Injury and participation information was collected over 5 years (1993–1997) on varsity men’s football players in the Canada West Universities Athletic Association. The locations of acute time-loss injuries or neurologic injures were coded as head and neck, upper extremity (shoulder to hand), or lower extremity (hip to foot). Poisson regression-based generalized estimating equations were used to estimate rate ratios and 95% confidence intervals. Injury rates were higher during games as compared with practice periods (for the head and neck, rate ratio (RR) = 9.75 (95% confidence interval (CI):​ 7.50, 12.67);​ for upper extremities, RR = 5.76 (95% CI:​ 4.46, 7.45);​ and for lower extremities, RR = 7.06 (95% CI:​ 6.03, 8.25)). In dry-field game situations, head and neck injury rates were 1.59 times higher on artificial turf than on natural grass (95% CI:​ 1.04, 2.42). Lower extremity game injury rates were higher on artificial turf than on natural grass under both dry (RR = 1.83, 95% CI:​ 1.35, 2.48) and wet (RR = 2.31, 95% CI:​ 1.18, 4.52) field conditions. Injury rates increased with every additional year of participation. Past injury increased the rate of subsequent injury. The effect of an artificial field surface may be related to infrequent use. Risk factors for injury included participation in a game, playing on artificial turf, being a veteran player, and having a past injury.