Numerous experiments with nanoparticles have recently led to a better understanding of the migration of colloids and larger particles in soils. However, it remains unclear how colloidal particles migrate in soil horizons without macropores, and whether they can move with the flow of capillary water. In this article, we tested the hypothesis that colloidal particles can be transported by water flow in capillary-sized soil pores. To test our hypothesis, column experiments with platinum nanoparticles were carried out. The columns contained undisturbed monoliths from the Luvisols and Phaeozems soil horizons in the southeast of Western Siberia. The lower part of the soil columns was immersed in a colloidal solution with platinum nanoparticles. Thus, we checked whether the nanoparticles would rise to the top of the columns. Platinum nanoparticles are a usable tracer of colloidal particle migration pathways. Due to the minimal background concentrations, platinum can be detected by inductively coupled plasma mass spectrometry (ICP-MS) in experimental samples. Due to their low zeta potential, nanoparticles are well transported over long distances through the pores. Our experiments made it possible to establish that the process of the transfer of nanoparticles with a flow of capillary water is possible in almost all the studied horizons. However, the transfer distances are limited to the first tens of centimeters. The number of migrating nanoparticles and the distance of their transfer increase with an increase in the minimum moisture-holding capacity and decrease with an increase in the bulk density of soil horizons and an increase in the number of direct macropores. The migration of nanoparticles in capillary pores is limited in carbonate soil horizons. The transfer of colloidal particles through soil capillaries can occur in all directions, relative to the gravity gradient. Capillary transport plays an important role in the formation of the ice composition of permafrost soils, as well as in plant nutrition.