Data were obtained from 376 domestic assault cases reported to one police department in 1986 and 1987. Acquaintance assaults were more than twice as likely to occur as either domestic or stranger assaults. Most victims experienced no injury or only minor injury, and the type of force or weapon used in most cases involved the assailant's hands, arms, or feet. As the relationship between the parties changed from more to less intimate, arrest became more likely. Although arrest was an unlikely event overall, it characterized stranger assault more than domestic assault. When an arrest occurred, it was more likely when situational elements were present or more evident. Arrest was 2.5 times more likely when a weapon was involved, 8 times more likely when the offender was present at the scene, 2 times more likely when the victim's injury was serious, and 3.5 times more likely when the victim's preference was arrest. A review of the police department's policy manual indicated a low level of interest in domestic violence. Despite the relatively low percentage of arrests made, police officers made fewer arrests for domestic assault than for stranger assault. Further research is recommended to assess whether police departments treat domestic violence cases differently from other similar cases. 57 references, 5 notes, and 4 tables