Objective: Circulating leptin and vitamin D are both associated with obesity and diabetes, but their rehtionship remains unclear. We conducted an analysis of the association between serum leptin and vitamin D in a nationally representative sample of American populations. Methods: From the third National Health and Examination Survey (NHANES III), we included 6,398 adult men and women who had measurements of serum leptin and vitamin D between 1988 and 1994. Leptin and vitamin D levels were quantified by radioimmunoassay. Multivariable linear regression analyses with sampling weights were performed to assess the associations between levels of leptin and vitamin D. Analyses were adjusted and stratified for potential confounders. Results: In the age-and sex adjusted model, each log-unit increase in leptin levels was associated with a 7.46 nmol/L (95% Cl: 5.99 to 8.92) decrease in vitamin D levels. Findings were similar when using sex-specific tertiles or leptin. In race-stratified analyses, inverse associations remained among non-Hispanic Whites (β = -7.61, 95% CI: -9.48 to -5.73) and Mexican-Americans (-2.70, -4.98 to -0.43) but not for other race/ethnicities, although no significant interaction was found (Pinteraction > 0.05). Results remained significant albeit slightly weaker when further adjusted by socioeconomic status, BMI, physical activity, or cigarette smoking (β = -4.09, 95% CI: -6.64 to -1.55). A sensitivity analysis excluding those previously diagnosed with diabetes did not alter the results. Conclusion: Our findings showed that higher serum vitamin D is associated with higher serum leptin in American adults, irrespective of potential confounders and diagnosis of diabetes. Further investigations should be conducted to assess whether this association is causal.