Introduction. The influence of the gait on the hip joint formation in patients with infantile cerebral palsy (CP) is an actual object of scientific research. The purpose of the research was to study the correlations between walking and clinical and roentgenometric parameters of the hip joint in the patients with CP. Materials and methods. There were examined 39 patients with CP and pathology of the hip joints (70 joints), who had been treated at the National Research Institute of Traumatology and Orthopedics for the period from 2018 to 2022. The patients had been divided into 2 groups depending on the ability to walk. We had performed a clinical and roentgenogramometric examination of the hip joints in positioning according to our own method and the standard anterior-posterior radiological position. All patients underwent a clinical assessment of femoral torsion using the Ruwe method. To assess the relationships between the studied indices Spearman's correlation analysis was performed. Results. The positive influence of gait on the hip joint formation had been revealed. The average values of the hip clinical and roentgenometric parameters in patients with CP who do not walk were significantly higher compared to those of subjects who walk. Correlation analysis revealed the relationship between walking and the studied parameters of the hip joint (cervical-diaphyseal angle, femoral torsion, acetabular angle, Sharp angle, Reimers index, the center edge angle of Wiberg) in both settings, as well as walking and other factors (age, level lesion, GMFCS scale (II-IV level), congenital dysplasia of the hip joints in the history). At the same time, there was no found any dependence between gait and adductor myotomy in the history. Conclusions. A significant influence of the gait on the following parameters of the hip joint was established: the true cervical-diaphyseal angle (p = 0.00001) in positioning according to our own method, femoral torsion (p = 0.01), acetabular angle (standard setting) (p = 0.00001), Sharpe angle (standard setting) (p = 0.018), Reimers index (standard setting) (p = 0.00007), center edge angle of Wiberg (standard setting) (p = 0.001) and lack of statistical significance of the influence of the adductor myotomy factor in history (p = 0.11) on the walking function.