Concretions in typical chernozem,gleyed chernozem-like,and solonetzic chernozem-like soils of the southern Tambov Lowland

A system for the diagnostics of chernozemic soils of the Tambov Lowland based on concretions is proposed for agricultural and reclamation purposes. The relationships between the structure and composition of the carbonate concretions, the long-term water regime of the soils, and the productivity of the crops have been established. The dense concretions in the typical chernozem testify to the depth of the seasonal wetting; the angular-rounded concretions in the deeply gleyed chernozem-like soil, to the upper boundary of the capillary fringe; and the angular concretions with sharp edges and cavities in the gleyic chernozem-like soils, to the groundwater table. In the chernozem-like soils that were waterlogged with bicarbonate-sodium water, the black angular concretions were formed in the solonetzic horizons, while the weakly compacted light-colored ones, in the zone of the capillary fringe. Humic acids were responsible for the color of the dark neoformations, and fulvic acids predominated in the light-colored ones. The appearance of black fine nodules indicated periodic surface water stagnation. Manganese predominantly accumulates in these nodules. The structure of the Mn-Fe concretions in the plow horizon observed at a magnification of 40–50 times has a diagnostic importance. The short-term (2–3 weeks) water stagnation leads to the formation of fine-stratified concretions, and the long-term (up to 1.5 months) stagnation promotes the formation of uniform porous ones. The solonetzic process induced by the bicarbonate-sodium water results in the appearance of mottled concretions.     

Clay minerals in chernozem-like soils of mesodepressions in the northern forest-steppe of European Russia

In the northern forest-steppe of European Russia, under the conditions of surface waterlogging (freshwater) and a stagnant-percolative regime, gleyic podzolic chernozem-like soils with thick light-colored eluvial horizons are formed. These horizons are close or similar to the podzolic horizons of bog-podzolic soils in many properties of their solid phase. They are bleached in color and characterized by the removal of Ca, Mg, Fe, Al, and Mn and the relative accumulation of quartz SiO₂. These soils differ from leached chernozems in their acid reaction and very low CEC, the presence of Fe-Mn concretions and coatings, and the significant decrease in the clay content in the A2 horizon as compared to the parent rock. The soils studied differ significantly from loamy podzolic and bog-podzolic soils by the composition of the clay minerals in the A2 horizons: (1) no essential loss of smectite minerals from this horizon was found as compared to the rest of the solum, (2) pedogenic chlorites (HIV and HIS) are absent, and (3) the distinct accumulation of illites is observed as compared to the subsoil and parent material, probably, due to the process of illitization.     

Identification of carbonate pedofeatures of different ages in modern chernozems

Carbonate pedofeatures of three chernozemic soils developed from loesslike loams in the foreststeppe zone of Lipetsk oblast under fallow plot (Luvic Chernozem (Clayic, Pachic)) and under forest (Calcic Chernozem (Clayic, Pachic)) and in the steppe zone of Dnepropetrovsk oblast (Calcic Chernozem (Episiltic, Endoclayic, Pachic)) were studied in the field and laboratory with the use of a set of methods, including the radiocarbon method, mass spectrometry, and micro- and submicromorphology. The morphological diversity of carbonate pedofeatures in these soils was represented by carbonate veins, coatings, disperse carbonates (carbonate impregnations), soft masses (beloglazka), and concretions. In the forest-steppe soils, disperse carbonates and soft masses were absent. The radiocarbon age of carbonate pedofeatures in the forest-steppe soils varied within a relatively narrow range of 3–4.3 ka cal BP with a tendency for a younger age of carbonate concretions subjected to destruction (geodes). In the steppe chernozem, this range was larger, and the ¹⁴C ages of different forms of carbonate pedofeatures were different. Thus, soft masses (beloglazka) had the age of 5.5–6 ka cal BP; disperse carbonates, 17.5–18.5 ka cal BP; and hard carbonate concretions, 26–27 ka cal BP. Data on δ¹³C demonstrated that the isotopic composition of carbon in virtually all the “nonlabile” carbonate pedofeatures does not correspond to the isotopic composition of carbon of the modern soil organic matter. It was shown that the studied chernozemic soils are polygenetic formations containing carbonate pedofeatures of different ages: (a) recent (currently growing), (b) relict, and (c) inherited pedofeatures. The latter group represents complex pedofeatures that include ancient fragments integrated in younger pedofeatures, e.g., the Holocene soft carbonate nodules with inclusions of fragments of the ancient microcodium.     

Clay illuviation and calcium carbonate accumulation along a precipitation gradient in Kansas

One of the significant features of loess-derived soils in Kansas is the occurrence of clay-rich subsurface horizons above a layer enriched with pedogenic carbonates. In order to examine the extent of clay increase and pedogenic carbonate enrichment in a precipitation gradient, ten soil profiles from three different precipitation regions were studied using micromorphological and mineralogical techniques. The precipitation gradient was divided into three groups: 400–550 mm, 550–750 mm, and 750–1100 mm regions. The objectives were to (1) understand the cause of clay orientation in clay-rich horizons (2) investigate the reasons for the clay increase, and (3) observe the interaction of clay and pedogenic carbonate accumulation features along a precipitation gradient in Kansas. Although clay films were identified in the field for soils in the 400–550 mm regions, illuvial clay films were not observed in thin section analysis. The clay accumulations mostly occurred as grain coatings. The rest of the clay accumulations observed were very thin, striated, and mostly associated with voids. The argillic horizons had a granostriated b-fabric, which indicates stress orientation of micromass caused by high shrink–swell activity. Thick and continuous illuvial coatings were observed in the buried horizons of paleosols. In the other two regions where precipitation exceeds 550 mm, illuvial clay coatings with strong orientation were observed along with thin and striated stress-oriented clay. Both types of clay orientations exceeded 1% of the cross-sectional area for the thin section. Although illuvial clay features and pedogenic carbonates were observed in all soils at approximately the same depth, complete obliteration of clay coatings was not observed in these horizons. In-situ weathering of biotite was one of the reasons for the clay increase in all soil profiles. In all soils studied, the clay increase and cause of clay orientation cannot be attributed to a single genetic process or event. Both illuviation and shrink–swell activity were involved in the orientation of clay. Although orientation of clay and pedogenic carbonates were observed in all soils at approximately the same depth, the decomposition of clay coatings was not observed in these horizons.     

悬着水位对铝硅酸盐的稳定性及土壤发生的影响

The mineral stability and solute activities of soil solution extracted from selected horizons of seven studied pedons of Alfisols in Kentucky, USA, and the relationship between distribution of iron-manganese concretions and the restrictive layers were investigated. The results showed that the genesis and development of these soils and mineral weathering trends were strongly influenced by the depth of bedrock and the presence of perched water tables at lithic (limestone) interfaces due to the dissolution and buffering effect of limestone bedrock. The extractable Mg/Ca ratio as depth function and soil depth above bedrock could be used as indices of weathering and degree of soil development. Maximum iron-manganese concretion accumulation was found to occur in the horizon overlying clay horizon (>40% clay) with a sharp increase in clay content (>10%), which suggested that zones of Fe-Mn concretion accumulation in soils of the Inner Bluegrass Region appeared to be a sensitive genetic indicator of argillic horizons with restrictive permeability.     

The influence of topography on the development of Alfisols on calcareous clayey till in Denmark

A detailed macro- and micro-morphological study of seven closely-adjacent soil profiles on calcareous clay tills in a long-established beech wood in central Denmark, shows that depth of decalcification, thickness of textural B horizon, development of umbric epipedons and development of albic sub-horizons are all related to slope inclination, slope form and to relative situation in the slope complex.The soils may be classified within four subgroups of Alfisols, while one is a Mollisol. The MAST of 8°C poses problems of typification at suborder level between frigid and mesic classes.Development of clay skins and of glaebules and concretions as indicators of the redeposition of mobilised components, are present to greater extent in stable slope situations; with least lateral subsoil water movement. Development of albic horizons is related to greater lateral water movement.Development of cutanic features by argilluviation - argillans - do not exclude features formed by redeposition of carbonates - calcitans. Soil plasmic fabrics with both features are termed calcisepic fabrics and prove deposition of clay and carbonates in the same soil horizons, usually the lower B and C(g) horizons.Morphologically the dark epipedons show intense humification and many fecal pellets because of the undisturbed and high plant productivity and intense biotic activity. The micromorphology is one of isotic argillasepic plasmic fabrics. The argillic horizons are dominantly insepic or vosepic, while the calcareous C horizons are either argillasepic, or calci-vosepic or calcisepic. It is possible that some of the carbonate reprecipitation post-dates the argilluviation, the carbonates derived during secondary dissolution in the suprajacent horizons.     

Influences of quaternary climatic changes on processes of soil development on desert loess deposits of the Cima volcanic field, California

Soils formed in loess are evidence of both relict and buried landscapes developed on Pliocene-to-latest Pleistocene basalt flows of the Cima volcanic field in the eastern Mojave Desert, California. The characteristics of these soils change systematically and as functions of the age and surface morphology of the lava flow. Four distinct phases of soil development are recognized: phase 1 - weakly developed soils on flows less than 0.18 M.y. old; phase 2 - strongly developed soils with thick argillic horizons on 0.18 – 0.7 M.y. old flows; phase 3 - strongly developed soils with truncated argillic horizons massively impregnated by carbonate on 0.7 to 1.1 M.y. old flows; and phase 4 - degraded soils with petrocalcic rubble on Pliocene flows. A critical aspect of the development of stage 1 soils is the evolution of a vesicular A horizon which profoundly affects the infiltration characteristics of the loess parent materials. Laboratory studies show that secondary gypsum and possibly other salt accumulation probably occurred during the period of phase 1 soil development. Slight reddening of the interiors of peds from vesicular-A horizons of phase 1 soils and presence of weakly developed B horizons indicates a slight degree of in situ chemical alteration. However, clay and Fe oxide contents of these soils show that these constituents, as well as carbonates and soluble salts, are incorporated as eolian dust. In contrast to phase 1 soils, chemical and mineralogical analysis of argillic horizons of phase 2 soils indicate proportionally greater degrees of in-situ chemical alteration. These data, the abundant clay films, and the strong reddening in the thick argillic horizons suggest that phase 2 and phase 3 soils formed during long periods of time and periodically were subjected to leaching regimes more intense than those that now exist. Flow-age data and soil-stratigraphic evidence also indicate that several major loess-deposition events occurred during the past 1.0 M.y. Loess events are attributed to past changes in climate, such as the Pleistocene-to-Holocene climatic change, that periodically caused regional desiccation of pluvial lakes, reduction of vegetational density, and exposure of loose, unconsolidated fine materials. During times of warmer interglacial climates, precipitation infiltrates to shallower depths than during glacial periods. Extensive, saline playas which developed in the Mojave Desert during the Holocene are a likely source of much of the carbonates and soluble salts that are accumulating at shallow depths both in phase 1 soils and in the formerly noncalcareous, nongypsiferous argillic horizons of phase 2 and 3 soils.     

Charred organic carbon in German chernozemic soils

Burning vegetation produces partly charred plant material which subsequently could contribute to the highly refractory proportion of soil organic matter. The presence of charred organic carbon (COC) was investigated in 17 horizons originating from nine soils from Germany and the Netherlands using a suite of complementary methods (high-energy ultraviolet photo-oxidation, scanning electron microscopy, solid-state ¹³C nuclear magnetic resonance, lignin analysis by CuO oxidation). Charred organic carbon could not be detected in the A horizons of an Alisol and a Gleysol, but it contributed up to 45% of the organic carbon and up to about 8 g kg^–1 of the soil in a range of grey to black soils (Cambisol, Luvisol, Phaeozem, Chernozem and Greyzem). All these soils have chernozemic soil properties (dark colour, A–C profile, high base saturation, bioturbation). A 10-km colour sequence of four chernozemic soils, which were very similar in chemical and physical properties, showed a strong relation between colour and the content of COC. This suggests that the COC affects mainly soil colour in the sequence studied. Finely divided COC seems to be a major constituent of many chernozemic soils in Germany. These results suggest that besides climate, vegetation and bioturbation, fire has played an important role in the pedogenesis of chernozemic soils.     

Dispersion and migration of fine particles in two palygorskite‐containing soils of the Jordan Valley

Migration of different mineral particles within columns of soil‐sand mixtures containing 10 or 20 mass % of soil was investigated by establishing differences in the mineral suite between the ”︁bulk clay” and the ”︁mobile fine material” fractions. The ”︁bulk clay” fractions of all soils contained smectite, palygorskite, kaolinite, quartz, feldspar, and calcite. The soils were saturated with sodium by leaching with NaCl solution, and then leached with distilled water. Clay dispersion and particle migration occurred in the columns. Values of SAR (sodium adsorption ratio) of the effluent decreased with time due to carbonate dissolution. At a certain SAR value, the clays apparently formed aggregates, and as a consequence particle migration stopped in the column. In addition to clay‐sized particles (< 2 μm), very‐fine‐silt‐sized particles (2— 5 μm) were able to migrate in the soil‐sand mixtures, too, and to some extent fine‐silt‐sized particles (5—10 μm) as well. Average size of mobile particles decreases with increase of soil content in the soil‐sand mixtures. The mineralogical composition of the ”︁mobile fine material” changed during the experiment. At the beginning of the experiment, the ”︁mobile fine material” was enriched in the non‐phyllosilicates (especially in calcite, and in some cases in quartz, feldspar and dolomite) and contained low concentrations of phyllosilicates (smectite, palygorskite and kaolinite). At the end of the experiment, the proportion of non‐phyllosilicates decreased, and as a consequence, the proportion of phyllosilicates increased. Among the non‐phyllosilicates, calcite was the most mobile mineral. Among the phyllosilicates, palygorskite was preferentially mobilized in topsoil horizons. In subsoil horizons, on the other hand, kaolinite was preferentially mobilized. This difference was explained by the different nature of carbonates in the topsoil and subsoil horizons. Palygorskite is preferentially occluded within the soil carbonates of lacustrine origin over smectite and kaolinite. These carbonates are present mainly in the subsoil horizons. As a consequence, the presence of these carbonates in the subsoil horizons decreases the migration of mainly palygorskite.     

Interrelationships between soil magnesium forms

Abstract

A second order equation best described the relationship between exchangeable Mg and both total and acid‐soluble Mg in select temperate and tropical soils. The relationship between acid‐soluble and total Mg was linear. Soil types differ in their Mg contents. On a per unit weight of material bases the finer soil particles contained more than 95 percent of the total soil inorganic magnesium. In tropical soils, as in temperate soils, the total Mg content of surface horizons tends to decrease with severe weathering, soil erosion and movement of soil colloidal particles down the profile.     

Chemistry of pedogenesis in Indo-Gangetic alluvial plains

Five soil pedons–two aquic and two udic Haplustalfs and one petrocalcic Natrustalf–from the Indo-Gangetic alluvial plain of Western Uttar Pradesh were investigated to evaluate the pedogenetic processes. Sand/silt ratios indicate that parent material discontinuities are insignificant. Higher K content and lower SiO₂/R₂O₃ ratios of the non-clay fractions in Bt, rather than in the A, horizons suggest maximum weathering at or near the surface.
An almost linear relationship between decrease in molar SiO₂/R₂O₃ and % increase in clay to about 100cm depth in all the pedons, presence of clay argillans in Bt horizons (where % clay, fine/coarse clay ratio and bulk density values are greatest), all indicate that the development of argillic horizons in these soils was due, at least partly, to lessivage of clay. Fe in clay fractions decreases with depth whilst Al increases, but in the fine earth both increase steadily with depth. This, together with crystalline iron concretions in the lower Bt horizons, suggests that in Haplustalfs these horizons are gaining clay by neoformation/ reorganization of illuviated constituents, especially A1₂O₃.     

The soil cover on a gentle watershed slope under conditions of the recent seasonal soil overwetting in the Kamennaya Steppe area

The soil cover pattern on a gentle watershed slope at the Kamennaya Steppe Experimental Farm consists of the areas of typical chernozems, zooturbated chernozems, and leached chernozems evolving into chernozemic-meadow soils; they are oriented along the slope. This soil combination is complicated by the presence of slightly eroded and slightly saline soils. The seasonal overwetting and inundation of the soils are favored by the discharge of the groundwater above the local aquiclude represented by the layer of dark brown calcareous clay at the depth of 1.0–2.2 m from the soil surface and by the concentration of snowmelt runoff in the concave parts of the slope. Seasonal overwetting of the soils leads to the appearance of olive tints and small iron concretions (quasigley features) in the lower horizons; in some cases, proper gley features (bluish coatings) appear in the chernozems. The veinlets of powdery gypsum are formed in the middle-profile horizons of some soils. Salt efflorescence on the surface in some places is also the result of the seasonal overwetting of the soils followed by their intense drying upon the hot and droughty weather conditions.     

The origin of sulfate salts in soils developed on Maikop clays in the North Caucasus region (According to data of the isotope composition of sulfur)

The ratio of ³²S/³⁴S stable isotopes was determined for soils developed on the Maikop clay sediments in the North Caucasus region in order to study the salt geochemistry in the soils. The soil sulfates were proved to be polygenetic. Three groups (generations) of sulfates of different geneses were distinguished according to their isotopic composition (³⁴S). In the underlying deposits and, partially, in the salt-bearing horizons of the soils, secondary sulfates originated from the oxidation of sulfides contained in the marine clay predominate. In the soil profiles, the sulfates inherited from the parent material (marine clay) are mixed with sulfates arriving with the atmospheric precipitation. The portion of the latter reaches about 30% of the total amount of sulfates in the soils. The salt (gypsum)-bearing horizons were formed in the course of the eluvial-illuvial redistribution of salts to the depth of seasonal wetting. The mobilization of sulfate salts from the underlying deposits and their transport to the upper parts of the soil profile with ascending water flows in the course of evaporation from the soil surface did not play a significant role in the accumulation of sulfates in the studied soils.     

Disintegration of the clay fraction in the eluvial horizons of some Danish podzolized soils

Abstract

Using sequential extractions, total elemental analysis, and X‐ray diffraction, we have investigated the impact of the podzolization process on component composition of the clay fraction in the eluvial horizons of eight more or less podzolized Danish soils. The results indicate that podzolization is highly aggressive towards all clay components in the eluvial horizons eventually leading to their disintegration. The 2:1 layer silicate clay minerals, illite and chlorite, are first transformed into other 2:1 layer silicate clay minerals. After passing through a microcrystalline phase high in Si but low in Al, Fe, Mg, and K, they finally disintegrate completely. Even gibbsite and kaolinite disintegrate under the aggressive conditions, caused among other things by the presence of dissolved complex forming organic molecules in these horizons. Application of lime and fertilizers seems to be able to reverse the process in case of the 2:1 layer silicate clay minerals.     

A study of argillic horizons in some soils in Morocco

Whether some soils observed in Morocco prior to the 1966 conference on Mediterranean soils held in Madrid had argillic horizons was the subject of active discussions. Textural differences between A- and B-horizons were obvious. Consequently, individual horizons of a number of profiles were sampled for micromorphological study. No clay films were observed in thin sections representing the B-horizons free in carbonate. Peds had pressure faces and internal striations in the matrix. Microchurning and swelling and shrinking are believed to have obliterated evidence of clay illuviation in the B-horizons, but clay coatings were observed at greater depth in well-developed B_ca-horizons.     

Schwermetallverteilung und -bilanzen typischer Waldböden aus nordischem Geschiebemergel

Heavy metal distribution and balances of typical forest soils out of German and Danish till The total contents of Cd, Cu, Fe, Mn, Pb and Zn were determined in different horizons as well as in clay coatings, mottles and concretions from 4 Orthic and Gleyic Luvisols around the Baltic Sea. Profile balances resulted in very small losses of Fe, small losses of Cu, Pb and Zn but higher losses of Cd and Mn. All elements were translocated by clay migration but corresponding enrichments in the B-horizons were covered by translocations in soluble form. Only Fe and Mn had been enriched in mottles of the water logged soils.     

The origin and early genesis of clay bands in youthful sandy soils along lake Michigan,U.S.A.

A beach ridge and dune complex with good radiocarbon control sampling the last 3500 radiocarbon years B.P. provides new insights on the early genesis of clay bands in sandy soils. Soil profiles were sampled by age groups, described in the field, and then subjected to laboratory analyses for particle-size distribution, pH, organic carbon, carbonate minerals, and extractable iron and manganese. This study suggests that small increases in pH, brought about by small increases in carbonate content within the soil profile, are responsible for flocculating small amounts of illuviated clay. This process, along with a transition to a greater hydraulic conductivity with soil depth due to coarser textures in any given profile, partly explains the existence and possible reason for the initiation of illuvial zones and eventually for clay-band horizons. A pronounced increase in the thickness of incipient clay-band horizons in soils older than 2300 years appears due to finer textures in the parent materials than are present in younger soils. Because of slightly reduced porosity and lower permeability, carbonates and a high pH are retained in both illuvial and eluvial horizons of some of these older soils. In addition, only in those profiles older than 2300 years do clay and iron oxide concentrations coincide and is there some suggestion of greater amounts of extractable manganese in horizons of minimum iron and clay. A pronounced segregation of clay-iron bands is not apparent at the study area but should occur in future years as additional amounts of iron and clay are deposited.     

Density fractionation of organic matter in dolomite‐derived soils

Dolomite (CaMg(CO₃)₂) constitutes half of the global carbonates. Thus, many calcareous soils have been developing rather from dolomitic rocks than from calcite (CaCO₃)‐dominated limestone. We developed a physical fractionation procedure based on three fractionation steps, using sonication with subsequent density fractionation to separate soil organic matter (SOM) from dolomite‐derived soil constituents. The method avoids acidic pretreatment for destruction of carbonates but aims at separating out carbonate minerals according to density. The fractionation was tested on three soils developed on dolostone parent material (alluvial gravel and solid rock), differing in organic‐C (OC) and inorganic‐C (IC) concentrations and degree of carbonate weathering. Soil samples were suspended and centrifuged in Na‐polytungstate (SPT) solutions of increasing density, resulting in five different fractions: two light fractions < 1.8 g cm^–3 (> 20 μm and < 20 μm), rich in OC and free of carbonate, and two organomineral fractions (1.8–2.4 g cm^–3 and 2.4–2.6 g cm^–3), containing 66–145 mg g^–1 and 16–29 mg g^–1 OC. The organomineral fractions consist of residual clay from carbonate weathering such as clay minerals and iron oxides associated with SOM. The fifth fraction (> 2.6 g cm^–3) was dominated by dolomite (85%–95%). The density separation yielded fractions differing in mineral compositions, as well as in SOM, indicated by soil‐type‐specific OC distributions and decreasing OC : N ratios with increasing density of fractions. The presented method is applicable to a wide range of dolomitic and most likely to all other calcareous soils.     

我国土壤放射性碳年龄

我国地域辽阔 ,土壤类型众多。土壤中不仅含有有机质和腐殖质 ,而且许多土壤中还含有丰富的钙质结核、分散碳酸盐以及贝壳、珊瑚等可供放射性碳断代的良好对象。土壤有机和无机1 4 C年龄研究表明 ,我国大多数土壤是全新世时期的产物。其中又以全新世中期和晚期的土壤占绝对优势。人为土纲中的土壤年龄与六千余年来我国悠久的农业耕种历史密切相关。相比之下 ,只有少数土壤形成于晚更新世晚期。而另一些土壤有数个形成、发育阶段 ,它们的年龄自然亦就跨越不同的地质时期 ,具有多元化的特点。     

DOWNWARD MOVEMENT OF DOLOMITE,KIESERITE OR A MIXTURE OF CaCO3 AND KIESERITE THROUGH THE UPPER LAYERS OF AN ACID FOREST SOIL

Liming and fertilization are important tools for improving the chemical status of acid, base poor forest soils. The downward movement of dolomite, kieserite and a mixture of CaCO₃ and kieserite was investigated by monitoring the leachates and exchangeable cation composition from single and combined horizon columns, reconstructed from an acid brown forest soil profile (0–15 cm). Upon entering the soil, Mg ions from kieserite displaced base cations and acidity (H and Al ions) from exchange sites, which subsequently moved down with the mobile SO₄ ^2- anions. Total leaching during the initial SO₄ ^2- pulse was similar with the CaCO₃ + kieserite mixture. Compared to the single kieserite treatment, the joint application of CaCO₃ greatly increased the proportion of Ca in the leachates from all horizons. It also decreased the leaching of acidity from the surface Oe horizon and prevented pH from dropping under this layer. With both treatments, the redistribution of magnesium with SO₄ ^2- anions resulted in a rapid increase in exchangeable Mg contents throughout the studied columns. Due to the important charge increase in the Oe horizon and to kinetic restraints imposed on dissolution, downward movement of Ca and Mg ions from dolomite was very limited. Mg was however much more mobile than Ca. In the CaCO₃ + kieserite and dolomite treatments, the migration of alkalinity and base cations with time was associated with a decrease in exchangeable acidity and an increase in ECEC in the two upper soil layers. By the end of the monitoring period, overall net Mg retention in the 0–15 cm columns increased in the order kieserite < CaCO₃ + kieserite << dolomite with respectively 20, 35 and 85% of cumulated inputs remaining in the columns. The corresponding net Ca retention amounted to 82 and 96% of cumulated inputs for the CaCO₃ + kieserite and dolomite treatments, respectively. Results from this study complement those obtained in the field by clearly demonstrating the mechanisms involved in the downward movement of some fertilizers commonly used to increase the base saturation of acid forest soils.