Introduction
The Poleta Formation (Cambrian Stage 3–Stage 4) at Indian Springs Canyon, in the northern Montezuma Range, Esmeralda County, Nevada, hosts a biota of exceptionally preserved fossils (a Konservat-Lagerstätte) including brachiopods with preserved nonbiomineralized soft parts (English and Babcock, Reference English and Babcock2010; Hollingsworth and Babcock, Reference Hollingsworth and Babcock2011). The discovery of exceptionally preserved brachiopods from the Indian Springs site was first reported by Babcock et al. (Reference Babcock, Hollingsworth, Peel and Rees2000). Until now, however, no detailed description of the purported stem brachiopod specimens referred to the genus Mickwitzia from this locality has been published. The presence of exceptionally preserved soft-part anatomy is particularly of interest in resolving the phylogenetic placement of these problematic stem brachiopods. Thus, a detailed description of the fauna allows us to both further elucidate the origin and affinity of Cambrian stem brachiopods and to determine the exact taxonomic identity of the specimens from this area. Taphonomic factors have been a particular problem in previous systematic treatments of Mickwitzia (Jensen, Reference Jensen1993; Nemliher, Reference Nemliher2001; Skovsted and Holmer, Reference Skovsted and Holmer2003; Balthasar, Reference Balthasar2004). With Mickwitzia specimens from both carbonate and shale lithologies, we attempted to comparatively address some aspects of this confounding taphonomic alteration with the newly described collection. For example, shell structure, which has conventionally been used as a diagnostic feature of these organisms, can be subject to diagenetic alteration, affecting or obliterating features otherwise systematically useful.
In terms of the wider significance of the described material, mickwitziids seem to occupy a unique phylogenetic position (Skovsted et al., Reference Skovsted, Brock, Paterson, Holmer and Budd2008, Reference Skovsted, Streng, Knight and Holmer2010), potentially linking tommotiids on the brachiopod stem with crown linguliform brachiopods. This is based on the presence of homologous characters including columnar shell ultrastructure and shell penetrating setae (e. g., Skovsted et al., Reference Skovsted, Brock, Paterson, Holmer and Budd2008, Reference Skovsted, Streng, Knight and Holmer2010; Balthasar et al., Reference Balthasar, Skovsted, Holmer and Brock2009). Here we document a new collection of Mickwitzia, referred to M. occidens Walcott, Reference Walcott1908 from the Indian Springs Lagerstätte in Nevada (English and Babcock, Reference English and Babcock2010; Hollingsworth and Babcock, Reference Hollingsworth and Babcock2011). The new material of Mickwitzia consists of both acid-etched shell fragments from limestone and whole specimens from shale horizons. In addition to the description of the new specimens, they are also compared with the type of Mickwitzia occidens.
The phylogenetic position and evolutionary significance of upper stem-group lophotrochozoans, in particular the mickwitziids, has been the subject of extensive debate and discussion (Holmer et al., Reference Holmer, Skovsted, Brock, Valentine and Paterson2008b; Skovsted et al., Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Petterson Stolk and Paterson2009b), much of which remains open. One aspect of this debate concerns the evolutionary origins of species assigned to Mickwitzia. Mickwitzia could represent either a clade or grade of similar but unrelated stem lophotrochozoans. Careful description of the new collections gives us insights into the morphological features, in particular homologous shell microstructures, which unite the mickwitziids as a biological and taxonomically valid unit.
In order to help elucidate the relationships of these stem brachiopods, the type material of Mickwitzia monilifera (Linnarsson, Reference Linnarsson1869) from Sweden, and M. muralensis Walcott, Reference Walcott1913 from the Mural Formation in British Columbia, Canada, are also discussed.
Locality, geological setting, and age
Specimens described herein as Mickwitzia occidens originate from two nearby sections in the middle member of the Poleta Formation as exposed in Indian Springs Canyon, northern Montezuma Range, Esmeralda County, Nevada (ca. 37.723°N, 117.322°W; Fig. 1). The middle member of the Poleta Formation encompasses an interval extending from the upper part of provisional Stage 3 to the lower part of provisional Stage 4 of Cambrian Series 2 (provisional; Hollingsworth, Reference Hollingsworth2011; Hollingsworth and Babcock, Reference Hollingsworth and Babcock2011). One section was described as IS-4 (Hollingsworth, Reference Hollingsworth1999; Hollingsworth and Babcock, Reference Hollingsworth and Babcock2011) and includes the boundary between the regional stages Montezuman and Dyeran. The second section (IS-W of Fritz and Hollingsworth, see Hollingsworth, Reference Hollingsworth2006) is situated ~200 m east of IS-4. The sections represent a facies typical for the middle member of the Poleta Formation, consisting of an alternation of shaly siltstones, claystones, and fine sandstones with intercalated lenses and beds of bioclastic limestones and dolomitic limestones. This studied part of the formation corresponds to the ‘lower siltstone unit’ described by Moore (Reference Moore1976), which in Esmeralda County contains features of subtidal deposition only (Moore, Reference Moore1976).
Figure 1 Map showing the location of the Indian Springs Lagerstätte in the Montezuma Range, Nevada.
Three levels of exceptional preservation have been recognized in the Poleta Formation in Indian Springs Canyon. They are at ~20 m, 68 m, and 103 m above the base of the middle member (English and Babcock, Reference English and Babcock2010; Hollingsworth and Babcock, Reference Hollingsworth and Babcock2011). Studied specimens are from all three of these levels, and from others within the middle member of the Poleta Formation. However, specimens illustrated herein are primarily from intervals at ~59 m, 62 m, 68–69 m, and 102.5–103 m.
According to Hollingsworth and Babcock (Reference Hollingsworth and Babcock2011), the middle member of the Poleta Formation in Indian Springs Canyon yields polymerid trilobites characteristic of three biozones. In ascending order, they are the Nevadella addyensis Zone (previously termed the Nevadella parvoconica Zone; Hollingsworth, Reference Hollingsworth1999), 0–58 m above the base of the middle member; the Nevadella eucharis Zone, 58–88 m; and the Olenellus Zone (alternatively, the Bonnia-Olenellus Zone), 88 m and above. The 88 m level corresponds to the Montezuman-Dyeran boundary as used regionally in Laurentia (Hollingsworth, Reference Hollingsworth1999, Reference Hollingsworth2011; herein, Fig. 2). This position also approximates the Stage 3-Stage 4 boundary of the developing global chronostratigraphy (Babcock et al., Reference Babcock, Robison and Peng2011).
Figure 2 Composite stratigraphy of the investigated sections of the Indian Springs locality with sample horizons and intervals of exceptional preservation indicated. Measurements refer to meters above the base of the middle member of the Poleta Formation. Stratigraphy compiled after Hollingsworth (Reference Hollingsworth2006, Reference Hollingsworth2011), English and Babcock (Reference English and Babcock2010), Hollingsworth and Babcock (Reference Hollingsworth and Babcock2011), and original measurements. Zonation according to Hollingsworth (Reference Hollingsworth2011) and Hollingsworth and Babcock (Reference Hollingsworth and Babcock2011).
Material and methods
Material and repository
Material from the Indian Springs locality described herein was recovered in 1999 by L. E. Babcock and in 2002 by L. E. Babcock and L. E. Holmer. Supplementary samples were collected by M. Streng in 2009. The material comprises ~50 complete or almost complete specimens from various levels of the lower half of the investigated interval (Fig. 2). All specimens represent single valves, except one specimen that is preserved in a bivalved state. In addition to the macroscopic specimens, 29 carbonate horizons and lenses were sampled (Fig. 2) to obtain shell material of Mickwitzia occidens and other microfossils by the dissolution of the collected samples. Individual samples were dissolved in 10% acetic acid and acid residues were subsequently rinsed, dried, and examined under a binocular microscope to identify and pick Mickwitzia shell fragments and other microfossils. About two-thirds of the samples yielded indisputable specimens of Mickwitzia, which were most common in the samples originating from the interval 53–68 m above the base of the middle member (ca. 20–35 m below the base of the Dyeran Stage; Fig. 2). Recovered Mickwitzia specimens from the carbonate horizons are all fragmentary with individual fragments not exceeding 4 mm in size. The interval with the highest abundance of shell fragments also yielded the most common macrospecimens. The youngest specimens of Mickwitzia within the sampled sections were found ~119 m above the base of the middle member, i.e., 31 m above the base of the Dyeran Stage.
Besides Mickwitzia, a rich macrofauna, including trilobites, linguliform brachiopods, helicoplacoids, chancellorids, sponges, and hyolithids, has been reported from the locality (see English and Babcock, Reference English and Babcock2010; Hollingsworth and Babcock, Reference Hollingsworth and Babcock2011). Nonbiomineralized tissues found include mantle setae of brachiopods, demosponge skeletons, hyolithid gut tracts, trilobite appendages, algae, cyanobacteria, and body parts of nonbiomineralizing arthropods (English and Babcock, Reference English and Babcock2010). Associated with the Mickwitzia shell fragments in the carbonate horizons are common trilobite fragments and chancellorid sclerites, and internal molds of hyolithids and rare linguiliform brachiopods, as well as agglutinated protists (Streng et al., Reference Streng, Babcock and Hollingsworth2005).
Material described herein is deposited at the Paleontological Research Institution, Ithaca, New York, USA (macrospecimens; PRI 68586 through 68637) and the Palaeontological Collections of the Museum of Evolution, Uppsala University, Sweden (acid-etched material; collective number PMU 27409). It should be noted that portions of the acid-etched material were accidently damaged or destroyed after being photographed. This involves all acid-etched specimens with specimen numbers starting with 016–019, 021, 023, 024, 030, 032, and 033. Other institutional abbreviations used in the text refer to specimens deposited in the collections of the U.S. National Museum of Natural History, Washington, D.C. (USNM) and and the Sedgwick Museum, Cambridge (MP).
Photography and compositional analysis
Imaging of fossil macrospecimens was performed with a digital camera (Axiocam; Zeiss, Jena, Germany) attached to a dissection microscope (Leica, Heerbrugg, Switzerland). Brightness, contrast, and color values were, when needed, corrected using image-processing software (Adobe Photoshop CS6, Adobe Systems Inc., San Jose, California, USA).
Preparation for scanning electron microscopy consisted of mounting the selected specimens on aluminum stubs with adhesive carbon backing and subsequently sputter-coated with gold. Scanning electron microscopy of fossil material was performed at the Microscopy Unit at the Evolutionary Biology Centre, Uppsala University. Images were collected using a Zeiss Supra 35VP field emission and a Philips XL30 scanning electron microscope (SEM) operating at 5 kV.
Compositional data was obtained on a Zeiss field emission SEM at 23kV. EDS X-ray spectra and elemental maps were collected on an EDAX Apollo X silicon drift detector system using Genesis Spectrum software (version 6.45, EDAX Inc., Mahwah, New Jersey, USA). Elements mapped in transverse, polished shell sections were C, O, P, Ca, Al, Si, Fe, Mg, and F, with a total of 64 frames per element.
Morphometrics
A set of comparative geometric morphometric comparisons were made in order to evaluate the potential for this technique as a diagnostic tool in refining the species-level characters and taxonomy of the genus Mickwitzia. Measurements from the major and minor axes of the shell extent, position of apex, and juvenile growth zones were used as landmarks (see Fig. 15.1).
Landmark based geometric morphometrics were performed on complete dorsal and ventral valves from Indian Springs (N=7) and on M. monilifera from Lugnås, Västergötland, Sweden (N=31). The data was subdivided into partitioned datasets depending on species and valve position. Landmarks (designated 1–8) corresponded to: the widest points laterally on the outer shell margin (5, 8), the anterior and posteriormost points on outer shell margin (1, 4), the shell apex (2), and the respective widest (6, 7) and longest margins at the boundary of the juvenile and adult shell zones (3) (Fig. 15.1).
Digitization of the landmark data was performed with tpsDIG (ver, 2.22, Rohlf, F.J., 2015, Department of Ecology and Evolution, State University of New York at Stony Brook, New York, USA) and the resulting file analyzed in MorphoJ (Klingenberg, Reference Klingenberg2011). A Procrustes superimposition was employed to normalize size scaling and orientation of the specimens. We generated a covariance matrix and then calculated PCA values in MorphoJ (see Klingenberg, Reference Klingenberg2011). For distinguishing species, we also performed canonical variance and discriminant analyses on the covariance matrix to determine the certainty that individuals of a species would be recovered in the correct taxonomic grouping.
Terminology
Currently three genera are assigned to the mickwitziid stem-group brachiopods, Mickwitzia Schmidt, Reference Schmidt1888, Heliomedusa Sun and Hou, Reference Sun and Hou1987 and Setatella Skovsted, Streng, Knight, and Holmer, Reference Skovsted, Streng, Knight and Holmer2010. The shells of Mickwitzia and Heliomedusa consist of two valves—a low to highly conical one and a flat to slightly convex one. The conical valve is typically referred to as ventral and the more flattened one as dorsal (e.g., Walcott, Reference Walcott1912; Balthasar, Reference Balthasar2004), although unambiguous proof for such an interpretation, such as a pedicle opening, muscle scars or lophophore-support structures, have not as of yet been described in these genera. Support for this traditional interpretation, however, comes from the recently described mickwitziid genus Setatella. Although Setatella differs in shape from Mickwitzia and Heliomedusa in having a biconvex shell with marginal apices, it displays structures interpreted to be equivalent to pseudointerareas in both valves along the posterior margin (Skovsted et al., Reference Skovsted, Streng, Knight and Holmer2010). These pseudointerareas can be continuous or divided by a longitudinal inset, interpreted as a pedicle groove. Accordingly, the valves with the divided pseudointerarea are interpreted as ventral, and those with an undivided posterior margin as dorsal. The flat valves of M. occidens from Nevada described below have a pseudointerarea comparable to that of the dorsal valves of Setatella. Consequently, they are considered as the dorsal valve and are analogous to the flat valves of other species of Mickwitzia as well as Heliomedusa, and are thus also best interpreted to represent dorsal valves.
Description
Preservation and taphonomy
The collection described falls into two distinct categories: (1) hand specimens of complete Mickwitzia, including dorsal, ventral, or both valves, and (2) acid-prepared shell fragments extracted from bulk, dissolved carbonate mudstone and wackestone layers and lenses. The mode of preservation of the shells of M. occidens differs greatly between specimens hosted in either limestone or shale units.
The hand specimens consist of weakly mineralized shells containing little original phosphate, with the most mineralized portion of the shell located at the umbonal region. Weathering of the source rocks could have contributed to this seeming lack of biomineralization by dissolution, and highlights the need for careful appraisal of taphonomic controls on the biomineralized components of shell and shell structures within organisms from this biota. The preservational style of the Indian Springs locality bears some overall similarity to the Chengjiang style of preservation (English and Babcock, Reference English and Babcock2010) in terms of the extent of lithology, weathering, and replication of shell and soft parts as pyrite.
In a small number of specimens, including three Mickwitzia and a few lingulid brachiopods, the marginal mantle setae are preserved. These structures are typically nonmineralized in analogous living and fossilized brachiopods and thus we can interpret this as an incidence of exceptional soft-tissue preservation of these structures.
Shale-hosted material contains numerous examples of complete specimens, retaining both valves in a few specimens although single valves comprise most of the recovered material. Specimens range in size from 10 to 47 mm in diameter, and include a range of ontogenetic stages interpreted as juvenile to adult forms. In carbonate-hosted material, the Mickwitzia shell fragments were easily recovered from acetic acid-treated limestone residues, identifiable by the characteristic pustulose ornament and presence of a distinctive, shell-penetrating pore system with inwardly pointing cones. Fragments collected range downward in length from a few mm to a few hundred microns. This shell hash includes pieces of individuals smaller than the macrospecimens collected from the shale facies.
General features and shape
Shale-hosted ventral valves are diagenetically flattened, being planar to slightly conical in lateral profile. The shell outline is ovoid in shape with the apex located posteriorly from the centroid of the shell and anterior to the posterior shell margin. A distinct, larval-shell growth area, lacking ornament is observed at the shell apex of the valve and abruptly ends at an ~100 µm radius from the apex point in adult specimens. Apices of ventral valves recovered from the carbonates (Fig. 6) are rather variable in shape, suggesting that ventral valves of M. occidens might also have been rather variable in lateral profile. Typically, the slopes of these conical fragments are moderately steep, more or less uniformly inclined in all directions, and have a straight to gently convex lateral profile. However, also rather flat apical shell fragments or fragments with distinct different inclinations between the anterior and posterior slope are known (Figs. 6, 7).
Dorsal valves appear to be more or less flat in lateral profile, but possess a thickly mineralized, marginal umbonal region that bends around on itself. More distal shell material is either nonbiomineralized, thus not present, or is weakly mineralized and has not survived preservation/acid processing. In dorsal valves, the position of the apex is typically located along the posterior edge of the shell margin, or is in close proximity to it. As in the ventral valve, the apex is free of ornamentation, indicating the larval shell.
All Mickwitzia shell material recovered by the dissolution of the limestone samples is fragmentary, with the individual fragments not exceeding 4 mm in maximum length (e.g., Fig. 5.16, 5.17). All fragments show the characteristic shell structure of the genus typified by a multilamellar fabric with inwardly pointing hollow cones and can therefore easily be distinguished from other phosphatic shell material in the acid residues (e.g., Fig. 6). Generally, three types of shell fragments can be distinguished: (1) common conical fragments, i e., apices of ventral valves (Figs. 6, 7, 13); (2) rare sickle-shaped fragments, i.e., apices of dorsal valves (Figs. 8–9); and (3) flat shell fragments, which represent the bulk of the recovered material and cannot be assigned to either valve (Fig. 5).
Ornamentation
Pustulose-reticulate ornament typical of mickwitziids is found on the outermost shell surfaces in well-preserved specimens (e.g., Figs. 5.10, 6.1, 6.4, 8.9, 11.14). Pustules form at the intersection of fine concentric and radial ridges, the latter of which multiply by intercalation. In exfoliated specimens, the reticulate pattern surface ornament becomes weaker depending on the degree of exfoliation (e.g., Figs. 5.7, 5.5, 5.15, 6.10, 8.10). There is seemingly no observed spatial connection between the surface pustules and the inwardly pointing cones, although this might not be the case.
Well-defined pustulose ornament is present in regular concentric bands, as observed in other species of Mickwitzia. Canals that contained tangential shell-penetrating setae are present in shell layers apart from the larval shell zones at the shell apices.
Whether the setae are retained throughout each growth stage or restricted to the shell margin with each previous generation being abandoned is unclear, although in exceptionally preserved specimens only the latter is observed, suggesting that the setae are restricted to the margin.
The ornamentation is not present within an area of ~100 µm radius from the nominal shell apex. We interpret this smooth area as the initial larval shell growth stage; it contrasts with the pustulose ornament on the rest of the shell. This feature is present to an equal degree in all observed shell apices, ventral and dorsal (Figs. 7.7–7.13, 9.10), suggesting it is not simply a function of ornament becoming abraded. There is, however, no clear discrimination between the smooth apex and the subsequent reticulately ornamented shell area; the transition between the areas is gradual.
Position of apex
The shell apex in ventral valves in all studied specimens is located along in a medial position lengthways that ranges in location from between the nominal centroid to close to the posterior shell margin. A scatterplot normalized for size and orientation differences with Procrustes superimposition shows the medial and distribution of apex locations, with standard deviations plotted (Fig. 15). In dorsal valves, the apex occurs at, or in very close proximity to, the posterior margin (Figs. 3.6, 8.3, 8.7, 9.6, 9.10).
Figure 3 Light photographs of shale-hosted macrospecimens of Mickwitzia occidens. (1) Exceptionally preserved, bivalved specimen with preserved mantle setae and remnants of the phosphatic shell (arrows) displaying characteristic cones; specimen PRI 68595 (level 102.5 m). (2) Preserved setae along right? posterolateral shell margin [detail of (1)]. (3) Preserved setae along left? posterolateral shell margin [detail of (1)]. (4) Exfoliated ventral valve with weakly mineralized adult shell, transversely elliptical in outline; specimen PRI 68632 (level 86.7 m). (5) Exfoliated apex [detail of (4)] showing four pairs of enigmatic, subradial grooves, which might correspond to ridges in the apical cavity. (6) Apical view of juvenile ventral valve showing preserved reticulate pattern of exterior shell surface around apex; specimen PRI 68597 (level 102.5 m; specimen coated with a sublimate of ammonium chloride). (7) Subspherical ventral valve, in which only phosphatic shell material of juvenile/proximal growth stages is preserved indicating a primarily poorly mineralized adult/distal shell; specimen PRI 68593 (level 69 m). (8) Left posterolateral margin of (7) with subradial short grooves interpreted as remnants of shell-penetrating setae (compare with Fig. 14.2). (9) Anterior margin of (7) lacking short grooves; fine pits indicative of characteristic shell structure of inwardly pointing cones. (10) Mold of elongate-elliptical ventral valve with original phosphatic shell almost entirely absent; specimen PRI 68605 (level 69 m). (11) Subspherical ventral valve with fine pitting, indicating remnants of original shell structure; specimen PRI 68621 (collected from talus, ca. level 104 m; specimen photographed submerged in water to increase contrast). (12) Subspherical dorsal valve with marginal apex and remnants of original shell structure; specimen PRI 68601 (level 102.5 m; specimen photographed submerged in water). (13) Posteriormost margin of (12) showing robust, well-mineralized shell zone (arrows); zone equivalent in size to the carbonate-hosted sickle-shaped shell fragments (Fig. 8). Scale bars=5 mm (1, 4, 7, 10–12), 2 mm (2, 3, 5, 8, 9), and 1 mm (6, 13).
Shell structure and composition
A degree of weathering and exfoliation affects most of the mineralized components of the shells in the shale facies with the original shell structure being significantly affected or absent as a result. The valves appear weakly phosphatized except for a distinctly thicker region that extends from the umbo of dorsal valves. In more distal portions of the shell, phosphate appears with a patchy distribution.
Exfoliated shell layer surfaces reveal the inwardly pointing cone structures. They manifest as a pustulose texture in the whole specimens (not to be confused with the external surface ornament within which they are interspersed; Fig. 3.2). These are equivalent to the inwardly pointing cones observed in the acid-recovered fragments.
The overall structure of the shell in cross section is dominated by a series of thinly laminated phosphatic sheets. These sheets are interrupted by the presence of regularly spaced, inwardly pointing phosphatic cones. At the interface of these cones with the parallel lamellae, there is a pronounced inward bending of the phosphatic lamellae around the cone structures (Figs. 4.6, 10.7, 10.13, 11.9). Thickness of the shell measured in a polished section of a shale-hosted specimen not affected by dissolution varies along the length from 150 to 300 µm but averages ~250 µm (Fig. 4.4). This contrasts with observations on the acid-etched material, where shell thickness does not exceed 100 µm (e.g., Figs. 5.12, 10.10–10.15). However, the shell of the polished shale-hosted specimen seems to be unusually thick also because the inwardly pointing cones barely elevate from the interior valve floor (Fig. 4.4). In the acid-etched material, a variety of fragments has been found that have irregular areas of similar shell thicknesses and barely elevated cones (Figs. 5.9, 5.13, 5.14, 13.6; potentially also Fig. 5.1, 5.4, 5.16). Inwardly pointing cones are commonly 200–250 µm apart (Figs. 5.6, 6.14). Arrangements of the cones in straight or bent rows have been observed (Figs. 3.2, 5.2), as well as denser groupings of cones and areas free of cones particularly in the ventral apical cavity (Figs. 5.3, 6.6, 6.8).
Figure 4 Backscatter electron (BSE) images of marginal setae and shell structural details in shale-hosted specimens of Mickwitzia occidens. (1–3) Specimen PRI 68595 with preserved mantle setae (see also Fig. 3.1); (1) anterolateral shell margin showing dispersed phosphate and setae; (2) close-up of exceptionally preserved mantle setae [detail of (1)]; bright spots correspond with iron pyrite framboids that comprise the exceptionally preserved setae; (3) detail of (2). (4–6) Polished shell cross sections of specimen PRI 68634 (collected from talus, ca. level 20 m); (4) micrograph showing compact shell lamellae bending around inwardly pointing cone structures (r=resin, sh=shell, mx=rock matrix); (5) internal structure of inwardly pointing cone showing faint transverse ridges (arrow), potentially equivalent to growth lines [detail of (4)]; (6) mapping of distribution of phosphorus (P), calcium (Ca), and silica (Si) in shell and rock matrix; note diagenetically formed silica deposits parallel to shell layers and phosphorus-free matrix. Scale bars=2 mm (1), 200 µm (2), 100 µm (4, 6), and 50 µm (3, 5).
Figure 5 SEM images of shell fragments of Mickwitzia occidens recovered by acid-dissolution of carbonate rocks. (1, 4, 9, 13, 14, 16) Shell fragments showing areas of thicker shell; (1, 4) specimen 062-10 (level 62 m); (9) specimen 024-01 (level 62 m); (13) specimen 032-03 (level 88.5 m); (14) specimen 039-05 (level 59 m); (16) specimen 021-03 (level 69 m). (2, 3, 6, 8) Shell fragments showing the inner shell surface with inwardly pointing cones; (2) specimen 062-11 (level 62 m) with aligned cones; (3) specimen 032-01 (level 88.5 m) with irregularly distributed cones; (6) specimen 019-03 (level 53 m) with regularly distributed cones; (8) specimen 019-04 (level 53 m) with worn cones. (5, 7, 10, 11, 15, 17) Shell fragments showing preservation of the outer shell surface; (5) exfoliated specimen 017-08 (level 69 m) showing no preserved ornamentation; (7) exfoliated specimen 024-08 (level 62 m) with radial sculpture still visible; (10) pristine shell surface without any pores, specimen 035-10 (level 62 m); (11) slightly exfoliated specimen 035-03 (level 62 m) with few pores; (15) specimen 018-08 (level 69 m) with exfoliated areas showing pores and pristine areas with reticulate ornament without pores; (17) large exfoliated fragment with weak reticulate pattern, specimen 030-04 (level 98 m). (12) Exfoliated shell fragment in cross section, specimen 017-09 (level 59 m). Scale bars=1 mm (16, 17), and 500 µm (remaining).
Figure 6 SEM images of apical fragments of ventral valves of Mickwitzia occidens recovered by acid-dissolution of carbonate rocks to show variability in shape of apices. (1, 2) Lateral and ventral views with commonly displayed shell breakage at shell edges, specimen 020-01 (level 59 m). (3, 4, 8–10, 13) Specimens displaying range of shape and preservation in recovered apices; note first-formed shell zone lacking ornamentation; (3, 8) slightly exfoliated conical specimen 018-01 (level 69 m) with few pores visible, revealing underlying inwardly pointing cone structure; apical (3) and profile (8) views; (4) rather flat specimen 024-11 (level 62 m) with pristine shell surface; (9, 10, 13) conical specimens with the lateral and posterior slopes evenly inclined and the anterior slope less so; pores only visible in exfoliated areas; (9) specimen 023-10 (level 62 m); (10) specimen 017-01 (level 59 m); (13) specimen 024-12 (level 62 m). (5–7, 11, 12, 14) Internal shell surfaces of preserved apices; note inwardly pointing cones present in all but absent in zone of first-formed shell in (5) and (7); (5) specimen 024-14 (level 62 m); (6) specimen 036-03 (level 62 m); (7) specimen 024-10 (level 62 m); (11) specimen 016-03 (level 55.5 m); (12) specimen 018-07 (level 69 m); (14) specimen 018-02 (level 69 m). Scale bar=500 µm.
Rather than penetrating the phosphatic laminations that run parallel to the shell surface, the cones are situated within a draped series of these laminations that consistently point inward away from the surface of the shell at both ends of the tube, bending to follow the direction of the associated perpendicular cone. The connection of lamellae between consecutive inwardly pointing cones resembles a series of arches in cross section (Fig. 4.4). By extension of this growth pattern to the shell surface, this explains how the pustulose ornamented texture observed on the surfaces of nonexfoliated shells develops, although an inward cone is not located at the ‘trough’ of every such depression. Thus inward cones, although closely allied to the pustulose ornament are not directly correlated in terms of the regularity of their appearance. More extensive surveying of thin sections is required to conclusively resolve the relationship between the inwardly pointing cones and surface ornament.
The core of the inwardly pointing cones appears to have no discernable structure, and there is no evidence of striated ornament or impressions on the wall of the canals that would point toward the presence of setae in these pores (Fig. 10.1–10.6). In some examples, the canals appear to change diameter along their length as they pass through the laminated shell layers.
In the carbonate-hosted specimens, the shell comprises a series of parallel-laminated phosphatic sheets penetrated by the hollow, infilled, phosphatic inwardly pointing cones. They only penetrate through to the surface when the outer original shell layers are exfoliated. The ornament of pustules is restricted to the outer shell layers and becomes weaker with increasing degree of exfoliation (e.g., Figs. 6.10, 11.1–11.3). Inner shell layers appear as massive sheets of phosphate with no conspicuous ornamentation (Figs. 5.5, 5.15, 7.4).
Figure 7 Additional SEM images of ventral valve preservation of Mickwitzia occidens. (1–3) Recrystallized specimen 023-09 (level 62 m) with corroded apex but with inwardly pointing cone structure preserved; (2) detail of (1); (3) detail of (2). (4) Exfoliated valve displaying pores corresponding with inwardly pointing cones, specimen 023-06 (level 62 m). (5, 6) Ventral valve showing depression at apex, specimen 023-01 (level 62 m); (6) detail of (5) showing circular pit of unknown origin [arrow in (5)]. (7–12) Well-preserved apices with little to no evidence of pores but with distinct, first-formed shell zone lacking ornament; (7) specimen 019-01 (level 53 m); (8, 9) same specimen as in Figure 6.1, 6.2; (10, 11) same specimen as in Figure 6.10; (12) same specimen as in Figure 6.4. (13–15) Specimen 018-03 (level 69 m) with ‘pore’ structures at apex; (14, 15) close-ups of ‘pores’ of (13). Scale bars=200 µm (1, 4, 7–9), 100 µm (2, 5, 10–13), 20 µm (14, 15), and 10 µm (3, 6).
The recovered shell fragments are generally well preserved, but many of them suffered recrystallization that has destroyed the original shell ultrastructure. In these specimens, shell cross sections appear massive, not displaying any original structural details (e.g., Figs. 7.2, 8.12, 10.11, 11.11). Most specimens, however, show the original multilamellar shell fabric, similar to that of Mickwitzia muralensis described by Balthasar (Reference Balthasar2004). Individual lamellae are 1–3 µm thick and separated by a narrow gap, most likely representing originally an organic membrane (Fig. 10.13). A silica infill in this location is noted in a number of specimens.
Figure 8 SEM images of dorsal valves of Mickwitzia occidens, acid-recovered specimens from carbonate samples. (1, 3, 5, 10) Robust, well-mineralized shell fragment with pustulose ornament, specimen 033-03 (level 62 m); (1) anterior view showing acrotretid-like laminated architecture and columns; (3) apical view; (5) oblique posterior view; (10) detail of (5) showing posteriormost shell margin with exfoliated area with pores and pristine area with reticulate ornamentation. (2, 4) Small, slightly exfoliated specimen 017-10 (level 59 m) showing inwardly pointing cones on interior valve surface; (2) anterior view; (4) apical view. (6) Oblique anterior view of robust specimen 023-08 (level 62 m) showing acrotretid-like shell structures. (7, 13) Remineralized dorsal valve fragment, specimen 023-11 (level 62 m); (7) obique posterior view of apical area; (13) apical view. (8, 9, 12, 14) Partly remineralized specimen 034-07 (level 59 m) with inwardly pointing cones (12, 14) and secondary phosphate growth over outer shell layers; (8) oblique ventral view; (9) posterior view; (12) shell cross section close to apex; (14) anterolateral view. (11) Fragment with poorly preserved apical surface area in oblique lateral vuew, specimen 037-03 (level 62 m). (15, 16) Exterior (15) and interior (16) of poorly preserved fragment still displaying pustulose ornament and inwardly pointing cones, specimen 037-04 (level 62 m); (15) posterior view; (16) anterior view. Scale bar=350 µm (16), 250 µm (8, 9, 14), 200 µm (10), 150 µm (12), and 500 µm (remaining).
Besides the hollow cones, straight canals that measure 40–50 µm in diameter pierce the shell at a low angle rather than perpendicularly (Fig. 12). Their function as setal canals is supported by the fine longitudinal striations of the canal walls (Fig. 12.8, 12.12), which are reminiscent of seta-forming microvilli, and the presence of phosphatized setal remains inside some of the canals (Fig. 12.11).
Exceptionally preserved structures
In a number of specimens (N=3), mantle setae are preserved in pyrite with a distinct framboidal texture. Mantle setae are composed of chitin secreted from setoblast cells in living brachiopods, thus the replication of the setae represents exceptional preservation of nonbiomineralized tissues.
The distribution of setal structures is restricted to the distalmost shell margin. Setae are typically oriented in a radial fashion originating from the centroid of the organism but with a posterior-facing tendency along the posterolateral margins (Figs. 3.1, 3.2, 14.1). Setae appear to be present around the whole circumference of the valve with the exception of the medial anterior margin. They are apparent as dark projections from the shell under stereoscopic light microscopy. Backscatter electron microscopy also delineates these structures from the surrounding matrix (Fig. 4.1–4.3). The distinction is facilitated by setal preservation as framboidal pyrite, which contrasts with the surrounding mudstone matrix. The observed direction and distribution of setae matches the distribution of small ridges on several macrospecimens (e.g., Fig. 14.2). These ridges measure 400–600 µm in length and most likely represent the setae-hosting canals that pierce the shell at a low angle. Their distribution indicates that such canals might have been an adult feature, because they only occur close to the shell margin and are absent from more juvenile growth stages of the shell.
Phosphatized setae contained within canals have previously been described in other fossilized linguliform and rhynchonelliform brachiopods (Jin et al., Reference Jin, Zhan, Copper and Caldwell2007; Popov et al., Reference Popov, Bassett, Holmer and Ghobadi Pour2009). Unlike in the siphonotretids described by Popov et al. (Reference Popov, Bassett, Holmer and Ghobadi Pour2009), in Mickwitzia occidens, the hollow canals containing the setae do not extend outward from the shell surface. The discovery of these shell-penetrating setae bearing canals in Mickwitzia further supports the view that the genera Micrina Laurie, Reference Laurie1986, Tannuolina Fonin and Smirnova, Reference Fonin and Smirnova1967, and Mickwitzia constitute members of a Cambrian stem group of Linguliformea because these structures are present in all of these taxa, although it should be pointed out that this is likely a symplesiomorphic character.
Acrotretoid-like shell structures
Within the apical part of both ventral and dorsal valves, structures are occasionally present that are reminiscent of columnar laminae of the acrotretoid shell structure (see, e.g., Williams and Holmer, Reference Williams and Holmer1992; Holmer et al., Reference Holmer, Popov and Streng2008a). These comprise hollow columns of calcium phosphate with an average diameter of ~15 µm bound by an inner and an outer lamella. Columns and lamellae define laminae of variable thickness (e.g., Figs. 9.13, 9.17, 13.13–13.15). The columns run perpendicular to the flat, phosphatic shell lamellae and are distributed in irregular fashion. The intralaminar space is typically filled with spherulitic phosphate granules (Figs. 9.13–9.17, 9.19, 13.7, 13.15) referred to as ‘cocci’ in previous descriptions (Holmer et al., Reference Holmer, Skovsted and Williams2002; Skovsted and Holmer Reference Skovsted and Holmer2003; Skovsted et al., Reference Skovsted, Brock, Holmer and Paterson2009a). These have been variably described as of bacterial, diagenetic, or decay-fabric origin. Cocci are present in both limestone- and shale-hosted examples and are often found clustered around phosphatic columns and lamellae alike. There seems to be a direct connection between the inwardly pointing cones and the columns of the laminae. The columns have about the same diameter as the cones and appear to be extensions of the cones (Fig. 13.9, 13.13). The ultrastructure of the columns is unclear; multilamellar (equivalent to cones; Fig. 13.14) and massive (potentially recrystallized; Fig. 9.14) structures have been observed.
Figure 9 SEM micrographs of additional dorsal valves of Mickwitzia occidens, acid-recovered specimens from carbonate samples. (1–3, 5) Heavily mineralized and recrystallized dorsal, sickle-shaped apex fragment; note pseudointerarea-like surface (2) and the W-shaped margin to the preserved shell area, specimen 039-02 (level 59 m). (4, 7, 9) Dorsal fragment with pseudointerarea-like surface, specimen 024-07 (level 62 m); (7) close-up of shell laminations showing an accretionary-like succession of phosphate layers with nick points presumably for setae canals, subsequently overgrown; (9) anterior view with poorly preserved interior surface. (6) Poorly preserved apical fragment with central apex, specimen 040-01 (level 62 m). (8, 11) Corroded specimen 037-01 (level 62 m) showing transition from ornamented posterior shell margin to pseudointerarea-like surface; (8) entire specimen in ventral view; (11) detail of (8). (10, 13, 17) Close-ups of specimen in Figure 8.6; (10) dorsal surface with unornamented first-formed shell; (13, 17) cross sections close to apex showing acrotretid-like columns interspersed between phosphatic lamellae; space between phosphatic layers is infilled with phosphatized cocci structures. (12) Inwardly pointing cone structures; same specimen as in Figure 8.2, 8.4. (14–16) Close-ups of specimen shown in Figure 8.1 showing cross section through acrotretid-like columns with surrounding cocci. (18) Detail of Figure 8.16 showing interior surface from near apex in a seemingly not-fully-grown specimen; note absence of acrotretid columns and poor development of inwardly pointing cones. (19) Details of phosphatic cocci of specimen 039-03 (level 59 m). Scale bar=500 µm (1–6, 8, 9), 200 µm (7, 10, 11, 13, 17), and 150 µm (18); thin scale bars=50 µm (12, 14–16) and 10 µm (19).
It is also currently unresolved during which ontogenetic stage the columnar layers would have formed. The columns might represent a juvenile feature that is lost or overgrown in subsequently deposited adult shell material, or they could be an adult/gerontic character that only would have been present in relatively large shells to provide extra stability.
The presence of columnar structures in members of the brachiopod stem group (Mickwitzia and Setatella) and in tommotiids (e.g., Micrina) strengthens the suggested link between one group of small shelly fossils and linguliform brachiopods (e.g., Holmer et al., Reference Holmer, Popov and Streng2008a; Skovsted et al., Reference Skovsted, Clausen, Álvaro and Ponlevé2014). We consider these as homologous structures in the stem Brachiopoda and closely allied tommotiids.
Growth patterns of valves
The conical ventral valve fragments with their centrally situated apices suggest a holoperipheral growth pattern in the ventral valve of Mickwitzia occidens. This is supported by complete ventral valves from the shale-hosted material that displays a subcentral apex. In contrast to the seemingly hemiperipheral growth of complete dorsal valves (Fig. 3), the sickle-shaped fragments tell a different story. The apex of the dorsal valve is not positioned on the margin itself but rather inset ~300 µm from the posterior margin. This implies that at up to a diameter of ~300 µm, dorsal valves grew in a holoperipheral fashion. Subsequently, a hemiperipheral growth pattern was then adopted. We interpret this as a shift in growth modes from a juvenile to adult pattern of shell growth.
Morphometric analyses
Although shell measurements and shape have been used as taxonomic criteria within stem Brachiopoda, no quantitative descriptive measure of such features has been yet utilized. The developmental basis of these features is also of interest in determining the character homologies of phosphatic-shelled stem lophotrochozoan taxa and possible links to tommotiid ancestors.
No discernible discriminant signal could be retrieved from the dorsal values, however, ventral valves from Västergötland and Nevada representing individuals of Mickwitzia monilifera and M. occidens, respectively, could be differentiated with principal component analysis, and discriminant function analysis. Based on these results, landmarks 1, 2, and 3 (Fig. 15.1), which indicate the rearward placement of the ventral shell apex, constitute strong signals for species diagnosis (recovering the separate taxonomic groups in 100% of cases). The outline landmarks also display a significant discriminant signal (length:width ratio) that can be more useful under an outline shape analysis. Cross-validation scores recovered the correct taxonomic groupings in 61.5% of Nevada specimens based upon comparisons of the ventral valve. Mickwitzia monilifera was recovered 66.67% of the time.
Comparison with the holotype
The holotype of Mickwitzia occidens (USNM 51518a) is an unevenly convex, irregularly shaped fragment of a variably exfoliated valve in exterior view, measuring ~6 mm in diameter. The unevenly convex shape of the fragment, with a steep and a gently inclined slope along its longest axis, suggests the position of an apex at the marginally situated culmination. No original shell margin or pristine shell surfaces are preserved. Instead, the fragment shows different degrees of exfoliation, with one-third of the shell area, which includes the presumed apex, being stronger affected than the other two-thirds. In the smaller area, pores related to the inwardly pointing cones are more distinct and no original shell ornament is evident (comparable with, e.g., Fig. 5.5 and exfoliated area in Fig. 5.15). In the remaining two-thirds, equivalent pores are smaller and remnants of the original external shell ornament are preserved as faint radial ridges (comparable with, e.g., Fig. 5. 7, 5.17). The convex shape of the fragment suggests the holotype to be a ventral valve, however, due to the small size of the fragment, an incomplete dorsal valve with a potentially marginal apex cannot readily be excluded.
The holotype was collected from sandstones at U. S. Geological Survey locality 174c on a small hill in the salt flat, 1.6 km northeast of Silver Peak Mill, Esmeralda County, Nevada (Silver Peak Quadrangle; Walcott, Reference Walcott1908, Reference Walcott1912), which is ~25 km west of Indian Springs Canyon. The type level is uncertain, but is likely to be within the Harkless Formation (Rowell, Reference Rowell1977; Mount, Reference Mount1980), which overlies the Poleta Formation and belongs to the Bonnia-Olenellus Biozone.
Based on the occurrence of the holotype and the Indian Springs specimens in the same general area and time interval, we consider them to be conspecific. However, the incomplete preservation of the holotype makes the comparison with the Indian Springs material not entirely unequivocal.
Systematic paleontology
Stem group Brachiopoda
Mickwitziids sensu Holmer and Popov (Reference Holmer and Popov2007)
[including Mickwitziidae Gorjansky, Reference Gorjansky1969]
Remarks
Mickwitziids differ from linguliform brachiopods by the combination of a columnar shell structure with striated shell-penetrating setal tubes, poorly developed pseudointerareas, pustulose shell ornament, a shell apex in a submarginal, posterior-displaced position, and the shell margin round in shape and much less elongate. However, the taxonomic composition and phylogenetic position of this informal grouping need to be further investigated.
Genus Mickwitzia Schmidt, Reference Schmidt1888
Type species
Lingula? monilifera Linnarsson, Reference Linnarsson1869, p. 344–345, pl. 7, figs. 1, 2 from the lower Cambrian (Series 2, Stage 4) Mickwitzia Sandstone Member of the File Haidar Formation near Lugnås, Västergötland (Sweden).
Diagnosis
Relatively large, ventribiconvex stem-group brachiopods, ovate to subcircular in outline, with organophosphatic shell composition. Shell ultrastructure multilamellar, typified by inwardly pointing cones perpendicular to shell surface containing hollow canals; shell can be penetrated by tangential canals bearing microvilli impressions and setae. Outer shell surface with a pustulose reticulate surface ornament. Both valves with rudimentarily developed pseudointerareas.
Occurrence
Mickwitzia has been reported from the Cambrian Series 2 of western Laurentia (e.g., Walcott, Reference Walcott1908; Mount, Reference Mount1976, Reference Mount1980; McMenamin, Reference McMenamin1992; Balthasar, Reference Balthasar2004), Baltica (e.g., Walcott, Reference Walcott1912; Jensen, Reference Jensen1993), and Australia (Skovsted et al., Reference Skovsted, Brock, Holmer and Paterson2009a). Material described as Mickwitzia from eastern Laurentia (Greenland; e.g., Skovsted and Holmer, Reference Skovsted and Holmer2003, Reference Skovsted and Holmer2005) is now considered to represent a distinct genus, Setatella (Skovsted et al., Reference Skovsted, Streng, Knight and Holmer2010).
Remarks
With the separation of the eastern Laurentian mickwitziids from Mickwitzia and their placement in the new genus Setatella by Skovsted et al. (Reference Skovsted, Streng, Knight and Holmer2010), a more precise and confined diagnosis of the genus Mickwiztia is now possible (compare Holmer and Popov, Reference Holmer and Popov2007). Mickwitzia is clearly defined and separated from Setatella and the crown linguliform brachiopods by its diagnostic shell ultrastructure of inwardly pointing cones. In this respect, Mickwitzia is only comparable to Heliomedusa, which in fact might represent a junior synonym of Mickwitzia (see Discussion and Interpretation below).
In addition to the type species, Mickwitzia occidens and M. muralensis are herein included in the genus. Mickwitzia lochmanae (nom. corr. herein, ex M. lochmana McMenamin, Reference McMenamin1992) from the Puerto Blanco Formation (Cambrian Series 2) of Sonora, Mexico is only known from a single, exfoliated macrospecimen, a supposed ventral valve (McMenamin, Reference McMenamin1992). Specimens described as M. multipunctata McMenamin, Reference McMenamin1992 were obtained by the dissolutions of limestone samples from the same formation. These specimens are nondiagnostic shell fragments, which are indistinguishable from any other Mickwitzia species and are most likely conspecific with M. lochmanae. Furthermore, the single known valve of M. lochmanae displays a central apex, which suggests reference to M. muralensis. Because of the unsatisfactory knowledge of both M. lochmanae and M. multipunctata, their likely synonymy, and the resemblance of the holotype of M. lochmanae with M. muralensis, both taxa described by McMenamin (Reference McMenamin1992) are herein considered nomina dubia.
Specimens described by Skovsted et al. (Reference Skovsted, Brock, Holmer and Paterson2009a) from Australia as Mickwitzia sp. have been compared with M. muralensis but are too fragmentary to allow a species determination. However, their ornamentation with smaller and seemingly wider-spaced pustules might suggest a new species.
Mickwitzia occidens Walcott, Reference Walcott1908
1908 Mickwitzia occidens Walcott, p. 54, pl. 7, fig. 1.
1912 Mickwitzia occidens; Walcott, p. 331, pl. 6, fig. 4.
?1976 Mickwitzia occidens; Mount, fig. 3.
1977 Mickwitzia occidens; Rowell, p. 79, pl. 1, fig. 12.
?1980 Mickwitzia occidens; Mount, fig. 3.
?1992 Mickwitzia muralensis Walcott, Reference Walcott1913; McMenamin, p. 181, figs. 4.1, 5.4–5.6.
2001 Mickwitzia, Babcock, fig. 2.
non 2002Mickwitzia cf. occidens; Holmer, Skovsted and Williams, text-figs. 2–3.
2003 Mickwitzia occidens; Skovsted and Holmer, fig. 2.
non 2003 Mickwitzia cf. occidens; Skovsted and Holmer, p. 3, figs. 3–5, 7–12.
non 2005 Mickwitzia cf. occidens; Skovsted and Holmer, p. 327–328, pl. 1, figs. 9–13.
2007 Mickwitzia occidens; Holmer and Popov, fig. 1711d–i.
non 2007 Mickwitzia sp. cf. occidens; Holmer and Popov, figs. 1712, 1713.
2010 Mickwitzia occidens; English and Babcock, fig. 4a, b.
2011 Mickwitzia occidens; Hollingsworth and Babcock, fig. 5.6, 5.7.
Holotype
Exfoliated valve (USNM 51518a) from the lower Cambrian (=Series 2) of the Silver Peak quadrangle, Esmeralda County, Nevada, U. S. Geological Survey locality 174c (Walcott, Reference Walcott1908, pl. 7, fig. 1; reillustrated by Rowell, Reference Rowell1977, pl. 1, fig. 12). According to Rowell (Reference Rowell1977) and Mount (Reference Mount1980), the holotype was collected from the Harkless Formation in the lower part of the Bonnia-Olenellus Biozone.
Diagnosis
Species of Mickwitzia with ventribiconvex shell and regularly distributed pustules. Marginal setae, where present, oriented in posterior facing direction on flanks. Position of apex variable but always posterior to shell centroid and anterior to posterior margin in ventral valve; in dorsal valves, apex typically coincident or in very close proximity to posterior shell margin.
Occurrence
Nevadia addyensis Biozone to lower? Bonnia-Olenellus Biozone (Cambrian Series 2, Stage 3–4), western Laurentia (northern Mexico, Nevada, and California; see synonymy; biozonation after Hollingsworth, Reference Hollingsworth2011).
Remarks
Balthasar (Reference Balthasar2004) interpreted the inwardly pointing cones with hollow canals to have hosted setae. However, no evidence for perpendicular shell-penetrating setae, such as striation in association with the canals, could be observed. We interpret this structure as a punctae-like outfolding of the mantle tissue.
Discussion
It has been suggested that mickwitziids occupy a unique position linking tommotiids to the brachiopod stem group and also crown linguliform brachiopods based upon the presence of homologous characters including shell composition, ultrastructure, and striated tubes related to the presence of shell-penetrating setae (e.g., Williams and Holmer, Reference Williams and Holmer2002; Holmer et al., Reference Holmer, Skovsted, Brock, Valentine and Paterson2008b; Skovsted et al., Reference Skovsted, Clausen, Álvaro and Ponlevé2014). Exceptionally preserved Mickwitzia specimens from Nevada provide new insight into morphology and shell structure, which in turn provides clues to their affinities.
One issue of contention concerning mickwitziids is interpreting which is the dorsal valve and which is the ventral valve. Because no evidence of a pedicle, pedicle groove, or sinus to accommodate a pedicle is known, we cannot use this as a diagnostic feature. The shell we interpret as ventral (following Balthasar, Reference Balthasar2004) is the valve that is more steeply curved and has a higher profile. The flattened valve is therefore interpreted as dorsal in orientation. This view is consistent with recent interpretation of the mickwitziid genus Setatella (Skovsted et al., Reference Skovsted, Streng, Knight and Holmer2010). Life positions of mickwitziids are unknown. The marginal setae could have provided support in softer sediments. As yet, no specimens of Mickwitzia occidens that are attached to skeletal matter or other hard substrates are known.
Compared with Mickwitzia monilifera, the overall outline shape of M. occidens is widest along the lateral margins rather than in a posterior-anterior direction. This results in a significant difference in the length:width ratio of these two species. Results of the geometric morphometric analysis (Fig. 15.3, 15.4) also confirm that there is significant interspecies variation with regard to the selected landmarks, particularly the apex position, enough to warrant separation of M. monilifera from M. occidens.
We currently lack sufficient data to include Mickwitzia muralensis in our comparative morphometric dataset. In terms of characters, ventral valves of M. occidens from Indian Springs are distinct from those of the Swedish M. monilifera in having a more central apex and appear to be generally flatter in lateral profile. Furthermore, both taxa differ in shell outline, with M. occidens being typically wider than long (length:width ratio of ~3:4), but considerable variability in the length:width ratio is evident in both taxa.
Compared with Mickwitzia muralensis (see Balthasar, Reference Balthasar2004), inwardly pointing cones are less common in M. occidens, and the acrotretid-type shell structure that is occasionally present in apical shell portions of M. occidens has not been described for M. muralensis. There is also an absence of a conspicuous inner layer/surface to the shell.
Ornament in Mickwitzia occidens consists of pustulose protrusions from the shell surface in regular concentric bands as is commonly observed in other species of Mickwitzia. These are less conspicuous in macrospecimens than in the acid residues, possibly due to exfoliation, secondary recrystallization, and diagenetic loss of features.
Shell composition of mickwitziids is thought to be organophosphatic, consisting organic layers with interspersed phosphatic portions (Skovsted and Holmer, Reference Skovsted and Holmer2003; Skovsted et al., Reference Skovsted, Streng, Knight and Holmer2010). Although a calcitic component has been suggested (Skovsted and Holmer, Reference Skovsted and Holmer2003; Balthasar, Reference Balthasar2004), we found no evidence of this in polished sections analyzed with EDS and X-ray mapping in the SEM. Rather we see a laminated series of phosphatic sheets (Fig. 4.6) similar to previously described mickwitziids with presumably diagenetic siliceous material interspersed between these. Acid-treated material was restricted to robust portions of what we interpret to be more heavily biomineralized areas of the shells corresponding to the shell apices. More distal portions of the shell were not recovered or were severely fragmented, suggesting that these portions were primarily organic in nature with only weakly or nonbiomineralized portions.
Interpreting the structure and function of inwardly pointing cones is the source of some uncertainty within these organisms. One possibility is that they are homologous to acrotretid-type columns, but this is easily discounted due to the significant structural differences (e.g., presence of laminated phosphatic structure in cones as compared to solid pillars of phosphate with a pore in acrotretid columns), the variable width of inwardly pointing cones compared to a near-constant diameter of columns, their position as a component of the outer shell layers bent inward rather than pillars interspersed between phosphatic layers, and finally the difference in size between these structures. Another interpretation of the inwardly pointing cones has been as a setae-bearing structures (Balthasar et al., Reference Balthasar, Skovsted, Holmer and Brock2009). In no instance are striated wall surfaces indicating microvilli (which would imply the presence of setae) observed in association with the inwardly pointing cone structures, suggesting they do not represent incidences of shell-penetrating setae as has been suggested previously (Balthasar, Reference Balthasar2004). This is despite the presence of exceptionally preserved setae along the margins of these organisms. Presumably we would expect to see some evidence of striations or intact setae if the preservational conditions were conducive to retention of these relatively robust structures in other portions of the shell. Furthermore, in specimens not subjected to exfoliation, it seems that many of these canals do not reach the shell surface, either from abandonment and subsequent overgrowth or because they never represented a shell-penetrating structure in the first instance. We interpret both these findings as conclusive evidence that the inwardly pointing cones are not setae-bearing in Mickwitzia occidens and other Mickwitzia species. Alternatively, if they connect to the exterior, we propose that these structures are more likely to represent structures convergent to, but likely not homologous with, punctae in crown-group brachiopods.
There appears to be a close relationship between the inwardly pointing cones and parallel lamellae. The lamellae bend inward perpendicular to the other shell-forming layers when they encounter a cone rather than simply intersecting the cone in a perpendicular fashion. The densely packed phosphatic lamellae themselves are very similar to those observed in tommotiid organisms including Camenella Missarzhevsky in Rozanov et al., Reference Rozanov, Missarzhevsky, Volkova, Voronova, Krylov, Keller, Korolyuk, Lendzion, Michniak, Pykhova and Sidorov1969, Sunnaginia Missarzhevsky in Rozanov et al., Reference Rozanov, Missarzhevsky, Volkova, Voronova, Krylov, Keller, Korolyuk, Lendzion, Michniak, Pykhova and Sidorov1969, Tannuolina, and Micrina (Balthasar et al., Reference Balthasar, Skovsted, Holmer and Brock2009; Skovsted et al., Reference Skovsted, Brock, Holmer and Paterson2009a, Reference Skovsted, Clausen, Álvaro and Ponlevé2014; Murdock et al., Reference Murdock, Bengtson, Marone, Greenwood and Donoghue2014). Many of these taxa also possess shell-penetrating, non-setae-bearing canals that might share a common origin with the inwardly pointing cone structures in Mickwitzia. We suggest a possible homologous origin and secretion mechanism for these laminated phosphatic fabrics, further strengthening the link between small shelly fossils and organisms that appear conspicuously more brachiopod-like, suggesting a stepwise acquisition of characters in the transition from a tommotiid-type organism toward Brachiopoda. The presence of acrotretid-like columns in the juvenile shell area of Mickwitzia, Setatella, and the other tommotiid taxa mentioned, further supports the case for a close phylogenetic affinity between these taxa and the crown-group linguliform brachiopods.
Many specimens preserved in shale show short, radially arranged structures in the distal posterior and lateral shell areas (Figs. 3.8, 14.2). These structures match the observed length of the shell-penetrating tubes observed in fragments recovered from carbonate matrices (e.g., Fig. 12.1). Preserved setae are restricted to the margin of the last-formed shell (Figs. 3.1–3.3, 14.1), suggesting that setal canals are abandoned as new shell layers are produced. Additionally, the absence of setae anterior of the median sector (Fig. 14.2) could have some significance in terms of feeding behavior. The same pattern is observed in Mickwitzia muralensis (Fig. 14.3).
In terms of growth mode, Mickwitzia occidens seems to possess a number of distinct ontogenetic stages, or growth episodes, which can be distinguished. These stages correspond to zones of varied geometric conformation and structural composition in the shell. The first-formed shell consists of a smooth zone ~200 µm in diameter that lacks accretionary growth and penetrating setae (Figs. 6, 7; i in Fig. 16). This is surrounded by a circular to oval zone that shows a reticulate-pustulose surface ornament without evidence of accretionary growth. Internally, this zone can possess acrotretid-type columns (Figs. 8.6, 9.13–9.17; f and j in Fig. 16) and widely spaced phosphatic lamellae reminiscent of the internal structure of Micrina and other tannuolinids. The shell is well mineralized in this area. A more lightly phosphatized ‘adult shell’ area follows, with a seemingly much thinner shell, that lacks the hollow multilayered structure found closer to the apex, the phosphatic laminations are tightly spaced, and the reticulate-pustulose ornament is less distinct (c in Fig. 16). Features suggesting accretionary or at least episodic growth that correspond with ring-like structures as the margin grows outward can be seen. Shell growth generally becomes more orthocline in the ventral valve as the shell grows outward, a possible gerontic feature.
Comparison of Mickwitzia with Heliomedusa from Chengjiang localities has been previously suggested due to similarities in preservation type, setae arrangement, and overall length:width ratio. The exceptionally preserved setae from Indian Springs further support this link. It is possible to interpret Heliomedusa as a junior synonym of Mickwitzia, although without original shell structure present in Heliomedusa for comparison, making this case will prove difficult. The role of taphonomy and taxonomic ambiguity as a result of diagenetic alteration plays a significant role especially in interpretations of shell structure and relates to the taxonomic uncertainty of the group. Confusion of the nature of the various shell layers by various authors is one example of this.
Confounding interpretation of the morphology of these organisms is a complex and varied taphonomic history associated with each locality and indeed even facies. The nature of preservation in these organisms, as in all fossils, is biased depending on depositional environment, porewater geochemistry, and subsequent diagenetic and tectonic alteration. Typically, each new fossil locality presents us with both novel data and a new set of biasing factors. Preservation with soft-part anatomy intact can be associated with dissolution of original shell structure or diagenetic recrystallization that obscures original morphology. Are structures then absent for preservational reasons or for evolutionary ones, sensu Donoghue and Purnell (Reference Donoghue and Purnell2009)? Such is the case of the alleged shell-penetrating setae associated with the inwardly pointing cones.
Distribution of organisms within or closely allied with the genus Mickwitzia in Cambrian Series 2 and 3 seems to be global (Holmer and Popov, Reference Holmer and Popov2007; Skovsted et al., Reference Skovsted, Brock, Holmer and Paterson2009a), and such taxa are found both in carbonate and siliciclastic facies. An older putative mickwitziid has recently been described from the Terreneuvian (Cambrian Stage 2) as Kerberellus marcouensis Devaere, Holmer, and Clausen in Devaere et al., Reference Devaere, Holmer, Clausen and Vachard2015. Fragmentary phosphatic mickwitziids, such as those recovered from the Indian Springs carbonate layers, are potentially a useful biostratigraphic tool if their stratigraphic range and taxonomy can be more precisely determined. Solidifying a definition of the characteristic features of Mickwitzia will be crucial to such an effort and we hope that the comparative data provided here can further this goal.
Figure 10 Detailed SEM micrographs of inwardly pointing cone structures of the shell of Mickwitzia occidens. (1–4) Inwardly pointing cones in different preservations; note central canal; (1) specimen 017-09 (level 59 m); (2) specimen 033-04 (level 62 m); (3) specimen 021-06 (level 69 m); (4) specimen 023-01 (level 62 m). (5–8) Cone-associated canal with laminated phosphatic wall structures; (5) specimen 030-01 (level 98 m), detail of (10); (6) specimen 018-07 (level 69 m); (7) specimen 032-04 (level 98 m); (8) specimen 017-14 (level 59 m). (9) Inwardly pointing cone in dorsal apical fragment; detail of Figure 8.2. (10–15) Cross sections of inwardly pointing cone structures; note inward bending of phosphatic lamellae (13) and equivalent laminated structures (14, 15); (10) specimen 030-01 (level 98 m); (11) specimen 021-05 (level 69 m); (12) specimen 019-04 (level 53 m); (13) same specimen as in Figure 5.16; (14) specimen 030-01 (level 98 m); (15) specimen 030-05 (level 98 m). Scale bars=50 µm (10, 11, 13–15), 25 µm (12), and 10 µm (1–9).
Figure 11 Detailed SEM micrographs of shell ultrastructure and ornament of Mickwitzia occidens. (1–3) Close-up of exfoliated specimens illustrating the multilamellar shell structure; ‘pores’ in (2) correspond with underlying inwardly pointing cone structures; (1, 3) specimen 023-02 (level 62 m); (2) same specimen as in Figure 7.13. (4, 5) Exfoliated pustules showing underlying rod-like phosphatic ultrastructure, specimen 023-10 (level 62 m). (6, 7) Pores on slightly (6) and more strongly (7) exfoliated shell fragments related to inwardly pointing cone structure; (6) same specimen as in Figure 6.3; (7) same specimen as in Figure 5.7. (8, 9) Exfoliated (8) and cross section (9) of shell surface showing densely packed phosphatic lamellae; (8) same specimen as (6) and Figure 6.3; (9) same specimen as in Figure 10.13. (10) Irregular canals penetrating shell (probable pseudobiological structures from pyrite crystal growth), specimen 018-04 (level 69 m). (11–14) Various stages of surface exfoliation revealing pores of inwardly directed cones; (11) specimen 024-10 (level 62 m); (12) same specimen as in Figure 6.1; (13) same specimen as in Figure 5.15; (14) specimen 024-03 (level 62 m). (15) Cross section through ventral apex with poorly preserved inwardly pointing cones overgrown by phosphatic cement, specimen 033-01 (level 62 m). Scale bars=100 µm (10–15), 20 µm (1–6), and 10 µm (7–9).
Figure 12 Striation comparison of setigerous tubes and inwardly pointing cones of Mickwitzia occidens. (1, 4, 7–9) Specimen 039-07 (level 59 m) with three tangential setal canals in association with inwardly pointing cones oriented perpendicular to shell surface; (1, 4) interior and exterior view, respectively; (7) close-up of (1) with microvilli striations visible on canal wall; (8) detail of (7) showing further details of individual microvilli and the phosphatic crystals that have replaced the original chitinous structure; (9) cross section of inwardly pointing cone; note that no striations are observed on the wall of this structure. (2, 5, 11, 12) Additional fragments showing association of inwardly pointing cones and setal canals; (2, 12) specimen 063-01 (level 99 m) with preserved microvilli striations (12); (5, 11) specimen 063-06 (level 62 m) with rudimentarily preserved, phosphatized, three-dimensional setae (11) inside setal canals. (3, 6) Specimen 034-08 (level 59 m) with inwardly pointing cones and remnants of setal canal. (10) Specimen 035-06 (level 62 m) with inwardly pointing cones and crossing linear structures, interpreted to represent mantle setae that have been overgrown by phosphate. Scale bar=500 µm; thin scale bars=200 µm (10), 100 µm (6, 7), 50 µm (11), 20 µm (8, 9), and 10 µm (12).
Figure 13 Detailed SEM micrographs of acrotretoid-like shell ultrastructure in ventral valves of Mickwitzia occidens. (1, 10–12) Ventral valve with thick, additional shell laminae in apical cavity; total number of laminae obscured by phosphatization interior to the first-formed additional lamina (= first lamina); specimen 036-05 (level 62 m); (10) close-up of apical cavity showing first lamina supported by columns followed by massive unit of phosphate; (11) first lamina supported by columns that appear to arise from inwardly pointing cone structures; (12) close-up of boundary between two successive laminae, i.e., the inner lamella of the first lamina and the outer lamella of the following lamina. (2) Specimen 024-02 (level 62 m) with acrotretoid-like laminae obscured by phosphatization. (3) Apical cavity with densely spaced inwardly pointing cones, potentially representing an early stage in the formation of acrotretoid-like columns; same specimen as in Figure 6.6. (4, 7–9) Valve similar to (1) showing first lamina and boundary to second lamina; second lamina and potential successive laminae obscured by phosphatization; specimen 040-02 (level 62 m); (7) close-up of columns of first lamina and boundary to second lamina; (8) close-up of inwardly pointing cones in proximity to the preserved first lamina; space between cones filled with phosphatized cocci; (9) transition between inner shell surface with inwardly pointing cones and preserved first lamina with columns seemingly arising from cones. (5, 6, 13–15) Ventral valve with a succession of laminae preserved in the apical cavity, specimen 063-07 (level 62 m); (5) dorsal view of apical cavity with patch of preserved laminae; (6) lateral view; note marginal thickened area (compare with Fig. 5.9, 5.13); (13) close-up of (6); (14) transition between two laminae (third and fourth) showing column penetrating two successive layers; note that boundary between laminae consists of two lamellae; (15) transition between first lamina and thin second lamina; inner lamella of first lamina and outer lamella of second lamina clearly visible. Scale bar=500 µm (1–6); thin scale bars=200 µm (10, 13), 100 µm (9, 11), 50 µm (7), and 20 µm (8, 12, 14, 15).
Figure 14 Distribution of marginal mantle setae in Mickwitzia occidens (1, 2) from Indian Springs and M. muralensis (3) from the Mural Formation (Cambrian Series 2), Canada. (1) Camera lucida drawing of specimen in Figure 3.1 with preserved marginal mantle setae (as in Fig. 3.2–3.3); note distinct absence of setae at anteriormost shell margin. (2, 3) Sketches of ventral valves (PRI 68593 counterpart and MP 2318) with fine radial ridges (short, dark gray, solid lines) interpreted to represent shell-penetrating setal canals; note the absence of the fine ridges in juvenile shell areas and the medial anterior sector. Dots in (2) and (3) indicate position of apex. Scale bar=1 cm.
Figure 15 (1) Diagrammatic representation of major axes of variation found by morphometric analyses of ventral valves from Mickwitzia occidens (Nevada, USA) and M. monilifera (Västergötland, Sweden); longer ‘lollipop’ indicates greater variation along that axis. (2) Percentage of shape variance encompassed by each respective principal component. (3) Results of discriminant analyses showing clear interspecies variation allowing covariance matrix to accurately recover species. (4) Plot of PC1 and PC2 displaying significant variation between M. occidens and M. monilifera ventral valves; ellipses correspond to 0. 95 frequency mean.
Figure 16 Diagrammatic reconstruction of Mickwitzia occidens in cross section; features not to scale. a, mantle cavity housing lophophore (not shown); b, coelomic cavity; c, lightly biomineralized, compact laminated phosphatic shell with inwardly pointing cones; d, distal cones (with canals open to the exterior?); e, proximal cones (with canals not reaching outer surface); f, reticulate shell ornament of juvenile shell; g, shell penetrating mantle setae; h, regular mantle setae; i, smooth apices of first-formed shell; j, acrotretid-like phosphatic lamellae with phosphatized cocci; k, pseudointerarea-like surface of dorsal valve.
Acknowledgments
We would like to thank J. S. and M. E. Hollingsworth for assistance in the field and G. Wife for support at the SEM. Constructive and thoughtful comments by N. Hughes, B. Pratt, M. Mergl, and L. Popov were much appreciated and helped to improve the final version of the article. The research has been funded by grants from the Swedish Research Council (VR) to MS and LEH and the National Science Foundation (EAR-0073089, 0106883) to LEB.