Complex formation and coordination of Fe in a solid soil sample, as well as in several iron containing solid phases, was qualitatively determined, using Fe K-edge extended X-ray absorption fine structure (EXAFS) spectroscopic measurements. Samples investigated were an untreated Oa horizon from a Haplic Podzol in Asa southern Sweden, lab-prepared goethite and two limonite samples from different locations. In addition, EXAFS data for two iron(III) spiked Oe samples will be presented. The measurements were performed at the wiggler beam line 4-1 at Stanford Synchrotron Radiation Laboratory (SSRL), USA or at wiggler beam line I811 at MAX-Lab, Lund University, Sweden Evaluation of the Asa Oe sample shows three significant contributions, with the most dominant being inner core Fe-O/N bond distances at 2.00(2) Å (using the EXAFS technique it is not possible to distinguish between O and N), indicating that iron is present as iron(III) and basically has a octahedral configuration; corresponding bond distances for iron(II) are ca. 0.10 Å longer. A second contribution is a second shell Fe-O-C scattering at 2.90(4) Å, indicating organic groups, such as phenols or carboxylates, to be bound to the iron(III). The third contribution found is a Fe--Fe distance at 3.39(3) Å, which unambiguously shows that the dominating iron species in the soil sample is a dimer or complex with higher degree of polynuclearity. Furthermore, this shows that iron(III) humic acid complexes in a soil with neutral pH and under aerobic conditions are hydrolyzed. Further refinements were derived from a common form of iron(III) complexes with ligands, (O₅Fe)₃O, where a central oxide ion binds to three iron(III) ions. In this complex, the Fe-O_oxo bond is shorter than the remaining five. The thus-suggested model increased the fit of the data and is in very close agreement with crystal structures containing the complex [(H₂O)₅Fe-O-Fe(OH₂)₅]⁴⁺, and other various complexes with a (O/N)₅-Fe-O-Fe-(O/N)₅ core. In addition, modeling results for a metal binding studies described in another paper improved significantly when allowing the suggested dimeric binding. Preliminary evaluation of the solid samples clearly shows the presence of a dominant inner core Fe-O/N interactions around 1.97-2.00 Å in all samples, as was found for the soil sample. For the two limonite samples, second scattering shell Fe--Fe interactions were found around 3.0 – 3.1 Å, as well as around 3.4 – 3.5 Å. The first distance suggests the presence of complexes with a double hydroxo bridge, whereas the latter indicates the presence of a single μ-oxo bridge, as found in the soil sample. In the goethite sample an additional Fe--Fe distance was found around 2.7 Å, as well as a Fe--O distance around 3.8 Å; goethite can therefore be excluded as a major iron species in the studied samples. A more detailed data treatment will be presented during the conference.