The surge in violent juvenile crime in the late 1980s coincided with an increase in drug arrests, which particularly increased among non-whites in urban areas. To meet the increased demand for crack cocaine, the drug industry had recruited young sellers, primarily nonwhite youths, many of whom viewed this as their only viable economic opportunity. For white youths, drug arrests declined during this period. Guns are increasingly involved in homicides by juveniles. From 1976 to 1985, when the homicide rate by juveniles was fairly steady, a gun was used more often than any other weapon (a constant 60 percent of the time). Seven years later, by 1992, the number of murders by juveniles in which a gun was involved had doubled, but the number committed without a gun remained steady. As more juveniles were recruited to sell crack, they armed themselves with guns. Of particular relevance to future crime rates is the increasing number of teenagers. The age group currently responsible for the highest homicide rate, the 18-year-olds, is about as small as it has been in recent years. As this cohort ages, its crime rate may level off, as it has in the past; however, children who are now younger (ages 5-15) represent the future problem, because they are larger cohorts than the current 18-year-old group. Even if current rates do not continue to increase, violent crime is likely to increase because there will be more 18-year-olds to commit violence at a higher rate. In the long term, a major effort must build a base of understanding by involving the research community in examining juvenile crime and its causes, the links between guns and drugs in the community and in their respective markets, and the rate of homicides among youth. The war on drugs should be assessed to determine its positive and negative effects, including its unintended consequences.