Daly and Wilson (1994, 2008) reported that rates of fatal assaults of young children by stepfathers are over 100 times those by genetic fathers, and they explain the difference in evolutionary terms. Their study was replicated by comparing updated homicide data and population data from 3 surveys. This indicated that the risk to young stepchildren was approximately 16 times that to genetic children, and stepfathers were twice as likely to kill by beating. However, when we controlled for father's age, the risk from cohabiting stepfathers was approximately 6 times greater. Above the age of 4 years, stepchildren were at no greater risk than genetic children. Children are at risk from fathers primarily when both are young and they do not live together; stepfathers' apparent overrepresentation results largely from their relative youth and from many nonresidential perpetrators being labeled stepfathers. Other factors are also influential, but if these include stepparenthood, its impact is considerably less than previous researchers have claimed. (PsycINFO Database Record (c) 2019 APA, all rights reserved).