Summary Copper sulphate (Bordeaux mixture) has been used as a fungicide against mildew in vineyards for more than a hundred years. This treatment has resulted in significant Cu accumulation in soils (from 100 to 1500 mg kg−1). It is desirable to determine the distribution of Cu in these soils to predict if this element is potentially toxic. Several speciation methods have been compared: sequential and single–step extractions, analytical transmission electron microscopy (ATEM, ASEM) and physical fractionation to study a profile of a vineyard acid soil in Beaujolais (France). Physical fractionation showed that copper is concentrated in the coarse organic fractions associated with plant residues and in the fine clay fraction. Analytical electron microscopy showed a great diversity of fixation sites (bacteria, amorphous organic and mineral compounds). Chemical sequential extractions showed that after the sequential extraction treatments, 60% of Cu was not extracted. Extraction data showed that in the case of an acid sandy soil, sequential chemical extraction did not seem to be sufficiently selective to speciate copper. In the single–step extractions study, the hydroxylamine treatment was the most selective. In the nonselective cases several phenomena may be responsible for the failure, the analytical electron microscopy study showed copper redistribution after certain single–step extractions were carried out. After comparing three different distribution methods we conclude that the distribution of this element in acid soils should be investigated by several analytical methods to establish the true speciation.

No abstract

The morphology of pure microbial extracellular polysaccharides (EPS) and of their association with a mineral was studied by scanning electron microscopy. Several methods were compared: cryo‐SEM, critical point dehydration (CPD), and TEM observation after solvent exchanges and inclusion in a resin. The EPS were observed as fibrils, interconnected into a network. Their dimensions were compared with literature data on the conformation and shape of the given EPS. It was deduced that these EPS have an undoubtably fibrillar structure when hydrated, and that some aggregation or collapse artifacts took place in both CPD or cryo‐SEM samples but were quite limited. Cryo‐SEM appears then as a relevant tool for the study of the secretion of EPS by microorganisms in soils.

The occurrence of 2:1 phyllosilicates in Andisols is variously ascribed to in situ pedogenic origin, aeolian addition, or the presence of hydrothermally altered rock fragments. We studied the origin of 2:1 phyllosilicates that occur in Holocene Hapludands on andesitic, sandy beach ridges in Costa Rica by micromorphological, mineralogical, and submicroscopical techniques. The 2:1 phyllosilicates also occur as pseudomorphs after primary minerals in fresh rock of the inland volcanoes, from which the parent material of the beach ridges was mainly derived. Hydrothermal processes are most likely responsible for the formation of such pseudomorphs. Rock weathering produces sand‐sized rock fragments with clay pseudomorphs and also liberates individual pseudomorphs. Subsequent erosion and alluvial transport affect their shape, but not their internal fabric. In the beach ridges, clay pseudomorphs appear as individual, sand‐sized clay bodies, and inside sand‐sized andesitic rock fragments. Submicroscopical analyses of these individual clay bodies and andesitic rock fragments with clay pseudomorphs indicate a predominance of 2:1 phyllosilicates. This implies that they are inherited from the parent material and are not due to postdepositional soil formation in the beach ridges. Weathering and biological activity affect the clay bodies and rock fragments with clay pseudomorphs, leading to the formation of clay‐sized particles consisting of 2:1 phyllosilicates. Toward the soil surface, these particles are incorporated into the allophanic groundmass resulting from actual soil formation. The geographically extensive occurrence of 2:1 phyllosilicates in Andisols suggests that the genetic processes described here may have more than regional validity.

Abstract--Transmission electron microscopy (TEM) and analytical electron microscopy (AEM) methods were used to study the crystal chemistry of phyllosilicates occurring in green grains of Miocene sediments from the Congo continental shelf. Using diagrams based on wt. % K and the (Fe + Mg)/A1 ratio, minerals were distinguished from mixed-layer phases. The most abundant detrital mineral is Fe-kaolinite. The morphology and composition identify this mineral as a component of ferralitic soils. This Fe-rich kaolinite has undergone a complex process of partial dissolution and recrystallization and further enrichment in Fe and, to a lesser extent, in Mg in the marine environment. The detrital mica observed with TEM retains the original morphology and chemistry of muscovite. Alteration processes resulted in the crystallization of 1:1 trioctahedral Fe 2+ and Mg-rich minerals and interstratified phases with 1:1 and 2:1 layers in varying proportions observed with the aid of hi~h-resolution transmission electron microscopy (HRTEM) imaging. Included among the newly formed 7-A phases are those apparently containing excess Si. The smectites are apparently neoform, and chemical analyses showed that these marine K-smectites belong to the beidellite-nontronite series and have tetrahedral substitutions similar to muscovite. Their compositions are closer to beidellite than to nontronite, although the latter was observed in association with goethite. The TEM observations and crystallochemical data show that mineral alteration ceased after forming mixedlayer minerals, and alteration did not reach the glauconitization stage. Apparently, the Miocene assemblages experienced rapidly changing environmental conditions and high sedimentation rates that continue today.