Konservat-Lagerstätten, such as the Toarcian (Early Jurassic) Posidonia Shale of southwestern Germany, are renowned for their spectacular fossils. Ichthyosaur skeletons recovered from this formation are frequently associated with soft-tissues; however, the preserved material ranges from three-dimensional, predominantly phosphatized structures to dark films of mainly organic matter. We examined soft-tissue residues obtained from two ichthyosaur specimens using an integrated ultrastructural and geochemical approach. Our analyses revealed that the superficially-looking ‘films’ in fact comprise sections of densely aggregated melanosome (pigment) organelles sandwiched between phosphatized layers containing fibrous microstructures. We interpret this distinct layering as representing condensed and incompletely degraded integument from both sides of the animal. When compared against previously documented ichthyosaur fossils, it becomes readily apparent that a range of preservational modes exists between presumed ‘phosphatic’ and ‘carbonized’ soft-tissue remains. Some specimens show high structural fidelity (e.g., distinct integumentary layering), while others, including the fossils examined in this study, retain few original anatomical details. This diversity of soft-tissue preservational modes among Posidonia Shale ichthyosaurs offers a unique opportunity to examine different biostratinomic, taphonomic, and diagenetic variables that potentially could affect the process of fossilization. Soft-tissue preservation in the Posidonia Shale likely was regulated by a multitude of factors, including decay efficacy and speed of phosphatic mineral nucleation; these in turn were governed by a seafloor with sustained microbial mat activity fuelled by high organic matter input and seasonally fluctuating oxygen levels.

Photography and light microscopy 

Photographs of fossils were taken with a Pentax WG-III camera. Both untreated and demineralized fossil samples were examined under an Olympus SZX16 microscope equipped with an Olympus SC30 digital camera. 

Transmission electron microscopy (TEM) 

Demineralized fossil matter was dehydrated in ethanol and then stepwise embedded in epoxy resin/ethanol mixtures, with the last step being pure epoxy. Embedded samples were allowed to polymerize at room temperature for 72 hrs and then placed in an oven at 60˚C for 48 hrs. Thin sections were made using a Leica EM UC7 ultramicrotome. A glass knife was used to produce semi-thin (200 nm) sections for initial quality control. Ultrathin (50 nm) sections were subsequently prepared using a diamond knife, and the sections were then mounted on pioloform-coated copper grids. TEM images were acquired using a JEOL JEM-1400 PLUS instrument at 80 and 120 kV with a bottom-mounted Matataki CMOS camera at the Department of Biology, Lund University, Sweden.

Synchrotron rapid-scanning X-ray fluorescence (SRS-XRF)

X-ray fluorescence was conducted at beamline 10-2, Stanford Synchrotron Radiation Lightsource, USA. Incident X-ray energy was set to 11 keV using a Si (111) double-crystal monochromator with the Stanford Position Electron Accelerating Ring storage ring containing 500 mA at 3.0 GeV. Fluorescence lines of the targeted elements and the intensity of the total scattered X-rays were observed using a silicon drift Vortex detector (SII NanoTechnology). A focused 100 × 100 μm beam was filtered through a 100 μm Ta pinhole aperture. Incident and transmitted X-ray intensities were measured with nitrogen-filled ion chambers, with the incident X-ray flux measured at around 8 × 10¹⁰ ph/s. The fossils were mounted on a vertically oriented aluminium bracket at 45˚ relative to the incident X-ray beam. The fluorescence detector was mounted at 90˚ in relation to the incident beam and was held at a distance of ~4 cm from the specimens at ambient atmospheric conditions. Data were collected by spatially rastering the X-ray beam over the mounted fossils as the stage was in continuous motion. Fluorescence signal was gated by an encoder signal at pixel sizes of 100 or 200 μm with beam exposure time at 20 ms per pixel. Image and data processing were completed on Sam’s Microprobe Analysis Toolkit (SMAK; Webb 2011).

Sam's Microprobe Analysis Toolkit (SMAK) for raw XRF Data