Study identifying possible correlates of child abuse, including: the stability of the child’s attachment to the abusing home; problems within the family unit; social and personal deviance of the child’s parent figures; and stresses facing the child’s mother figure. The paper asks whether seriously injured children (17%) form a particular subgroup of abused children. The sample used in the study comprised all cases (255) of established child abuse coming to the attention of the Child Welfare Department during 1967, 16% of whom died or suffered serious injury. This sample was not necessarily representative of the population of child abuse cases occurring in NZ at the time and it was suspected that the sample was biased towards the inclusion of children coming from inadequate or unstable family backgrounds. The variable that best discriminated between the seriously and non-seriously injured and abused groups was age with children under 5 being 5x more likely to be seriously injured than children over 5. Within the under-5s, the variable most associated with serious assault was whether or not the child was living with their ‘natural parents’. Young children living with no natural parents were twice as likely to be severely abused than young children living with both natural parents, and 9x more likely to be seriously abused than children 5 and over. Though race was a variable in this study, it seemed to generate no statistical differences for discussion.