Smoking rates among college students is increasing, yet little is know about the development of students' smoking. This longitudinal analysis focuses on a national sample of high school students who were college students 4 years later (n = 1,479). SUDAAN statistical analysis procedures were used to weight and adjust for sampling design and nonresponse. Approximately 37% of the college students at follow-up who were never smokers at baseline had initiated smoking within the 4 years. Among experimenters at baseline, 25% had progressed their smoking behavior over the 4 years. Multivariate logistic regression analyses showed that there were different predictors of these two transitions among college students. Students who were more likely to initiate smoking were White, did not like school as much, and were more rebellious. Students who were more likely to progress from experimentation to a higher level of smoking thought peers approved of smoking and believed experimentation with smoking was safe. Several predictors of smoking initiation and progression during the transition from high school to college were identified that could be important components of interventions targeting high school and college-bound students. Increased efforts should be employed to include college students in national smoking prevention and cessation programs.