The phenomenon of infant homicide (infanticide) has been examined, explained, justified, and treated according to physiological, psychiatric, and psychological correlates, but there has been little examination of the social correlates that directly pertain to infant homicide; however, social correlates are often indirectly addressed in the medical and psychiatric literature. This study tested relationships among social correlates often asserted, but typically not tested, in the medical and psychiatric literature. Using a sample of 380 infant homicides in Texas from 1981 through 1991, a multivariate analysis was used to predict the age at which an infant (less than 34 months old) is likely to be fatally injured. A multivariate analysis was conducted on the victim/offender relationship, cause of death, and victim's age at the time of the fatal injury. The findings support the following hypotheses: as the age of the victim increases, the level of violence used to fatally injure the infant increases; and as the level of relational intimacy decreases, the level of violence used to fatally injure the infant increases. The leading cause of death was head trauma (54.2 percent), and the second-largest category was body/abdomen trauma (10 percent). The fatal injuries were most often reported as inflicted by the victim's mother (32.4 percent), followed by the victim's father (24.7 percent) and the mother's boyfriend (16.4 percent). The mean age of the victim was 11.06 months, and the median age was 9.25 months. The risk of fatal injury decreased with each additional day of life. Head trauma had its lowest occurrence at birth and tended to have its highest occurrence at age .06 to 3.75 months (78.26 percent). 4 figures, 1 table, and 36 references