A major challenge for the toxicologist involved in safety evaluation of chemical mixtures is to test the hypothesis that as a rule exposure to mixtures of chemicals at (low) non-toxic doses of the individual chemicals is of no health concern. A series of repeated dose studies in rats with defined mixtures of chemicals with the same or different target organs revealed that exposure to a combination of chemicals compared with exposure to the individual compounds did not constitute an evidently increased hazard, provided each chemical was administered at a level similar to, or slightly lower than, its own 'No-Observed-Adverse-Effect-Level'. The results of subacute oral toxicity studies in rats with defined mixtures of nephrotoxicants with similar mode of action underlined the applicability of the additivity assumption for a mixture of chemicals with simple similar action. Safety evaluation of complex chemical mixtures is a challenge that can be tackled as follows: first, identify the (e.g. ten) most risky chemicals in the mixture, and, second, assess the hazard and the potential health risk of the mixture of the most risky chemicals, using procedures developed for defined mixtures. To identify interactions between individual compounds, a most promising testing strategy appeared to be a statistical approach using a fractional two-level factorial design. A challenge for today and the future is to gradually substitute mixture-oriented (real life-oriented) standard setting for (unrealistic) single chemical-oriented standard setting.