The purpose of the work was to study the micropermeability between portions of restorative materials under different surface preparation in laboratory conditions. Materials and methods. The study used 50 samples of cylindrical restorative materials with a diameter of 5.0 mm and a height of 4.0 mm, divided into five groups 10 samples each. The samples of group 1 consisted of two portions of photocomposite without preliminary preparation. The samples of group 2 were produced from two portions, between which the V generation adhesive system was applied. In the samples of group 3, the vertical surface of the first portion of the photocomposite was moistened, adhesive preparation was performed, and the second portion was added. In the samples of group 4, the vertical surface of the portion was polished and similar measures were taken. Group 5 samples consisted of glass ionomer cement and photocomposite. The surfaces of the samples, except for the border, were isolated, the samples were immersed for a day in the dye, sprayed and the depth of its penetration was estimated through the border on vertical cuts in points and by a computer program in millimeters. Results and discussion. Micropermeability in the samples of group 1 was 1.3 ± 0.15 points and was significantly (p<0.05) the lowest among all groups except group 2. The indicator in the samples of group 2 was 1.7 ± 0.15 points with an unreliable (p>0.05) difference from the given indicator, in the samples of group 3 – 1.9 ± 0.18 points with a significant (p<0.05) difference, the last two indicators had an unreliable difference (p>0.05) between them. Significantly (p<0.05) the highest micropermeability was in the samples of group 4 – 4.1 ± 0.18 points. In samples of group 5, the indicator was equal to 2.6 ± 0.16 points (p<0.05). According to the computer analysis of the image in the samples of group 1, the dye deepened by 0.55 ± 0.15 mm, which is reliably (p<0.05) the lowest indicator. In the samples of group 2, it was 1.20 ± 0.18 mm, in the samples of group 3 – 1.31 ± 0.12 mm (the difference was not significant, p>0.05). The indicator was significantly (p<0.05) higher in the samples of group 5 – 2.13 ± 0.16 mm, in the samples of group 4 it was 3.65 ± 0.19 mm and was significantly (p<0.05) the highest, exaggerating by 6.6 times the indicator of the samples of group 1. A significantly high indicator of micropermeability was found in the samples consisting of glass ionomer cement and photocomposite, and this should be a certain warning, however, in clinical conditions, when using an open “sandwich technique”, as a rule, photocomposite material overlaps a certain area of glass ionomer cement without marking a clear border between materials. Nevertheless, it is necessary to pay special attention to the formation of this border during the direct restoration of teeth with damage to their contact and occlusal surfaces in the “sandwich technique” and to ensure the tightest adhesion of the photocomposite to the glass ionomer with somewhat limited visual control. Conclusion. The lowest indicators of micropermeability according to two estimates were established between portions of the photocomposite material without any preparation of their surfaces. The highest micropermeability was found in the case of applying the photocomposite to the polished surface of the material