Skeletal Carbonate Mineralogy of Brachiopods: Smith, Freeman, Dixon-Anderson and Lee

Published: 31 March 2022| Version 1 | DOI: 10.17632/j74twbhjxv.1
Contributors:
,
,
,

Description

Here we combine published and new mineralogical data on brachiopods from all over the world, including most major taxa, to investigate patterns in and controls on brachiopod carbonate mineralogy. Measurements of 1711 specimens in 147 species included 253 fossil specimens: living species range from 79°N to 74°S and from intertidal to almost 4000 m deep. Calcareous brachiopods create strong and resilient valves, formed of very low-Mg calcite (χ ̅ =1.3 wt% MgCO3 in the phylum). The Craniida (χ ̅ = 8.9 wt% MgCO3) and Thecideida (χ ̅ = 6.5 wt% MgCO3) are unusual in precipitating calcite with higher Mg content; these taxa are unusual in that they cement themselves to the substrate. This is the first study to find bimineralic brachiopods; a few species show a combination of low-Mg (χ ̅ = 0.8 wt% MgCO3) and intermediate-Mg calcite (χ ̅ = 7.2 wt% MgCO3). While Mg in calcite varies systematically among valve layers and sometimes along the growth axis, we found no consistent difference between valves of the same individual. A weak latitudinal signal indicates some overall temperature control of Mg, but in general brachiopods are strongly controlling, precipitating low-Mg calcite even when Mg:Ca ratios is seawater are high. The drivers and influences on brachiopod mineralogy range from individual to environmental to phylogenetic – the resulting variability is complex. Assuming simple mineralogy in brachiopods for paleoceanographic studies may need to be reexamined.

Files

Steps to reproduce

Each valve was photographed, scraped clear of any encrustation, soaked for 2-3 hours in 5% household bleach (if collected alive), rinsed and air-dried prior to sectioning along a hinge-to-margin axis. One half of each valve was labeled and archived. The other half was hand-ground to fine powder in an agate mortar along with 0.1 g NaCl as an internal standard for X-ray diffraction (XRD). 0.5 g samples of shell powder were mixed with ethanol and spread on glass slides to dry. Each slide was run through a Phillips X-ray diffractometer at 50 counts per degree with a count time of 1 sec, over the range of 26 to 33 °2. Peak heights (in counts) and positions (in °2) were determined, the halite peak position was standardized to 31.72 °2, and other peak positions corrected. Weight percent Mg in calcite was calculated from calcite peak position (in °2) after Chave (1967) (y = 30x -882). Each spectrum was visually inspected as well, and locations of ragged or asymmetrical peaks were confirmed. Relative peak height (ht) counts of aragonite (A1 at 26.213 °2 and A2 at 27.216 °2) and calcite (C1 at 29.4 to 29.8 °2) were used to calculate Peak Height Ratio (PHR) for each graph: PHR = (ht A1 + ht A2)/(ht A1 + ht A2 + ht C1). Wt% calcite was calculated using the machine-specific calibration of Gray and Smith (2004): Wt% Calcite = 80.4 (PHR)2 – 180.9 (PHR) +101.2. In the case where two distinct calcite peaks were seen, relative proportions were determined using the two-calcite calibration of Smith & Lawton (2010): Wt% High-Mg Calcite = (111.5 * CPHR) -1.03, where CPHR is ratio of the two calcite peak heights (High/Low+High).

Institutions

University of Otago

Categories

Geochemistry, Brachiopoda, Carbonate Mineral

Licence