Interactions between bacteria and protozoa in soil were studied over 2-week periods in the field and in a pot experiment. Under natural conditions the total biological activity was temporarily synchronized by a large rainfall, and in the laboratory by the addition of water to dried-out soil, with or without plants. In the field, peaks in numbers and biomass of bacteria appeared after the rain, and a peak of naked amoebae quickly followed. Of the three investigated groups-flagellates, ciliates, and amoebae-only populations of the latter were large enough and fluctuated in a way that indicated a role as bacterial regulators. The bacterial increase was transient, and the amoebae alone were calculated to be able to cause 60% of the bacterial decrease. The same development of bacteria and protozoa was observed in the pot experiment: in the presence of roots, amoebic numbers increased 20 times and became 5 times higher than in the unplanted soil. In the planted pots, the amoebic increase was large enough to cause the whole bacterial decrease observed; but in the unplanted soil, consumption by the amoebae caused only one-third of the bacterial decrease.