Families which had experienced two or more unexpected infant deaths were the subject of detailed confidential enquiries, including necropsy examination. Cases were derived from two main sources: first, deaths occurring during a nationwide programme of support for families with a subsequent baby (8 families) plus 2 families from a scries of confidential enquiries in Sheffield, and second, direct referrals from paediatricians (17 families). Fifty‐seven deaths were studied. Twenty‐four families had experienced 2 and three had experienced 3 deaths: 11 deaths (19%) were found to be adequately explained by history or post‐mortem findings; 7 (12%) were probably accidental; 31 (55%) were most probably due to an action by one of the parents (filicide); only 5 (9%) were considered to be true or idiopathic sudden infant death syndrome; in 3 (5%) cases there was insufficient information to draw a conclusion. Five (18%) of the families lived in circumstances of serious social deprivation. A history of psychiatric illness was present in one or both parents in 18 (67%) of the fam