The study investigated the contribution of information leaflets to advice and practice in infant care within the wider context of parent learning. Two leaflets about reducing possible risks of cot death were of particular interest. The practices selected for study were infant feeding, sleep position and warmth assessment choices. As well the clothing, bedding and heating arrangements of babies in bed at night were described and compared with practices of Christchurch parents in 1986. Ninety seven maternity patients at Christchurch Women's Hospital were interviewed in hospital and 53 of these, who were new mothers, were interviewed again at home when their babies were about three months old. The hospital interviews gathered information about promoted behaviours, sources from which written and verbal advice were received, women's perceptions of written and verbal advice received and their intended choices for infant feeding, sleep position and warmth assessment. The home interviews gathered information about written and verbal advice sought and received since the baby's birth, women's perceptions of this advice and of their learning experiences, actual infant feeding, sleep position and warmth assessment choices. Evidence of behaviours was obtained by three consecutive days of diary recording for infant feeding and sleep position, and by observation for infant warmth assessment. Details of clothing, bedding and heating arrangement for babies in bed at night were also gathered. Promoted behaviours were breastfeeding , the side sleeping position, and to a lesser degree, feeling the torso to assess an infant's warmth. These behaviours were consistent with those promoted in the cot deathrelated leaflets. Doctors in particular, and health professionals in general were women's most trusted sources for advice about infant care. Written information reported to have been seen was generally reported to have been read by hospital subjects and readers were representative of the total sample. Materials in the Plunket information pack were well read and more than half the home subjects reported reading most of the materials. The cot death related leaflets were amongst those most often reported to have been read (by more than three quarters of subjects), perceived to have been helpful and thought worth recommending to other parents. Hospital subjects' intended choices of infant care were breastfeeding for 85, the side sleeping position for 89, and feeling the torso to assess infant warmth for 34 subjects. By the time their babies were three months old a significant number of home subjects had changed from their intended practices. Twenty five women had changed from fully breastfeeding (chi²=26.0, df 1, p<O.OO5), 16 from intending the side sleep position only to including the back (chi²=21.0, df 1, p<O.005) and 31 subjects changed to feeling the torso to assess infant warmth (chi²=40, df 1, p<O.005). Knowing that the prone sleep position was a risk factor for cot death was the reason given by 17 of 19 experienced mothers for changing from choosing the prone sleep position for their last baby to choosing the side or back for the new baby. There were significant changes in clothing and bedding practices for home subjects compared to practices in 1986 with 1990 parents more likely than 1986 parents to have used 6 or less layers of total coverings on their infants in bed at night (chi²=20.9, df 1, p<O.OO5). Readers of the cot death related leaflets were representative of the total hospital sample and were more likely than non-readers to have had an intended warmth assessment method (chi²=9.3, df 1, p<O.OO5). Actual infant sleep position and thermal care practices of readers of the two leaflets were consistent with those promoted in the leaflets. The study's findings suggest that information leaflets have made a useful contribution to advice and practice in infant care. In this way leaflets have the potential to contribute to reducing infant deaths and they deserve an important place in education programmes to improve infant care and prevent cot deaths.