AbstractBackgroundInorganic arsenic has been used by many laboratories to study the pathogenesis of exencephaly in rodents. These studies, which used predominantly injection exposures, coupled with the paucity of epidemiology data, resulted in the erroneous inference that inorganic arsenic should be considered a human teratogen.MethodsThis study assembles and assesses literature analyses of older human and animal investigations together with the results of new experimental studies. These recent studies were performed according to modern regulatory guidelines, and relevant exposure routes (inhalation and ingestion) were used to evaluate the potential risk of developmental effects in humans.ResultsThe existing epidemiological data are inadequate to support risk assessment because of the failure to confirm or measure arsenic exposure during early gestation and the deficiencies in accounting for potential confounding factors. The animal data revealed that inorganic arsenic caused malformations in offspring only when it was injected into the veins or peritoneal cavity of pregnant animals during early gestation. Exposure via inhalation or oral ingestion, even at concentrations that were nearly fatal to pregnant females, caused no arsenic‐related malformations.ConclusionsInorganic arsenic poses virtually no danger to developing offspring when maternal exposure occurs by relevant routes (oral and inhalation) at concentrations that are likely to be experienced in the environment or in the workplace. Teratology 64:170–173, 2001. © 2001 Wiley‐Liss, Inc.