Objective-To see whether non-professional volunteer community mothers could deliver a child development programme to disadvantaged first time mothers for children aged up to 1 year.Design-Randomised controlled trial. Setting-A regional health authority in Dublin. Subjects-262 first time mothers who were delivered during six months in 1989 and who lived in a deprived area of Dublin; 30 experienced mothers from the same community recruited as community mothers.Interventions-All the first time mothers received standard support from the public health nurse. In addition, those in the intervention group received the services of a community mother, who was scheduled to visit monthly during the first year ofthe child's life.Results-232 (89%) first time mothers completed the study-127 in the intervention group, 105 controls. At the end of the study children in the intervention group were more likely to have received all oftheir primary immunisations, to be read to, and to be read to daily, played more cognitive games; and were exposed to more nursery rhymes. They were less likely to begin cows' milkc before 26 weeks and to receive an inappropriate energy intake and inappropriate amounts of animal protein, non-animal protein, wholefoods, vegetables, fruit, and milk.Mothers in the intervention group also had a better diet than controls. At the end of the study they were less likely to be tired, feel miserable, and want to stay indoors; had more positive feelings; and were less likely to display negative feelings.Conclusion-Non-professionals can deliver a health promotion programme on child development effectively. Whether they can do so as effectively as professionals requires further study.