Microstructural data of six recent brachiopod species: SEM, EBSD, morphometric and statistical analyses

Here, we provide the dataset associated with the research article “Mapping of recent brachiopod microstructure: A tool for environmental studies” [1]. We present original data relative to morphometric and statistical analyses performed on the basic shell structural units (the secondary layer fibres) of brachiopod shells belonging to six extant species adapted to different environmental conditions. Based on SEM micrographs of the secondary layer, fibres from ventral and dorsal valves, and from different shell positions, showing regular and symmetrical cross sectional outlines, were chosen for morphometric measurements using Adobe Photoshop CS6, Image-Pro Plus 6.0 and ImageJ. To work out the reliability of the measurements, the most significant parameters were tested for their probability density by distribution plots; for data visualization and dimension reduction, principal component analysis (PCA) was performed using R 3.3.0 [2] and independent-samples t-tests were performed using SPSS Statistics (IBM Version 22.0. Armonk, NY). Besides a quantitative analysis, a qualitative description of the shell microstructure is provided by detailed SEM imaging and EBSD measurements.

Brachiopod shells were embedded in epoxy resin (not all), cut along the longitudinal (or transversal) axis, and immersed in 36 volume hydrogen peroxide (H 2 O 2 ) for 24 hours to remove organic matter. Sectioned surfaces were smoothed with silicon carbide (SiC) powder, etched with 5% hydrochloric acid (HCl) for 3 seconds, and then rinsed in deionised water and dried [3]. Then they were: 1) gold coated for SEM analysis; 2) mechanically grinded and polished down to a grain size of 1 μm, etch-polished with colloidal alumina (particle size0 .06 μm) in a vibratory polisher and coated with 4-6 nm of carbon for EBSD analysis.

Experimental features
Morphometric measurements and analysis of fibres of the secondary layer based on SEM micrographs, EBSD and statistics (distribution plots, principal component analysis and independent-sample t-tests These data provide a quantitative and qualitative description of the microstructure of recent brachiopod shells using several tools: SEM, EBSD, morphometric and statistical analyses.
These methods may be applied to other invertebrates and to fossil shells to objectively describe and compare their microstructures.
These data are valuable to researchers investigating invertebrate biomineralization patterns.

Data
Brachiopod calcite shells are high resolution biomineral archives used to reconstruct global marine environments in the recent and deep past [4][5][6][7][8][9][10]. Biominerals, the hard parts produced by organisms for support and protection, are one of the best tools to use, as they are high-resolution archives of the environmental conditions prevailing during their growth. Here, we focus on the basic structural units (fibres) of the secondary calcite layer of six recent rhynchonelliformean brachiopods. Based on SEM and EBSD analyses, 1197 morphological measurements of the fibres were performed and statistically analyzed, comparing the size and shape of the fibres in different valves of the same specimen, at different positions within the valve, in different shell layer successions, in different species and in different environmental conditions.

Sample collections
Six extant rhynchonelliformean brachiopod species (21 adult specimens) were chosen for microstructure analyses (Table 1). They have either a two-layer shell sequence or a three-layer shell sequence, both comprising a fibrous secondary layer, and are adapted to different environmental conditions, from Signy and Trolval Islands, Antarctica, to Doubtful Sound and kaka point, New Zealand to the Tuscan Archipelago, Mediterranean Sea.

SEM
We followed SEM sample preparation as suggested by Crippa et al. [3]. The specimens were embedded in a transparent bicomponent epoxy resin and cut along the longitudinal (or transversal) axis using a low speed saw with a thin diamond blade. To remove the organic matter within the shell,

EBSD
For EBSD measurements brachiopod shells were embedded in epoxy resin and were cut along and perpendicular to the median plane of the investigated shells. Surfaces of the embedded specimens were subjected to several sequential mechanical grinding and polishing steps down to a grain size of 1 μm. The final polishing step was carried out with colloidal alumina (particle size~0.06 μm) in a vibratory polisher. Sample surfaces were coated with 4-6 nm of carbon. EBSD measurements were carried out at the Department of Earth and Environmental Sciences, LMU Munich, Munich, Germany, on a Hitachi SU5000 field emission SEM, equipped with a Nordlys II EBSD detector and AZTec acquisition software. The SEM was operated at 15 and 20 kV; measurements were evaluated with CHANNEL 5 HKL software [11,12]. EBSD data are presented as band contrast measurement images, a grey scale component that gives the signal strength of the EBSD Kikuchi diffraction pattern in each measurement point. Accordingly, the strength of the diffraction signal is high when a mineral is detected whereas it is weak or absent when a polymer is scanned. A high diffraction signal is shown with light, while a weak signal is visualized with dark grey colors in the band contrast measurement image. Plate 5 shows EBSD band contrast measurement images of two layer shells (L. uva, C. incospicua, M. sanguinea, N. nigricans).

Statistical analyses
Based on SEM micrographs, each fibre, with regular and symmetrical cross sectional outline, was chosen for morphometric measurements (1197 measurements) from different ontogenetic stages; fibres were first outlined using Adobe Photoshop CS6, and then all parameters (e.g. Max Feret diameter, Min Feret diameter, Area, Perimeter, Convex area and Convex perimeter) were measured by Image-Pro Plus 6.0 and ImageJ.
The frequency distribution plots of the most significant parameters (Area, Perimeter, Max Feret diameter, Convex Area) were calculated and drawn by Excel 2013 (FREQUENCY function and NORM. DIST function) (Figs. 1-3 Based on the six measured parameters, five shape descriptors were calculated: Formfactor (circularity, 4π × Area/Perimeter 2 ), Roundness (4Area/π × Max Feret diameter 2 ), Aspect Ratio (Max Feret diameter/Min Feret diameter), Convexity (Convex Perimeter/Perimeter), and Solidity (Area/Convex Area) [14]. For data visualization and dimension reduction, principal component analysis (PCA) was performed on the five shape descriptors using R 3.3.0 (Figs. 4-6) [2]. We used the function prcomp for      Table 3 T-test of fibres size and shape data of the ventral valve vs the dorsal valve in different positions of the shell (pe: posterior external; cm: central middle; ai: anterior internal). Significant values (p-value r 0.05) are marked in bold style.  Table 5 T-test of fibres size and shape data in different positions of the ventral valve. See captions of Fig. 5 and Table 2 for the legend. Significant values (p-value r 0.05) are marked in bold style.
principal component analysis and fviz_pca_biplot for plot; the biplots were created using the package factoextra [15]. Independent-sample t-tests were performed using SPSS Statistics (IBM Version 22.0. Armonk, NY) (Tables 2-9). A p-value r 0.05 is considered significant.

Funding
This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 643084.