@WFS,World Fossil Society,Riffin T Sajeev,Russel T sajeev

Citation: Kim J-K, Kwon Y-E, Lee S-G, Kim C-Y, Kim J-G, Huh M, et al. (2017) Correlative microscopy of the constituents of a dinosaur rib fossil and hosting mudstone: Implications on diagenesis and fossil preservation. PLoS ONE 12(10): e0186600. https://doi.org/10.1371/journal.pone.0186600

Editor: Dong Hoon Shin, Seoul National University College of Medicine, REPUBLIC OF KOREA

The Boseong fossil site is a rich Upper Cretaceous fossil site from South Korea based on the discovery of abundant dinosaur eggs and egg clutches [1–3], post-cranial skeletal remains of the small basal ornithopod Koreanosaurus boseongenesis [4], and partial skeletal remains of the large anguimorph lizard Asprosaurus bibongriensis [5]. While it has been presumed that rapid burial took a key role in preserving the fossils from the local region based on large-scale depositional features [2], deeper insights in specific factors that may have contributed to bone preservation has not been thoroughly explored. Analytical studies on both skeletal fossil material and the hosting geological matrix provide insights into the interaction and relationship of composing elements [6,7]. Chemical analysis on a microscopic scale of fossil bone allows further evaluation of elements that have native and/or external origins [7–12]. Although the general nanostructure of fossilized bone can be investigated through novel X-ray techniques [13], transmission electron microscopy (TEM) analysis directly provides information on the morphology, arrangement, and chemistry of fossil apatite nanocrystals [9,14–18]. Such data are crucial for understanding how these apatitic crystalline phases have originated and also how the bone structure was maintained during fossilization. It should be noted that TEM data is based on an extremely small and localized scale, and such shortcomings of TEM investigation can be mitigated through correlative microscopy [17–19]. We have previously investigated structural and chemical features at micro- to nanoscale of a dorsal rib portion from Koreanosaurus [17] obtained from the holotype specimen (referred to as KDRC-BB2: Korea Dinosaur Research Center-Boseong Bibong 2) which consist mainly the “torso” region discovered in an articulated state in a large mudstone block that may have originated directly from the main outcrop of Site 5 [4]. In this study, the distal region of a fully preserved seventh left dorsal rib bone was obtained from KDRC-BB2, and we have specifically selected it based on the following reasons; i) information on exact original position, ii) simple morphology, and iii) intact hosting mudstone (S1 Fig).

The section was divided into three regions–hosting mudstone (yellow arrow), boundary (red arrow), and rib bone (blue arrow). Clusters of calcite microcrystals can be directly observed in all regions. The mudstone and boundary region primarily contains detrital clasts of quartz and feldspars. Due to the compressed nature of the bone matrix, specific osteohistological features were not discernible from the rib bone besides the vascularization pattern.

Here, through correlative microscopy techniques, we aimed to investigate the microstructure and chemistry of key constituting phases of the fossil rib bone, hosting mudstone, and the boundary in-between. We evaluated the distribution and interaction of the key phases between these regions with focus on unraveling features involved in diagenesis and bone preservation. The frequent occurrence of “platy phases” within the rib bone matrix from our initial study was a compelling feature [17], and we intended to fully reveal the identity and origin of these phases and their presumed role in bone preservation. These phases may have affected the arrangement of apatite crystals from the rib bone, and without TEM investigation, such assumptions cannot be thoroughly assessed. Although our research sample represents only a small fraction of the entire skeletal fossil and fossil site, the preservation of the brittle and highly porous dorsal rib bones of Koreanosaurus was intriguing. We also considered it as an ideal research material without inflicting considerable damage to the holotype specimen. Due to the small size and fractured state of the rib bones, correlating microstructural and nanostructural preservation of osteohistological features were mainly performed on the larger and more intact femoral bones from the paratype specimens [4,18].

Correlative microscopy is a technique for performing progressive structure and chemical analyses on specimens from macro- to nanoscale and involves the use of optical and electron microscopes. The key importance of this technique is that a specific region of interest observed from the optical thin section should be reexamined in detail by electron microscopes in higher resolution. A crucial step in correlative microscopy is the use of proper and effective sample preparation methods for the corresponding analytical procedures. Typically, two types of samples—optical thin sections and nanopowders—are prepared directly from the bulk specimen. As conventional powder preparation methods result in an inevitable loss of spatial information, in this study we employed an ultrasonic drilling device (S2A Fig) which allowed us to acquire very small amounts of powders (in ng range) from predetermined specific regions either from the bulk specimen or optical thin section. We also designed an ultrasonic spraying device (S2B Fig) that is capable of dispersing powder samples on a TEM grid via a masking tool. This device is derived from a previously developed multi-sample loading device by our research team [20], which can load up to four samples on a single TEM grid using more conventional methods. We also utilized half-masking techniques on TEM grids by coating Au nanoparticles as a standard on half of the grid. For acquiring TEM samples from optical thin sections, we employed focused ion beam (FIB) milling. The biggest advantages are its precision and capability in preparing TEM samples from various directions [21]. The following methods were used in this study. 1) Initially, we used a stereoscopic zoom microscope to observe the bulk sample before carrying out specific preparation procedures. We also used the microscope to evaluate the overall sample quality after preparation using reflected light. 2) We used a polarizing optical microscope to identify mineral constituents of the hosting mudstone, and to discern the microstructural features of the rib bone. 3) We carried out initial phase identification by X-ray diffraction (XRD) analysis on both the optical thin sections and powder samples. 4) An electron probe microanalyzer (EPMA) equipped with wavelength dispersive spectroscopy (WDS) was primarily used for chemical mapping of optical thin sections to characterize the distribution of distinctive and common elements from the rib bone, hosting mudstone, and boundary region. 5) Scanning electron microscopy (SEM) imaging and chemical analysis with energy dispersive spectroscopy (EDS) was performed for investigating microstructures and elemental distributions from both thin sections and powder samples in general, and for characterizing the constituting phases. 6) We used TEM as the main technique for evaluating the specific structure and chemistry of the identified phases. For electron crystallographic analysis, we obtained and analyzed selected area electron diffraction (SAED) patterns and high-resolution TEM (HRTEM) images along with corresponding fast Fourier transform (FFT) data. We mainly carried out TEM–EDS analysis to identify the distribution of specific clay phases from the FIB-milled samples. All microscopes were equipped with charge-coupled device (CCD) cameras, and we captured and stored the micrographs directly using the respective software of each CCD camera.

@WFS,World Fossil Society,Riffin T Sajeev,Russel T sajeev