During the past two decades, a small but convincing body of evidence has been collected to demonstrate the importance of disaggregating homicide rates into more significant and homogeneous subcategories. This chapter reviews various classification strategies that have been proposed and examined by homicide researchers, along with the results of several studies that support the utility of this approach. Among those who study homicide, especially from a macrolevel perspective, there is generally broad support for the utility of categorizing homicides into more specific and homogeneous subgroups. Common themes on how best to accomplish this have focused on the victim-offender relationship and the somewhat loosely and flexibly defined concept of circumstances. Researchers have proposed a variety of specific typologies, however, but as yet there does not appear to be one approach that is commonly supported or used. Important conceptual issues regarding how disaggregation is approached and operationalized thus still must be resolved, and several significant methodological issues require more attention than they have received to date. The likelihood of further advances in this area will depend largely on whether researchers are able to develop and agree on theoretically significant classification systems and whether the necessary data are available to support such strategies. 2 tables and 46 references