A bedside technique for evaluating the behavioural response of healthy neonates to pain was assessed. Thirty six term infants (median gestational age 40 weeks; median postnatal age 4 days) and 31 preterm infants (median gestational age 34 weeks; median postnatal age 4 days) were assessed at the cotside for their response to heel preparation and heel lance for routine blood sampling. The facial actions of brow bulge, eye squeeze, nasolabial furrow, and open mouth were noted, and also the presence or absence of crying. Thirty five (97%) term and 26 (84%) preterm infants showed an increase in the number of behaviours in response to heel lance. Brow bulge and nasolabial furrow were seen most often, and occurred more often than crying in the two groups. There was good interobserver agreement (94%). The consistency of response and the high degree of interobserver agreement makes this method of behavioural assessment of acute pain of use in healthy neonates.