STUDY QUESTION: Does cord blood androgen level obtained at birth affect the AGD in human newborns? SUMMARY ANSWER: In human newborns, though males have a significantly longer AGD compared to females (as early as 22 weeks of gestation) the AGD is not affected by androgen levels at birth in both the sexes. WHAT IS KNOWN ALREADY: Animal studies have reported a critical time period in early fetal life, termed the masculinization programming window (MPW) during which AGD is fixed by in utero androgen action and is unaffected by testosterone levels later during gestation. Thus, AGD may serve as a lifelong biomarker of androgen exposure during this window. This MPW is hypothesized to occur in humans at 8-14 weeks of gestation during which AGD is fixed. The effect of androgens (testosterone) on AGD after the MPW in humans is not known. Furthermore, altered AGD has been associated with various human reproductive health disorders in both males and females. STUDY DESIGN, SIZE, DURATION: A prospective descriptive cohort study was performed using data from randomly selected neonates (n = 205) born at a single center over a period of 1 year (August 2015 to August 2016). PARTICIPANTS/MATERIALS, SETTING, METHODS: AGDs in male (n = 117) and female infants (n = 88) together with penile width, glans girth and stretched penile length were measured by trained caregivers. Gestation ranged from 22 to 41 weeks and infants were examined within 24 h of birth (within 48-72 h in very sick preterm infants after clinical stabilization). AGD-1 was measured from the center of the anus to the posterior base of scrotum in males or to the posterior fourchette in females. AGD-2 was measured from the center of the anus to the anterior base of the penis in males or to the clitoris in females. Sex steroid hormones (testosterone, 17-OH progesterone (17-OHP) and androstenedione) were measured in serum prepared from umbilical cord blood samples taken at birth, using liquid chromatography-tandem mass spectrometry. MAIN RESULTS AND THE ROLE OF CHANCE: Males had a significantly lower gestational age (mean ± SD; 34.6 ± 4.9 versus 36.1 ± 4.1 weeks, P = 0.04), and a significantly longer AGD-1 (mean ± SD; 21.6 ± 6.0 versus 12.7 ± 3.8 mm, P < 0.001) and AGD-2 (41.9 ± 8.7 versus 33.9 ± 7.1 mm, P = 0.004) compared to female infants, respectively. The cord serum testosterone levels were significantly higher for male than female infants [median, interquartile range; 13.0 (7.3, 20.5) versus 4.1 (2.5, 5.9), ng/dl, P < 0.001]. There was no difference in levels of 17-OHP (P = 0.697) or androstenedione (P = 0.601) between the two sexes. On multiple regression analysis after adjusting for potential confounders, none of the AGD's in both males and females correlated with any sex steroid hormonal levels. We also provide normative charts for penile length, penile width and glans girth in preterm and term infants. LIMITATIONS, REASONS FOR CAUTION: No data were collected on family history of genital malformation, infertility or hormonal disorders, parental endocrine-disrupting chemical exposure or diet pattern, any of which might have influenced the AGD and/or sex steroid hormone levels in the offspring. WIDER IMPLICATIONS OF THE FINDINGS: Our results suggest that AGD in humans, like animals, is fixed in early gestation (likely during the hypothesized MPW) and is unaffected by androgen levels thereafter. Thus, AGD can serve as a biomarker of in utero androgen action during early gestation (likely 8-14 weeks) in humans. As such, causes of human newborn and adult reproductive health disorders, such as endocrine disruptors, should be explored during early gestation. However, further larger studies are needed to help corroborate these findings. STUDY FUNDING/COMPETING INTERESTS: No specific funding was obtained for this study, and all authors have no conflict of interest to declare.