A dose response study was carried out with pigs in order to examine the effects of increasing dietary deoxynivalenol (DON)-concentrations on performance, clinical serum characteristics, nutrient digestibility and DON-metabolism. For this purpose, wheat contaminated naturally with Fusarium toxins was incorporated into pig diets at increasing proportions to give calculated dietary DON-concentrations of 0, 2.3 and 4.6 mg/kg during the starter period of phase 1 (14 d) of the experiment, and 0/0, 1.2/1.4, 2.3/3.7 mg/kg starter/grower diet during phase 3 (56 d) of the experiment. Each diet was tested on 16 pigs of both sexes with an initial average live weight of approximately 28 kg. A recovery phase (phase 2, 21 d) was intercalated between phase 1 and 3 of the growth experiment where all groups were fed with the uncontaminated control diet since some pigs exposed to the highest dietary DON-concentration during phase 1 nearly completely refused the offered feed. Affected pigs completely recovered during this phase. In phase 3, when diets with lower DON-concentrations were fed, no differences in performance could be detected. Serum clinical characteristics (enzymes indicating liver damage, total protein, immunoglobulins) did not respond to increasing DON-concentration in the diets. DON-concentration in serum increased in a dose-response-related manner as dietary DON-concentration increased. However, this parameter was not or only weakly correlated to any of the examined performance parameters or serum characteristics. Also, nutrient digestibility of the diets and N-retention were not affected by treatments with the exception of crude fat digestibility which was not consistently influenced. Concentration of DON and its metabolite de-epoxy-DON increased in urine with increasing dietary DON-concentration in a strongly linearly related fashion. The proportion of the excretion of de-epoxy-DON of the total urinary excretion of DON plus de-epoxy-DON rose linearly up to approximately 4%. Total recovery of DON plus de-epoxy-DON as percentage of DON-intake varied between 45 and 57% and was not influenced by dietary DON-concentration. Only a very small fraction of approximately 0.1% of ingested DON was recovered in faeces.