Abstract Background: Laboratory and epidemiologic research suggests a protective role of magnesium in colorectal cancer development. We estimated the associations of serum and dietary magnesium with colorectal cancer incidence in the Atherosclerosis Risk in Communities (ARIC) study. Methods: Serum magnesium concentration was measured in blood collected twice (1987–1989 and 1990–1992) and averaged. Dietary magnesium was assessed by food-frequency questionnaire administered twice (1987–1989 and 1993–1995) and averaged. For both dietary and serum magnesium, the averaged measures were categorized into quintiles for analysis. Analyses included 315 colorectal cancer cases among 13,009 participants for serum magnesium (followed for a median of 20.4 years), and 256 cases among 10,971 participants for dietary magnesium (followed for a median of 17.5 years). Cox proportional hazards regression was used to calculate multivariable-adjusted HRs and 95% confidence intervals (CI). Results: Multivariable-adjusted HRs (95% CI) of colorectal cancer for the highest four quintiles compared with the first quintile of serum magnesium were as follows: Q2: 0.70 (0.49–0.99); Q3: 0.68 (0.47–1.00); Q4: 0.87 (0.62–1.21); and Q5: 0.79 (0.57–1.11; Ptrend = 0.04). An inverse association was present in females (HR for Q5 vs. Q1: 0.59, 95% CI: 0.36–0.98, Ptrend = 0.01), but not males (HR for Q5 vs. Q1: 1.10, 95% CI: 0.67–1.79, Ptrend = 0.92; Pinteraction = 0.34). Dietary magnesium was not statistically significantly associated with colorectal cancer risk. Conclusions: Our study found a higher risk of colorectal cancer with lower serum magnesium among females, but not males. Impact: If our findings are confirmed, maintaining adequate serum magnesium levels may be important for colorectal cancer prevention.