Eighteen important Quaternary loess paleosoil sites have been studied across the European loess belt. They included approximately 50 buried soils, the over and/or underlying loess and the present-day surface soils (where possible). From the numerous types of secondary CaCO3 accumulations recognized, only the small scale accumulations are discussed in this paper, including calcified root cells, CaCO3 hypocoatings, needle-fiber calcite, powdery calcite coatings, earthworm biospheroliths, and calcite pseudomorphs on gypsum. Most of the features studied are found in the rhizosphere microecosystem. Each of these features showed a unique relationship with organic matter, micro-organism and plant distribution, moisture and leaching processes. For example, the exceptional abundance of calcified root cells is an indicator of a long stability of soil surface during favorable climatic conditions. Needle-fiber calcite is related to biomineralization of saprophytes in soils with sufficient moisture and decomposing organic matter. The presence of powdery calcite coatings covering clay and Mn coatings in non-calcareous soils indicates a change from leaching to non-leaching soil moisture regimes. Earthworm biospheroids provide arguments for a relative stability of the soil surface, without rapid burial processes or drastic, prolonged water stress. Finally, gypsum pseudomorphs are indicators of changes in local climatic parameters. In addition to a discussion of the genesis and the dynamics of the various accumulations, several examples are given to show how these accumulations can be used as indicators of a changing environment, such as renewed loess deposition, the extent of drought or varying intensity of leaching during the year. To conclude, various forms of small scale secondary CaCO3 accumulations are common in the loess paleosoils and each type has the potential to contribute to a better understanding of the evolution of the paleoenvironment. Therefore, it is proposed that their individual presence or absence throughout the pedostratigraphical levels and throughout the (paleo)topography should be recorded at all scales of observation.