In order to elucidate the mechanisms of nutrient supply in soil, dynamics of soil solution composition in the rooting zone was investigated. 1) For nondestructive and repeated soil solution sampling from the heterogeneous rooting zone, a looped hollow fiber technique was first developed. The improved sampler, composed of a hollow fiber in loop shape, a silicon tube and a disposable syringe, enabled investigation of the temporal and spatial dynamics of soil solution composition. 2) The effect of plant growth on the dynamics of soil solution was studied in the presence of maize by pot experiments. Soil solutions were sampled 5-6 times during the 50 day growing period. EC and concentrations of NO3-, Ca2+, Mg2+ and K+ decreased and pH increased with plant growth. The activity ratio for K (ARK), a index of the soil's cation ratio and availability of K, also increased, suggesting that an ARK gradient was formed in the vicinity of roots by the greater uptake of K compared with Ca and Mg. 3) During plant growth, electrical neutrality was maintained and NO3- was the main controlling factor as the dominant anion. Accordingly, decrease of N in the soil solution was almost equal to the amount of plant uptake. On the contrary, decrease of Ca and Mg in the soil solution was several times greater than the amount of plant uptake, whereas, decrease of K in the soil solution amounted less than 10% of the amount of plant uptake, suggesting intensive replenishment of K from soil. 4) To investigate the mechanisms of nutrient supply in soil, the amount of nutrient supply by mass flow was calculated as the sum of the product of the amount of transpiration and average concentration of the nutrient in the soil solution for several pot experiments with a variety of soils and fertilizer applications. In general, the relative contribution of mass flow to the total amount of plant uptake was far greater than 100% for Ca and Mg, 50-150% for N, 5-38% for K and 1-6% for P, indicating that the supply of N, Ca and Mg was mainly by mass flow and the supply of P and K was mainly by diffusion (not mass flow). 5) By the analysis of inorganic and organic fractions of N and P in the soil solution separately, it was further indicated that organic fraction of N and P also contributed considerably to their total supply, not only for Andisols but also for Fluvisols. In conclusion, these findings directly lead to the understanding of the nutrient dynamics in the rhizosphere and also provide fundamental basis for rational soil/fertilizer management for sustainable agriculture.