PURPOSE: To examine whether similar risk factors influenced episodic and persistent gun-carrying among urban African-American adolescents. METHODS: The sample consisted of 705 African-American youths (48.9% male; mean age at baseline = 14.56 years) who were interviewed annually throughout high school as part of a larger study on students who leave school before graduation. Episodic gun-carrying was defined as carrying a gun during one or two waves of the study. Persistent gun-carrying involved carrying a gun during three or four waves. Data were analyzed using multinomial logistic regression to test how risk factors assessed at ninth grade influenced the persistence of gun-carrying. RESULTS: Fifteen percent of students reported carrying a gun episodically, and 5% persistently. "Male gender" (OR = 3.61, 95% CI = 2.16-6.04), "adult weapon-carrying" (OR = 1.58, 95% CI = 1.20-2.09), "marijuana use" (OR = 1.03, 95% CI = 1.01-1.06), "selling drugs" (OR = 3.24, 95% CI = 1.52-6.92), and "fighting" (OR = 1.57, 95% CI = 1.14-2.15) distinguished noncarriers from episodic carriers. Frequency of fighting (OR = 1.60, 95% CI = 1.00-2.57) and selling drugs (OR = 3.29, 95% CI = 1.16-9.35) distinguished episodic gun-carriers from persistent gun-carriers. Variables associated with victimization did not uniquely differentiate among the patterns of gun-carrying. These results were similar for males and females. CONCLUSIONS: Similar risk factors characterize episodic and persistent gun-carrying. Specifically, selling drugs and fighting had a strong dose-response relationship with the persistence of gun-carrying. In this population, episodic gun-carrying should be viewed as very risky and not merely as youthful experimentation or a defensive behavior.