Depression is common and a major cause of morbidity and mortality and is also known to have serious effects on quality of life. Both clinical and pharmacologic studies have implicated the role of brain-derived neurotrophic factor (BDNF) as a susceptibility locus for the development of mental illness, including depression, bipolar disorder, and schizophrenia. Population-based genetic studies have examined the association between BDNF and a variety of depression outcomes, but the results have not clearly established the role of BDNF in the development of this complex disorder. The aim of this study was to test for associations between two genetic variants in BDNF, Val66Met (rs6265) and -270 C > T, and depression measured in two independent samples. In this analysis we included 3,548 participants from British Women's Heart and Health Study (BWHHS) and 6,836 mothers from Avon Longitudinal Study of Parents and Children (ALSPAC) who had complete data on genotype and depression outcomes. We did not detect any strong evidence of associations between any of the two polymorphisms and indicators of depression in either BWHHS or ALSPAC samples. Further, we carried out a systematic review and meta-analysis of all association studies of these two BDNF polymorphisms and depression. The meta-analysis of Val66Met in depression obtained an overall summary OR of 1.06 (95% CI: 0.89-1.26, P = 0.537) comparing MM with VV genotypes and an OR of 0.97 (95% CI: 0.89-1.05, P = 0.403) comparing MV with VV genotypes. Our findings suggest that BDNF genotype does not exert a major influence on the development of depression.