The induction of X-chromosome malsegregation, sex-linked recessive lethals and II–III autosomal translocations by aniline·HCl was investigated in Drosophila melanogaster. Nondisjunction was tested in 2 and 4 d old virgin females fed on aniline·HCl solutions (3, 5, 10 and l5%) using a system where exceptional females (XXY) and only 1/4 of the expected regular progeny are viable. After mating, the females were subcultured daily. Similarly treated 7-day-old wild-type males were used to run classical II–III translocation and recessive lethal tests; for the latter, the solutions were also injected intraabdominally. In all cases, five broods were obtained. A direct correlation was observed between concentration and toxicity. Furthermore, males were more sensitive than females, and the latter's sensitivity was higher at 4-day-old than at 2-day-old. This could be attributed to a decrease with age in the efficiency of a detoxifying mechanism, or to the generation of a toxic metabolite in older flies. Significant increases in nondisjunction were observed with 5, 10 and 15% solutions suggesting the existence of a threshold. No dose effect was detected within the range of the effective concentrations used. The increases were observed in the first subculture (representing mostly stage 14 oocyte, i.e., cells in metaphase I) and in the third subculture, representing cells in which the spindle has not yet formed, thereby pointing to a direct effect of the chemical on the chromosomes and not on the spindle. It is proposed that the second sensitivity peak detected might be the outcome of the transient loss of a protective configuration provided by the karyosome, due to its expansion in stages 9 and 10 of the developing oocytes. No sex-linked lethals or translocations were induced.