The achievements of soil hydrophysics are significant, however many aspects are still not clear. Among them-the mechanisms of sorption of water by soils and role of exchange cations in this process, in particular. Parts of a moisture, sorbed by exchange cations and neutral surface of soil particles, were not precisely known till now. While adsorbing water, as a result of mutual attraction of molecules of water and solid phase of soil (both soluble, and insoluble components), there is a change of power characteristics of the system: decreasing of specific free energy resulting in decreasing of total soil-moisture pressure (P, atm), its temperature of freezing, equilibrium relative humidity of air, heating of soils at wetting (HW). It is possible to estimate the role of exchange cations in sorption of water, determining, how the energy of a soil moisture changes at saturation of soils with various cations, as its heats of hydration (HH) are different. It was possible to show, that for loamy and clay soils in a range -5>P>-200 atm there is a close relationship between the soil-moisture content (W) and lg |P|: lg |P|= A - B W. An analysis of the data for clay (published by H.Kuron) has expanded this range up to -5400 atm, and later - up to -20000 atm. Besides, at P=-5400 atm W of clay is directly proportional to HH of cations. All this can be due to the dominant role of exchange cations of the diffusion layer in sorption of moisture by soils. One more argument for the benefit of this hypothesis is the direct proportionality and high correlation (r = 0.96 ± 0.03) between W and cation exchange capacity (CEC) of soil colloides. At CEC=50 mmol-eq./100 g, W can reach 50-150%. It is interesting, that at CEC=0, HW=6 cal/g of colloides. Hence, even if there is low content of exchange cations in soil colloides (for example, quartz sand containing low content of humus), they can sorbe some moisture on the neutral surface of a solid phase. The analysis of experimental data has shown, that at CEC<0.5 mmol-eq./g sorption of water occurs mainly on the neutral surface of colloid particles; at CEC>0.5 mmol-eq./g the role of exchange cations prevails; at CEC=0.5 mmol-eq./g their contributions are equal. The relation HW/CEC is inversly proportional to CEC. This relation varies from 53.5 in sandy soils up to 8.8 in clay soils. As CEC of soil colloides depends on mineralogical composition of soils, HW also depend on it. For understanding of nature of these processes it is important to understand, why various exchange cations, even having equal charges, sorbe water with different energy (for value HH they form a sequence: H - 247; Mg - 230; Ca - 175; Na - 94; K 75 cal/mmol-eq.). For the value of potential of ionization (I, electron-volt), they form the same sequence: H - 13.5; Mg - 11.3; Ca - 9; Na - 5.1; K - 4.3 e-v. If in calculation for Mg and Ca one uses I average for two electrons which are different there is a direct proportionality between HH and I of metals: HH =(23 I 25) cal/mmol-eq. High correlation (r = 0.97 ± 0.02) of this dependence allows to assume, that it is I that causes cation's ability to sorbe moisture. I, in turn, is a result of electronic envelope structure of atoms and, hence, a result of position of atoms in the Periodic System of Elements by Mendeleev. In it's each period I increases as nuclear weight increases. However, within any group, on the contrary, the increase of nuclear weight results in reduction of I. It is caused by different energy of electron attraction to a nucleus, taking place on various outside levels and sublevels of atoms electronic. Normal (standard) electrode potential (E, volt) is equal for K - 2.92; Ca - 2.84; Na - 2.713; Mg - 2.38; H - 0. For the first time it was possible to find out close (r=-0.98 ± 0.02) relationship: E = 6 - 0,64 I. In result there is an close relationship between HH and E (r = -0.97 ± 0.02): HH = 75 (4 - E). It's physical sense: if for a hypothetical ion E = 4 v, it's HH = 0 (this ion can not be hydrated); if E = 0, HH = 300 cal/mmol-eq. If one use for Mg and Ca average I for two electrons, there is a direct proportionality between HH and (I E) of metals (r = 0.98 ± 0.02): HH = 20.5 (I E) + 45. Its physical sense: if for a hypothetical ion I = E, it's (I E) = 0 and HH = 45 cal/mmol-eq.