The influence of haemodilution with colloids on somatosensory evoked potentials in non-premedicated volunteers is reported. In seven volunteers (randomized crossover design), blood (20 mL kg−1 within 30 min) was removed and simultaneously replaced by gelatin 3% or hydroxyethylstarch 6%. After 30 min, blood was retransfused within 30 min. Median and posterior tibial nerve somatosensory evoked potentials were recorded from the cortex, second cervical vertebra, Erb's point and 1st lumbar vertebra, respectively. One volunteer experienced a severe allergic reaction to gelatin, therefore only six gelatin trials were evaluated. Haemodilution decreased the haematocrit from 39.8 ± 1.6% (mean ± SD) to 31.1 ± 2.0% (gelatin) and from 40.7 ± 1.7% to 29.8 ± 1.5% (hydroxyethylstarch), respectively. Retransfusion increased haematocrit to 34.4 ± 0.9% (gelatin) and to 34.2 ± 1.3% (hydroxyethylstarch). Neither haemodilution with gelatin nor haemodilution with hydroxyethylstarch or retransfusion influenced evoked potentials. In conclusion, the treatment of blood loss up to 30% of estimated blood volume with gelatin or hydroxyethylstarch will not affect somatosensory evoked potential monitoring provided normovolaemic conditions are maintained.