Objectives Androgen receptor gene CAG repeat, AR (CAG)(n), polymorphism is thought to have an effect on male reproductive functions and a relationship between long AR (CAG)(n) and decreased androgenic activity has been shown. Therefore, we hypothesized that in adolescents with long AR CAG repeat the prevalence of pubertal gynecomastia (PG) will be higher and we aimed to investigate the association between AR (CAG)(n) polymorphism and PG in Turkish adolescents. Methods Adolescents with PG between 11 and 19 years of age were enrolled as the study group and healthy individuals without a history of PG, who were at least 14 years of age and Tanner 4 or 5 were enrolled as the control group. The AR (CAG)(n) length was detected by direct DNA sequencing analysis and reproductive hormones were measured by standardized analyses. Results The mean AR (CAG)(n) was 22.3 +/- 2.6 (mean +/- SD) in the PG group (n=101) and 21.9 +/- 3.1 (mean +/- SD) in the control group (n=88) (p=0.276). The adolescents with short AR (CAG)(n) had lower body mass index standard deviation scores (BMI SDS) compared to the adolescents with intermediate and long repeat numbers (p=0.029). Conclusions The results of this study showed a lack of direct association between AR (CAG)(n) and PG. However, the significant relationship between the AR (CAG)(n) quartiles and BMI SDS suggests that long AR (CAG)(n) might cause PG indirectly. Further studies are needed to better clarify this relationship.