A thorough review of the literature revealed no published repeated-dose oral developmental toxicity studies of inorganic arsenic in rats. In the present study, which was conducted according to modern regulatory guidelines, arsenic trioxide was administered orally beginning 14 days prior to mating and continuing through mating and gestation until gestational day 19. Exposures began prior to mating in an attempt to achieve a steady state of arsenic in the bloodstream of dams prior to embryo-foetal development. Groups of 25 Crl:CD(SD)BR female rats received doses of 0, 1, 2.5, 5 or 10mg/kg/day by gavage. The selection of these dose levels was based on a preliminary range-finding study, in which excessive post-implantation loss and markedly decreased foetal weight occurred at doses of 15 mg/kg/day and maternal deaths occurred at higher doses. Maternal toxicity in the 10mg/kg/day group was evidenced by decreased food consumption and decreased net body weight gain during gestation, increased liver and kidney weights, and stomach abnormalities (adhesions and eroded areas). Transient decreases in food consumption in the 5mg/kg/day group caused the maternal no-observed-adverse-effect level (NOAEL) to be determined as 2. 5mg/kg/day. Intrauterine parameters were unaffected by arsenic trioxide. No treatment-related foetal malformations were noted in any dose group. Increased skeletal variations at 10mg/kg/day were attributed to reduced foetal weight at that dose level. The developmental NOAEL was thus 5mg/kg/day. Based on this study, orally administered arsenic trioxide cannot be considered to be a selective developmental toxicant (i.e. it is not more toxic to the conceptus than to the maternal organism), nor does it exhibit any propensity to cause neural tube defects, even at maternally toxic dose levels.