The evidence that measures of obesity and stature are associated with prostate cancer is weak and inconsistent. We performed a systematic review and meta-analysis of the relationship between body mass index (BMI), height, weight, waist circumference and waist-to-hips ratio (WHR) and the risk of prostate cancer. Study-specific dose-response slopes were obtained, and random effects rate ratios (RRs) were computed from linear meta-regression models. We included 55,521 cases identified among 2,818,767 men from 31 cohort studies, and 13,232 cases and 16,317 controls from 25 case-control studies. The overall RR for BMI was 1.05 per 5 kg/m2 increment, 95% CI 1.01-1.08. For studies that reported results by stage of disease, the RRs were stronger for advanced disease (RR 1.12 per 5 kg/m2 increment, 95% CI 1.01-1.23) compared with localized disease (RR 0.96 per 5 kg/m2 increment, 95% CI 0.89-1.03), p = 0.02. Height was also positively associated with risk (RR 1.05 per 10 cm increment, 95% CI 1.02-1.09), but the evidence was weak for weight (RR 1.01 per 10 kg increment, 95% CI 0.97-1.04), waist circumference (RR 1.03 per 10 cm increment, 95% CI 0.99-1.07), and WHR (RR 1.11 per 0.1 unit increment, 95% CI 0.95-1.30). Stronger associations were observed among cohort studies compared with case-control studies for BMI (p = 0.006), height (p < 0.001) and weight (p = 0.02). This meta-analysis indicates that obesity is weakly associated with an increased risk of prostate cancer (particularly advanced stage tumors). While increased stature may also increase risk, there is little evidence for an association with central obesity.