BACKGROUND We assess the longitudinal associations between marijuana and cigar (little cigars and cigarillos [LCCs] and large cigars) use on subsequent initiation of marijuana and cigar use. METHODS Data are from a cohort study of 2189 young adults recruited in fall 2010 from 11 colleges in the Southeast. We used discrete-time survival analysis to examine whether ever use of marijuana at baseline (spring 2011, freshman year) predicted initiation of LCCs and large cigars and whether ever use of these cigar products predicted initiation of marijuana use across 10 waves of data collection (2011-2018). RESULTS The sample was 65.3 % female, 83.6 % White, 5.9 % Hispanic, and 61.8 % had college-educated mothers. At baseline, 70 % reported never using LCCs, 71 % reported never using large cigars, and 74 % reported never using marijuana. Ever use of marijuana at baseline was associated with an increased risk of LCC initiation (Incident rate ratio [IRR] = 1.6, 95 %CI = 1.0, 2.5) but not large cigar initiation. Ever use of LCCs (IRR = 1.4, 95 %CI = 1.1, 1.8) and ever use of large cigars (IRR = 1.3, 95 %CI = 1.1, 1.8) at baseline both predicted initiation of marijuana use. CONCLUSIONS Our findings support growing evidence that marijuana and LCCs are strongly associated and use of one substance predicts use of the other. In contrast to studies of adults, we also found that young adults who have tried large cigars may be at increased risk for subsequent marijuana use. These findings highlight the need to consider each product as a potential gateway of the other when developing interventions for young adults.