The antinociceptive effects of ethanolic extracts of Pancratium maritimum L., Narcissus tazetta subspecies tazetta and Leucojum aestivum L. bulbs have been investigated in mice using the p-benzoquinone-induced abdominal constriction and hot-plate tests. In the p-benzoquinone-induced abdominal constriction test the ethanolic extracts of P. maritimum (300, 600 or 1200 mg kg-1, s.c.) and N. tazetta subsp. tazetta (5, 50, 100 or 200 mg kg-1, s.c.) caused dose-dependent inhibition of abdominal constrictions whereas a fluctuating response was obtained from ethanolic extracts of L aestivum (2.5-500 mg kg-1, s.c.). In the hot-plate test P. maritimum and L. aestivum caused a significant increase of latency only at the highest concentrations used (1200 mg kg-1 and 500 mg kg-1, i.p., respectively). However, at these concentrations they also caused significant toxic effects. In contrast with P. maritimum and L. aestivum, N. tazetta subsp. tazetta (5-500 mg kg-1, i.p.) extracts had no antinociceptive effect in this test. These findings indicate that the antinociceptive effect of Amaryllidaceae plants differs depending on the model of nociception investigated.