Objectives The second‐to‐fourth digit ratio (2D:4D ratio) can be assumed a valid indicator of sexual differentiation, provided that it is stable once it develops, or eventual changes remain uniform in a respective cohort throughout ontogenesis. The main goal of this study was to determine whether the metacarpal 2M:4M and the digit 2D:4D ratio change during the period of pubertal/adolescent growth. Materials and Methods The metacarpals and digits were measured on radiographs of left hands in the sample of 328 individuals (96 pairs of male and 68 pairs of female twins) from the Wrocław Longitudinal Study of Twins (1967–1983). Five consecutive annual measurements were done for each individual within a 4‐year‐interval somewhere between 7 and 18 years of age. Age‐related changes in both ratios were studied using a set of mixed‐effects linear models. Three types of correlation coefficients were used for assessment of stability between repeated measurements at different ages. Results An overall decrease in the average 2M:4M ratio was observed, attributable to a much larger extent to males than to females. On the contrary, a slight overall increase in the average 2D:4D ratio was observed, attributable to a much larger extent to females than to males. The rank order of the ratios remained highly stable within the monitored period (the correlation coefficient mostly ranged between 0.85 and 0.95). In spite of these findings, we recorded significant intraindividual changes in both ratios. In some individuals the 2D:4D ratio can undergo changes comparable to average sex differences and much higher than average age‐related changes. Discussion Relatively slight overall changes in digit ratio in puberty and adolescence themselves are not inconsistent with the use of the 2D:4D ratio as an indirect marker of prenatal sexual differentiation. Nevertheless, individual changes in the ratios varied substantially in this study and differed from the average trends. Future studies should focus on the nature of interindividual developmental differences in the digit and metacarpal ratios.