays. The high dose part of the curve is sensitive to oxygenation during irradiation and is affected significantly by the mutants with low fertility ( mei-9, mus101 and mus302). The low dose component is not sensitive to oxygenation during irradiation and seems influenced by all seven repair deficient mutants. The sensitivity of the high dose part to oxygenation suggests that this part is related mainly to DNA break damage, while in the low dose part base damage seems more important. Existing recessive lethal and translocation data were plotted against the surviving fraction for a reassessment. In excision repair deficient mutants translocation induction is lower compared to repair proficient flies at the same level of survival (i.e., dominant lethality). Likewise in post-replication repair deficient mutants induction of recessive lethals is decreased. However the frequency of respectively induced recessive lethals and translocations obtained at the same level of X-rays was the same in repair deficient and proficient backgrounds. It is concluded that genetic damage recovered in a repair deficient background is likely to be qualitatively different even if the frequency of the damage induced by a given dose is not altered.