The influence of the maternal genotype (Canton-S, proficient in the repair of X-ray-induced chromosome breaks and ebony, less proficient in this regard) on the recovery of X-ray-induced autosomal (II-III) translocations and ring-X chromosome losses in mature spermatozoa was studied. In the first series of experiments, males carrying appropriate markers on their second and third chromosomes were irradiated and mated to Canton-S or ebony females and the frequencies of II-III translocations were determined. In the second series of experiments, males carrying ring-X chromosomes were irradiated in N2 or in O2, mated to Canton-S or ebony females and the frequencies of XO males were determined; additionally, under similar gas-treatment and radiation conditions, the pattern of egg-mortality was also assessed. The data on translocations show that the yields are higher with ebony than with Canton-S females; these and earlier results on dominant lethals and sex-linked recessive lethals support the interpretation that the maternal repair system in the ebony strain is less proficient and more error-prone than that of the Canton-S strain. Those on the losses of ring-X chromosomes demonstrate that (i) the absolute yields of XO males are lower with ebony than with Canton-S females irrespective of whether the parental males are irradiated in N2 or in O2; (ii) the exposure-frequency relationships are all linear, but the slopes are higher when the males are irradiated in O2 and are consistent with an oxygen-enhancement-ratio of about 1.5 and (iii) the relationships between the logarithm of egg-survival and XO male frequency are also linear, but the slopes for the O2 groups are lower than those for the N2 groups (slope ratios of 0.86-0.87). The finding that at given survival levels, the XO frequencies are lower in the O2 than in the N2 groups of both the Canton-S and ebony series viewed in the context of the mechanisms that have been postulated to explain the loss of ring-X chromosomes in irradiated mature spermatozoa permits the following interpretation for the observed results: (i) a higher proportion of potential XO zygotes is lost through dominant lethality in the O2 groups than in the N2 ones presumably because the chromosome breaks induced in O2 are qualitatively different in the sense that they have a higher probability to undergo reunions relative to restitution, compared with breaks induced under anoxia and (ii) this leads to lower than expected oxygen-enhancement ratios (i.e., expected on the basis of published data on sex-linked recessive lethals, another kind of genetic damage which shows a linear exposure-frequency relationship.