Zinc is an essential micronutrient for plants and animals, but can also become a potentially toxic metal contaminant in soils at high concentrations. The bioavailability of Zn to soil organisms and plants cannot be predicted based on total concentrations, because it strongly depends on the chemical speciation of Zn in the soil. Recent studies have demonstrated that Zn sorbed to clay minerals at neutral or slightly alkaline pH is slowly incorporated into newly formed mineral structures, such as layered double hydroxides (Zn-LDH) or phyllosilicates containing Zn in octahedral coordination. However, little information is presently available about long-term speciation changes of Zn in soils under field conditions. In this study, we investigated the speciation of Zn in three oolitic Jurassic limestones of the Swiss Jura mountain range and in the soils developed from these rocks in order to obtain new information about the long-term fate of Zn in soils. The rock and soil samples were analyzed by X-ray fluorescence analysis (XRF) for total elemental composition. The speciation of Zn was characterized using a combination of synchrotron micro-X-ray fluorescence (µ-XRF) and micro-X-ray absorption fine structure (µ-XAFS) spectroscopy (Beamline 10.3.2, Advanced Light Source, Berkeley). XAFS spectra were analyzed by linear combination fitting based on an extensive set of reference spectra of known Zn species. In addition, total Zn in the soil samples was fractionated using a classical six-step sequential extraction method. The rocks contained between 43 and 207 mg/kg total Zn. In the soils, total Zn concentrations ranged from 237 to 864 mg/kg, clearly exceeding the Swiss guideline value of 150 mg/kg. These Zn concentrations are considered to be lithogenic rather than a result of environmental pollution. During soil formation from limestone, trace metals including Zn are enriched due to preferential dissolution and leaching of calcium carbonate and residual enrichment of less mobile elements. XAFS spectroscopic results indicated that Zn in two of the three limestones (Gurnigel and Schleifenberg) is present mainly as Zn-substituted goethite. A small fraction of total Zn in both rocks was present as sphalerite (ZnS). These results were supported by extraction of powdered rock samples with 1M NH4-acetate solution (pH 6), which suggested that most of the Zn was bound in non-carbonaceous components not extractable with NH4-acetate. In contrast, XAFS results and extraction with NH4-acetate showed that nearly all Zn in the third limestone (Dornach) was present as Zn substituting for Ca in the calcite structure. The Gurnigel and Schleifenberg soils contained considerable amounts of Zn-bearing goethite, probably stemming from the parent rock, and smaller amounts of newly-formed Zn species, most likely Zn-containing phyllosilicates with Zn bound in octahedral sheets. These results suggest that Zn in goethite was very stable during pedogenesis. Sphalerite was detected in the soils only in minute quantities. For both soils, the results of sequential extractions confirmed that Zn was associated mainly with iron oxides and a residual mineral fraction, which included sphalerite and phyllosilicates. In contrast, XAFS analysis of the Dornach soil suggested that Zn is primarily bound in octahedral sheets of phyllosilicates. This was supported by the sequential extraction results, exhibiting the largest fraction of Zn in the residual fraction. In conclusion, our results demonstrate that Zn-substituted goethite occurring in limestones is very stable during pedogenesis, while Zn bound in carbonates is released and subsequently incorporated into newly formed minerals. Sphalerite also dissolves during pedogenesis, but was still detected in very small quantities in the soils. The long-term speciation of Zn in soils developed from limestones therefore depends on the Zn speciation in the parent rocks.