Introduction
Tommotiids are an extinct group of metazoans represented by organophosphatic sclerites (individual elements of multicomponent skeletons or scleritomes; Bengtson, Reference Bengtson1970, Reference Bengtson1985) of various shapes that are often very abundant in lower Cambrian deposits around the world (Rozanov et al., Reference Rozanov, Missarzhevsky, Volkova, Voronova, Krylov, Keller, Korolyuk, Lendzion, Michniak, Pykhova and Sidorov1969; Landing, Reference Landing1984; Missarzhevsky, Reference Missarzhevsky1989; Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990; Conway Morris and Chen, Reference Conway Morris and Chen1990; Esakova and Zhegallo, Reference Esakova and Zhegallo1996; Skovsted et al., Reference Skovsted, Balthasar, Brock and Paterson2009a, Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Pettersson Stolk and Patersonb, Reference Skovsted, Brock, Topper, Paterson and Holmer2011; Kouchinsky et al., Reference Kouchinsky, Bengtson, Runnegar, Skovsted, Steiner and Vendrasco2012). The sclerites of tommotiids are generally fairly distinct, but until recently they were exclusively known from disarticulated assemblages, the structure of the skeleton and body plan of the animal remained unknown, and tommotiids were commonly regarded as a phylogenetically problematic group. Reconstructions of the tommotiid animal were largely based on a worm- or slug-like model (e.g., Bengtson, Reference Bengtson1970, Reference Bengtson1977; Landing, Reference Landing1984; Evans and Rowell, Reference Evans and Rowell1990), and the discovery of the vagrant, slug-like, sclerite-bearing mollusk Halkieria evangelista Conway Morris and Peel, Reference Conway Morris and Peel1995, from the lower Cambrian Buen Formation of North Greenland provided apparent support for this model (Williams and Holmer, Reference Williams and Holmer2002; Ushatinskaya, Reference Ushatinskaya2002; Demidenko, Reference Demidenko2004; Li and Xiao, Reference Li and Xiao2004).
The recent discovery and detailed description of ontogenetically fused specimens of the tommotiid Eccentrotheca helenia Skovsted, Brock, Topper, Paterson and Holmer, Reference Skovsted, Brock, Topper, Paterson and Holmer2011 from lower Cambrian carbonates in South Australia by Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011) demonstrated that the scleritome of this tommotiid was constructed as a tube composed of vertically stacked sclerite rings. The scleritome exhibited an open-basal aperture that attached to hard substrates and the animal is interpreted to have been a sessile filter feeder related to modern lophophorates (i.e., Phoronida and Brachiopoda; Skovsted et al., Reference Skovsted, Brock, Paterson, Holmer and Budd2008, Reference Skovsted, Brock, Topper, Paterson and Holmer2011). Subsequently, the tannuolinid tommotiid Micrina etheridgei (Tate, Reference Tate1892) has been shown to share many characters (shell morphology, ultrastructure, and larval shell morphology) with Cambrian linguliform brachiopods (Holmer et al., Reference Holmer, Skovsted, Brock, Valentine and Paterson2008, Reference Holmer, Skovsted, Larsson, Brock and Zhang2011; Balthasar et al., Reference Balthasar, Skovsted, Holmer and Brock2009), whereas a third tommotiid taxon, Paterimitra pyramidalis Laurie, Reference Laurie1986 appears to be closely allied with paterinid brachiopods (Skovsted et al., Reference Skovsted, Balthasar, Brock and Paterson2009a; Topper et al., Reference Topper, Holmer, Skovsted, Brock, Balthasar, Larsson, Pettersson Stolk and Harper2013; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014), thus further strengthening the proposed tommotiid-lophophorate link.
Paterimitra pyramidalis exhibits three different sclerite types: (1) a bilaterally symmetrical, pyramidal sclerite (S1), (2) a triangular and bilaterally symmetrical (S2) sclerite, and (3) laterally compressed asymmetrical (L) sclerites (Skovsted et al., Reference Skovsted, Balthasar, Brock and Paterson2009a; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). Ontogenetically fused specimens of P. pyramidalis suggest a cone- or tube-like, bilaterally symmetrical scleritome. The basal part of the scleritome consists of one S1 sclerite and one S2 sclerite defining a circular posterior opening, and an unresolved number of L sclerites lining the distal margin of the S1/S2-composite (Skovsted et al., Reference Skovsted, Balthasar, Brock and Paterson2009a; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). The similarities between Paterimitra L sclerites and laterally compressed sclerites of Eccentrotheca suggest a similar arrangement of sclerites forming rings lining the distal margin of the Paterimitra scleritome (Skovsted et al., Reference Skovsted, Balthasar, Brock and Paterson2009a; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). The similarities in mode of shell formation, scleritome construction, and microstructure of Eccentrotheca and Paterimitra led Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011) to informally establish a group of tommotiid taxa referred to as the eccentrothecimorphs (a group that also includes Porcauricula Qian and Bengtson, Reference Qian and Bengtson1989, Sunnaginia Missarzhevsky in Rozanov et al., Reference Rozanov, Missarzhevsky, Volkova, Voronova, Krylov, Keller, Korolyuk, Lendzion, Michniak, Pykhova and Sidorov1969, Micrina Laurie, Reference Laurie1986, Tannuolina Fonin and Smirnova, Reference Fonin and Smirnova1967, and Kulparina Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990) and are considered to represent a group of fossils of key importance for our understanding of the evolution and early divergence of the lophophorate phyla Brachiopoda and Phoronida (see Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). The implication of these discoveries is that the brachiopod body plan, with its two bilaterally symmetrical valves, probably evolved through a stepwise specialization and reduction in the number of sclerites (Skovsted et al., Reference Skovsted, Brock, Paterson, Holmer and Budd2008, Reference Skovsted, Balthasar, Brock and Paterson2009a, Reference Skovsted, Brock, Topper, Paterson and Holmer2011; however, see Murdock et al., Reference Murdock, Bengtson, Marone, Greenwood and Donoghue2014 for a slightly different interpretation).
Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990 is a poorly documented tommotiid, endemic to the Arrowie and Stansbury basins in South Australia, which has recently been suggested to be a close relative of Paterimitra (Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011). The potentially pivotal role of Paterimitra in deciphering the evolution of brachiopods has prompted a detailed reinvestigation of K. rostrata to help elucidate the evolutionary changes within the tommotiid lineage that led to the establishment of the Brachiopoda.
Material and methods
The sclerites and ontogenetically fused specimens of Kulparina rostrata documented in this paper were retrieved from acetic-acid-resistant residues of samples from measured stratigraphic sections intersecting key lower Cambrian (Terreneuvian Stage 2 to Cambrian Series 2, Stage 3) shallow-water platform successions of the Wilkawillina Limestone at Wilkawillina Gorge in the Bunkers Graben (type section) and at Bunyeroo Gorge in the Heysen Range, Arrowie Basin, South Australia (Fig. 1). For detailed descriptions of localities, see Paterson and Brock (Reference Paterson and Brock2007), Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011), and Topper et al. (Reference Topper, Brock, Skovsted and Paterson2011a, Reference Topper, Brock, Skovsted and Patersonb). The material, which includes isolated sclerites and ontogenetically fused specimens, was examined under a binocular microscope. Selected specimens were mounted and coated with gold-palladium for Scanning Electron Microscope (SEM) imaging using a JEOL JSM 6480LA SEM at the Microscopy Unit in the Department of Biological Sciences, Macquarie University, Sydney; a Zeiss Supra35-VP SEM at the unit for Biological Structure Analysis (BSA), Uppsala University; or a Hitachi S-43000 at the Swedish Museum of Natural History (Naturhistoriska Riksmuseet, NRM), Stockholm. All illustrated specimens are housed in the paleontological collection of the South Australian Museum, Adelaide (SAMP).
Locality map. (A) Overview map of Australia, marking South Australia. AB, Arrowie Basin. (B) South Australia; dark-shaded areas indicating Arrowie and Stansbury basins, with ★1–5 indicating approximate positions of sampled sections and spot samples. ★1 area includes CD2, SYC101, and Minlaton-1; ★2 represents Horse Gully section; ★3 represents Bunyeroo Gorge; ★4 represents Wilkawillina Gorge and includes Wilkawillina limestone type section/section E and section H, and 10MS-W and 10MS sections; ★5 area includes Elder Range, Chace Range (CR-1) and Druid Range spot sample locality. YP, Yorke Peninsula. Modified from Bengtson et al. (Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990), Gravestock et al. (Reference Gravestock, Alexander, Demidenko, Esakova, Holmer, Jago, Lin, Melnikova, Parkhaev, Rozanov, Ushatinskaya, Zang, Zhegallo and Zhuravlev2001), and Jago et al. (Reference Jago, Zang, Sun, Brock, Paterson and Skovsted2006, Reference Jago, Gehling, Paterson, Brock and Zang2012).
Geological setting, sampled localities, and additional occurrences
Wilkawillina Limestone type section + section E, Wilkawillina Gorge, Bunkers Graben
The type section for the Wilkawillina Limestone was measured and defined by Brian Daily (Reference Daily1956) near the northwest margin of the Bunkers Graben. The fossiliferous packstones and grainstones of the section were previously described by Gravestock (Reference Gravestock1984) and Clarke (Reference Clarke1986a, Reference Clarkeb, Reference Clarkec, Reference Clarke1990). Samples yielding specimens of Kulparina rostrata were collected in 1982 by David Gravestock along the type section [Gravestock Reference Gravestock1984, section E, corresponding to PIRSA (Primary Industries and Regions South Australia) locality register, 6635RS samples; for details, see Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011: p. 258 and supplementary material]. The L sclerites of K. rostrata were retrieved from nine samples from section E: E13 (6635RS310) and E15–E22 (6635RS312–6635RS319). At least three of these samples also contained S sclerites of K. rostrata: E18 (6635RS315), E21 (6635RS318), and E22 (6635RS319). All these samples predate the Abadiella huoi Trilobite Zone of South Australian strata and probably correspond to Terreneuvian Stage 2 of the revised Cambrian timescale (Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011).
Wilkawillina Limestone, section H, Wilkawillina Gorge, Bunkers Graben
Section H was collected through the Wilkawillina Limestone less than 2 km south of the type section (=section E of Gravestock, Reference Gravestock1984), and some of Gravestock’s samples from this section contained abundant specimens of K. rostrata: samples H81, H83, H86, H87, H88, and H89, located respectively at 83, 125, 135, 146, 156, and 164 m true thickness above the base of the section. All six samples contained L sclerites, whereas S sclerites were restricted to samples H83, H86, and H87.
Bunyeroo 12, Daily sample, Bunyeroo Gorge
A single sample (Bun 12) from a suite of samples collected from the Wilkawillina Limestone at Bunyeroo Gorge yielded specimens of Kulparina rostrata. These samples were collected by Brian Daily in the early 1970s but, unfortunately, no stratigraphical information exists for the majority of these samples (for details see Skovsted et al. Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Pettersson Stolk and Paterson2009b, Reference Skovsted, Brock, Topper, Paterson and Holmer2011). Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011, fig. 5a, 5c) described laterally compressed sclerites from this sample as Eccentrotheca helenia, stressing the strong resemblance to columnar L sclerites of K. rostrata. This is the only sample where possible sclerites of K. rostrata and E. helenia have been recorded to co-occur, although the identification of these specimens as E. helenia might have to be revised (see discussion below).
Additional occurrences of Kulparina rostrata
In addition to the sections detailed above, sclerites of Kulparina rostrata have been reported from several sections, spot locality samples, and drill cores in South Australia. Bengtson et al. (Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990) recorded the occurrence of K. rostrata and Eccentrotheca guano Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990 in the Kulpara and lowermost Parara limestones of their Horse Gully section, Yorke Peninsula, Stansbury Basin samples 6429RS103 (K. rostrata and E. guano): L1858 (K. rostrata), 6429RS104 (K. rostrata and E. guano), 6429RS105 L1857 (K. rostrata and E. guano), and 6429RS106 L1761 L1856 (E. guano). Gravestock et al. (Reference Gravestock, Alexander, Demidenko, Esakova, Holmer, Jago, Lin, Melnikova, Parkhaev, Rozanov, Ushatinskaya, Zang, Zhegallo and Zhuravlev2001) reported E. guano and possible K. rostrata in the transition between the Kulpara and Parara limestones in their HG00/HG0 (Horse Gully) section, and E. guano from CD-2 (Kulpara Formation), SYC-101 (Parara Limestone), Minlaton-1 (Parara Limestone), all on Yorke Peninsula, Stansbury Basin. Two sclerites from HG0 referred to E. guano by Gravestock et al. (Reference Gravestock, Alexander, Demidenko, Esakova, Holmer, Jago, Lin, Melnikova, Parkhaev, Rozanov, Ushatinskaya, Zang, Zhegallo and Zhuravlev2001, pl. VII, figs. 12–13) were reinterpreted as sclerites of E. helenia by Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011). Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011) also recorded sclerites of K. rostrata in the Arrowie Basin: 10MS-W and 10MS sections, Wilkawillina Limestone, Bunkers Graben, L sclerites (E. guano), samples 215, 225, 235, 246.1, and 255.5; BALK section, Balcoracana Creek, Bunkers Range, L sclerites (E. guano), samples 340, 350, 362 (also containing S sclerites of K. rostrata), 370, 384, 390, 401, 410, 420, 430, 440, and 460. Topper et al. (Reference Topper, Holmer, Skovsted, Brock, Balthasar, Larsson, Pettersson Stolk and Harper2013) also reported presence of Kulparina sclerites in the Wirrapowie Limestone in Chace Range (CR-1) and Druid Range in the Arrowie Basin.
Kulparina rostrata and Eccentrotheca guano
The tommotiid species Eccentrotheca guano and Kulparina rostrata were both originally described from assemblages of disarticulated sclerites from the Kulpara and lower Parara limestones of the Horse Gully section in the Stansbury Basin of South Australia (Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990). The two forms share an extremely irregular surface sculpture and the morphological resemblance between them was already noted in the original description of K. rostrata (see Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, p. 136, figs. 86–91). The sclerites of K. rostrata were originally described as asymmetrical, occurring in dextral and sinistral symmetry variants, much like the sclerites of Sunnaginia (see Murdock et al., Reference Murdock, Donoghue, Bengtson and Marone2012). However, recently retrieved material, partly discussed by Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011) and described in detail herein, show that Kulparina sclerites were actually close to bilaterally symmetrical, with a morphology closely resembling S sclerites of Paterimitra. The sclerites described as E. guano by Bengtson (in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, figs. 71–73) resemble the L sclerites of Paterimitra in general morphology, although the E. guano sclerites are considerably more irregular and unpredictable in shape and mode of growth. Sclerites of E. guano and K. rostrata exhibit the same kind of irregular mode of growth, surface structure, and shell structure (Murdock et al., Reference Murdock, Bengtson, Marone, Greenwood and Donoghue2014), and sclerites of both taxa are usually recorded from the same samples. More conclusively though, the new material contains fused specimens representing natural associations of sclerites of E. guano and K. rostrata. Consequently, the two taxa appear to represent two separate sclerite morphs of a single biological species.
The original descriptions of Kulparina rostrata and Eccentrotheca guano were part of the same publication (Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, p. 136 and 120, respectively). Because both taxa are here reinterpreted to represent different parts of the same organism, the names are considered to be synonymous. The only applicable generic name for the revised species is Kulparina Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990 because the species cannot be accommodated in Eccentrotheca Landing, Nowlan, and Fletcher, Reference Landing, Nowlan and Fletcher1980. According to Article 24.2.2 of the International Code for Zoological Nomenclature (ICZN, 1999), the precedence of names published in the same work is the choice of the first reviser. The most characteristic sclerite in the species is the pyramidal S sclerite, which formed the basis for the original definition of K. rostrata, and acting as first revisers we hereby give precedence to this name, with the holotype SAMP 30698 from the Kulpara Limestone, Horse Gully (Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, fig. 90) as best representing the concept of the revised species.
Terminology and description
The sclerites of Kulparina rostrata show strong resemblance to sclerites of the recently re-described tommotiid Paterimitra pyramidalis (see Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). This is most obvious for sclerites originally described under the name K. rostrata. These sclerites have previously been considered to be asymmetrical (Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, p. 136), but material investigated herein demonstrates that they were bilaterally symmetrical (Fig. 2.1, 2.3, 2.5, 2.7) like the S sclerites of Paterimitra (Fig. 2.2, 2.4, 2.6, 2.8). The asymmetry of the sclerites described and illustrated by Conway Morris and Bengtson (in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, figs. 86–91) probably represent fragmentation, because protruding parts of the fragile, poorly preserved sclerites have been broken off on either the left or the right side. This certainly seems to be the case in the holotype (SAMP 30698; Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, fig. 90) in which the right lateral side of the sclerite appears to be missing. The previously used terminology (originally defined for and applied to sclerites of Sunnaginia by Landing et al., Reference Landing, Nowlan and Fletcher1980, fig. 5) was developed to illustrate the relationship between the sides (‘S1, S2, S3’) and lobes (‘L1, L2, L3’) of asymmetrical, essentially triangular sclerites (Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, fig. 91) and is consequently incompatible with the reinterpreted bilaterally symmetrical sclerites of Kulparina. Due to the strong similarities with P. pyramidalis, it is more appropriate to apply the terminology developed for this tommotiid (see Skovsted et al., Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Pettersson Stolk and Paterson2009b; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014) for Kulparina. This means that we refer to the symmetrical sclerites (K. rostrata sensu Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990) as S sclerites and the laterally compressed sclerites (E. guano sensu Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990) as L sclerites, despite the risk of confusion with the Sunnaginia terminology previously applied to Kulparina. In P. pyramidalis, the S sclerites occur in two distinct morphs, referred to as S1 and S2, respectively. In contrast, despite showing a large degree of morphological variation, distinct S sclerite morphs are not recognized in K. rostrata. The morphological terminology of S sclerites of K. rostrata is illustrated in Figure 2, where specimens, to enhance comparisons, are displayed together with corresponding sclerites of P. pyramidalis.
Terminology, S sclerites of Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990 and S1 sclerites of Paterimitra pyramidalis Laurie, Reference Laurie1986 from the Flinders Ranges, South Australia. (1, 3, 5, 7) S sclerites of K. rostrata, from sample E22 (6635RS319), Wilkawillina Limestone type section (Daily, Reference Daily1956)/section E (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben; (1) SAMP 44786, apical view; (3) SAMP P53721, posterior view; (5) SAMP 44786, lateral view; (7) SAMP P53722, anterior view. (2, 4, 6, 8) S1 sclerites of P. pyramidalis; (2) SAMP 46315, apical view, from sample AJX-M/267.5, Ajax Limestone, AJX-M section, Mt. Scott Range; (4) SAMP 47840, posterior view, from sample Bunyeroo 9, Wilkawillina Limestone, Bunyeroo Gorge; (6) SAMP 47846, lateral view, from sample WILK/S, Second Plain Creek Member, Wilkawillina Limestone, Wilkawillina Gorge type section (WILK), Bunkers Graben; (8) SAMP 47842, anterior view, from sample 10MS-W/390, Winnitinny Creek Member, Wilkawillina Limestone, 10MS-W section.
Kulparina rostrata comprises at least two sclerite types, S (Figs. 2.1, 2.3, 2.5, 2.7, 3) and L (Fig. 4), both of which exhibit a high degree of plasticity. All sclerites of K. rostrata grow by basal-internal accretion in highly irregular intervals, which is reflected externally in irregular, but prominent growth lamellae (Fig. 2.3, 2.5). The external surface is coarsely ornamented by irregular growth lines (Figs. 2.7, 3.2), whereas the internal surface is smooth to finely granular or pustular (Fig. 3.15).
S sclerites of Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990 from the Flinders Ranges, South Australia. (1, 2) SAMP 53723; (1) posterior view, box indicating area enlarged in (2); (2) close-up of (1), external shell structure. (3) SAMP 53724, posterior view. (4) SAMP 53725, posterior view. (5) SAMP 44786, posterior view. (6) SAMP 53726, cross section, anteroapical view. (7) SAMP 53727, posterior view. (8) SAMP 44788, posterior view. (9) SAMP 53728, posterior view, with damaged subapical flange. (10) SAMP 53729, anteroapical view. (11) SAMP 53730, posterior view. (12) SAMP 53731, posterior view. (13) SAMP 53732, posterior view. (14, 15) SAMP 53733; (14) posterior view, box indicating enlarged area in (15); (15) close-up of boxed area in (14), internal surface structure. Specimens in (1, 2, 9) from sample Bunyeroo 12, Wilkawillina Limestone, Bunyeroo Gorge; (3, 5–8, 11) from sample E22 (6635RS319), Wilkawillina Limestone type section (Daily, Reference Daily1956)/section E (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben; (4, 10) from sample H83, Wilkawillina Limestone, section H (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben; (12) from sample E21 (6635RS318), Wilkawillina Limestone type section (Daily, Reference Daily1956)/section E (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben; (13–15) from sample E18 (6635RS315), Wilkawillina Limestone type section (Daily, Reference Daily1956)/section E (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben.
S sclerites
Typical S sclerites of Kulparina rostrata are characterized by their broad bilateral symmetry (Figs. 2.1, 2.3, 3.1, 3.3–3.14). The sclerites are pyramidal to triangular (apical angle ~90º) with a concave anterior side, which results in a compressed outline in lateral view and a cross section reminiscent of a broad-based isosceles triangle (Figs. 2.1, 2.5, 3.6). The apex (Fig. 2.3) is slightly displaced toward one side, which, in accordance with the Paterimitra terminology, is referred to as the posterior (Fig. 2.1; corresponding to the direction of lobe L3 of Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, fig. 91) and in well-preserved specimens is drawn out into a prominent subapical flange (Fig. 2.1, 2.3, 2.5; rostrum L3 sensu Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, fig. 91, p. 142).The flange can project slightly beyond the posterior side but can alternatively be developed merely as a zone of apically deflected growth laminae on the posterior field (Fig. 3.12–3.14). The posterior side is highly variable in shape but is typically dominated by the subapical flange and exhibits a wide triangular notch (Fig. 2.3). Laterally, the flanks of the sclerite are formed by two narrow, elongate lateral plates (Fig. 2.3, 2.5), strongly flexed toward the posterior. One or both of the lateral plates are often damaged, which in many specimens results in apparent asymmetry (Figs. 2.1, 2.3, 3.3, 3.8–3.14; Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, figs. 86–90). The anterior edges of the lateral plates form two radial, ridge-shaped anterior boundaries bordering a concave, subtriangular anterior plate (Fig. 2.1, 2.7) with a curved anteroapical edge that defines a widely U-shaped anterior sinus (Fig. 2.7). Some sclerites are strongly flattened in the anterior-posterior direction (Fig. 3.13, 3.14). S sclerites of K. rostrata range in width (defined as the distance between the distal edges of the lateral plates) from a few hundred microns up to ~1.5 mm. The height of S sclerites generally ranges up to ~1.5 mm, although extremely large sclerites exceeding 2 mm as well as comparably low sclerites (0.6 mm) do occur. The width/height ratio ranges from ~0.6 to 1.8, although most specimens exhibit a width ~0.8 of the height.
L sclerites
L sclerites of Kulparina rostrata are typically high, often strongly compressed sclerites that exhibit extremely irregular morphology (Fig. 4). Single sclerites range from simple columnar or triangular forms (Fig. 4.1–4.4, 4.19), through palm- or finger-like forms (Fig. 4.5–4.9, 4.15), to somewhat exaggerated, irregular morphologies (Fig. 4.16, 4.21, 4.23). Less flattened, almost cone-shaped sclerites also occur (Fig. 4.14). For this reason, general patterns in sclerite morphology are difficult to define. The L sclerites can be straight, but more frequently they exhibit a slight to strong curvature, both in the plane of flattening and perpendicular to it (Fig. 4.1–4.2, 4.8, 4.10–4.12, 4.15–4.16, 4.19). Changes in growth direction occur frequently through the growth of individual sclerites (Fig. 4.3, 4.6, 4.8). The length of the L sclerites usually exceeds the width, occasionally by up to 10 times (Fig. 4.3). The width of the sclerite base generally expands evenly throughout growth, but the expansion can in some cases be temporarily halted, resulting in straight sclerite sides (Fig. 4.4, 4.9–4.10, 4.17) and some sclerites even exhibit apparent phases of base constriction (Fig. 4.21) or in extreme cases restriction of the basal cavity (Fig. 4.20, 4.22, 4.24). Fusion of adjacent L sclerites during growth into complex, sometimes digitate composites is a common feature and the majority of specimens represent fusion of at least two incipient sclerites with the fused sclerites often intimately entwined with each other. In flattened sclerites, fusion usually occurs in the plane of flattening, resulting in either long, narrow composites (Fig. 4.4–4.5) or broad, large sclerites with curved bases (Fig. 4.18, 4.20, 4.23–4.24). Conical or tubular sclerites are usually fused along their shortest sides (Fig. 4.14). L sclerites of K. rostrata range in width from ~100 µm up to 1.5 mm (small sclerites ~50 µm wide occur in articulation with S1 sclerites, see section below). The length of the L sclerites generally ranges up to ~1.5 mm, although extremely large sclerites exceeding 2 mm are known.
L sclerites of Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990 from the Flinders Ranges, South Australia. (1) SAMP 44770, lateral view. (2) SAMP 44769, lateral view. (3) SAMP 44773, lateral view. (4) SAMP 44772, lateral view. (5) SAMP 44771, lateral view. (6) SAMP 44775, lateral view. (7) SAMP 53734, lateral view. (8) SAMP 44777, lateral view. (9) SAMP 53735, lateral view. (10) SAMP 44776, lateral view. (11) SAMP 53736, oblique lateral view. (12) SAMP 53737, oblique lateral view. (13) SAMP 53738, lateral view. (14) SAMP 53739, lateral view. (15) SAMP 53740, lateral view. (16) SAMP 53741, lateral view. (17) SAMP 44779, lateral view. (18) SAMP 44782, lateral view. (19) SAMP 53742, lateral view. (20) SAMP 53743, lateral view. (21) SAMP 53744, lateral view. (22) SAMP 44780, lateral view. (23) SAMP 53745, lateral view. (24) SAMP 44783, lateral view. Specimens in (1–6, 8, 10, 16–18, 22, 24) from sample E22 (6635RS319), Wilkawillina Limestone type section (Daily, Reference Daily1956)/section E (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben; (7, 11–12, 14–15, 19–21) from sample H83, Wilkawillina Limestone, section H (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben; (9, 23) from sample H87, Wilkawillina Limestone, section H (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben; (13) from sample H81, Wilkawillina Limestone, section H (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben.
Ontogenetically fused specimens and implications for scleritome architecture
The identity of S and L sclerites as part of the scleritome of Kulparina rostrata is inferred by their co-occurrence in several samples from various sections and by the strong similarities in microstructure (Murdock et al., Reference Murdock, Bengtson, Marone, Greenwood and Donoghue2014), growth mode, and surface ornament. However, the strongest support is provided by the occurrence of rare, ontogenetically fused specimens exhibiting L sclerites directly attached to the distal margins of S sclerites (Fig. 5). The attached L sclerites in these composites are considerably smaller than any recorded individual L sclerite, but apart from the size discrepancy, they exhibit morphological features identical to the isolated L sclerites described above. The absence of individual small L sclerites can be explained by taphonomic sorting or selective sampling and sorting during the fossil extraction process.
Ontogenetically fused S and L sclerite composites of Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990 from the Flinders Ranges, South Australia. (1–4) SAMP 44787, two L sclerites fused to lateral plate of S1 sclerite; (1) posterior view, box indicating area enlarged in (2); (2) close-up of (1); (3) anteroapical view, box indicating area enlarged in (4); (4) close-up of (3). (5–7) SAMP 44785, L sclerites fused along anterior sinus of S1 sclerite; (5) anterior view; (6) apical view, box indicating area enlarged in (7); (7) close-up of (6). (8–10) SAMP 53746, L sclerite fused along anterior sinus of S1 sclerite; (8) apical view; (9) anterior view; (10) lateral view. (11–13) SAMP 53747, three L sclerites fused along anterior sinus of S1 sclerite; (11) anterior view, box indicating area enlarged in (12); (12) close-up of (11); (13) apical view. Specimens in (1–4, 8–13) from sample E22 (6635RS319), Wilkawillina Limestone type section (Daily, Reference Daily1956)/section E (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben; (5–7) from E18 (6635RS315), Wilkawillina Limestone type section (Daily, Reference Daily1956)/section E (Gravestock, Reference Gravestock1984), Wilkawillina Gorge, Bunkers Graben.
The L sclerites in the composites are attached to the S sclerites either along the posterior margin (Fig. 5.1–5.4) or along the margin of the anterior plate (Fig. 5.5–5.13). This pattern is very similar to the way L sclerites attach to S1 sclerites of Paterimitra (Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014, fig. 16A–H). In SAMP 44787 (Fig. 5.1–5.4), two minute, elongated L sclerites are fused to the distal edge of the posteriorly facing side of the lateral plate in an extremely high Kulparina S sclerite (compare Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014, fig. 16). In SAMP 44787 (Fig. 5.5–5.7), SAMP 53746 (Fig. 5.8–5.10), and SAMP 53747 (Fig. 5.11–5.13), L sclerites of variable morphology, but generally with a broader base, are fused to the margin of the anterior sinus of the S sclerite and appear to overlap the anterior plate (compare Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014, fig. 17). All L sclerites fused to S sclerites are oriented in the same general direction, with their apices directed toward the apex of the S sclerite. This corresponds closely to the observed direction of L sclerites in composites of P. pyramidalis (Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014, figs. 16–17).
Two large specimens, consisting of elongate plate-like structures with an arcuate base, appear to have been formed by the coalescence of multiple L sclerites (Fig. 6). The general morphology of these specimens is suggestive of longitudinal sections of a tube and the margins of the specimens appear to have continued to grow as a unit after the merging of the constituent sclerites. Individual sclerites in these composites differ dramatically both within and between specimens. In SAMP 53748, a large number of incipient sclerites of vastly differing sizes and morphologies are present in a roughly trapezoidal structure ~1.5 mm long and 1 mm wide (Fig. 6.1–6.6). The majority of sclerites are narrow, almost straight-sided, either curved or straight, but a single large, broad sclerite with multiple apices are present in the center of the composite. The sclerites partly overlap, but the larger sclerites are separated quite substantially from each other (~60–70 µm), often leaving spaces between sclerites and the underlying surface (Fig. 6.2, 6.3, 6.5). Although all sclerites are inclined in the same general direction and are oriented toward the wider of the short sides of the composite, the curvature of longer sclerites appears to be random with sinistral and dextral sclerites overlapping. The roughly quadrate SAMP 53749 is a composite specimen of similar size but is composed of fewer sclerites (Fig. 6.7–6.10). The bulk of the composite is composed of three, large broad sclerites, each with multiple apices. One side displays two more or less identical and partly overlapping broad sclerites with two apices whereas the opposing side is dominated by a single, broad sclerite apparently lacking an apex. Smaller sclerites are also present in two zones at either end of the composite. A very similar sclerite arrangement has been observed in L sclerite composites of Paterimitra (Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014, fig. 15).
Ontogenetically fused L sclerite composites forming arched plate-like structures of Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, from sample 6635RS319, Wilkawillina Limestone type section, Wilkawillina Gorge, Bunkers Graben, Flinders Ranges, South Australia. (1–6) SAMP 53748; (1) overview; (2, 5) lateral views; (3, 6) oblique views of curved margin; (4) oblique view of opposite margin. (7–10) SAMP 53749; (7, 9, 10) oblique views of curved margins; (8) overview.
The striking similarities between ontogenetically fused S/L as well as L/L composite specimens of Kulparina and Paterimitra leads to the conclusion that the scleritome of K. rostrata was constructed in a similar sclerite arrangement as suggested for P. pyramidalis (Skovsted et al., Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Pettersson Stolk and Paterson2009b, Reference Skovsted, Brock, Topper, Paterson and Holmer2011; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). The main difference between the two taxa is the apparent lack of a distinct S2 sclerite morph in K. rostrata. S sclerites of K. rostrata do vary considerably in morphology and some individual sclerites approach a general shape that could be morphologically comparable to the S2 sclerites of Paterimitra (Fig. 3.6, 3.9, 3.13, 3.14). However, morphological variation among S sclerites appears gradational and no distinct morphs can be definitely recognized. There could be a number of potential reasons for their absence; S2 sclerites could potentially be morphologically similar to S1 sclerites and consequently have been misidentified; the absence of S2 sclerite morphs could be because of sampling bias, either by taphonomic factors or during acid preparation, because they can be much smaller in size compared to the S1 sclerite morph, similar to the case in Paterimitra (see Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014); or Kulparina does not possess a S2 sclerite morph and the irregular L sclerites could have articulated to form a suitable ‘posterior opening’ (similar to the opening in Paterimitra) to facilitate attachment to the substrate. Regardless, the morphological similarities between Kulparina and Paterimitra are compelling and following the Paterimitra model, the base of the Kulparina scleritome was composed of one or a pair of bilaterally symmetrical S sclerites with the remainder of the scleritome composed of L sclerites that formed a conical or tube-shaped structure. L sclerites lined the anterior margins of the S sclerite/sclerites in subsequent rows and were arranged with the apices directed toward the S sclerite and partly overlapping each other, presumably resulting in a robust scleritome.
Discussion
Ontogenetically fused tommotiid specimens are notoriously rare in the fossil record and Kulparina rostrata represents just the third tommotiid taxon (after Paterimitra pyramidalis and Eccentrotheca helenia) in which ontogenetically fused sclerites representing a sizable part of the scleritome have been recovered. Specimens composed of a few adjacent sclerites are also known in Lapworthella Cobbold, Reference Cobbold1921 (Landing, Reference Landing1984; Demidenko, Reference Demidenko2004), Sunnaginia (Landing, Reference Landing1995), and Tannuolina (Qian and Bengtson, Reference Qian and Bengtson1989; Li and Xiao, Reference Li and Xiao2004), as well as from E. kanesia Landing, Nowlan, and Fletcher, Reference Landing, Nowlan and Fletcher1980 (Landing et al., Reference Landing, Nowlan and Fletcher1980; Landing, Reference Landing1995), but these specimens yield little information on scleritome construction. The scleritome of E. helenia is a slowly expanding tube, constructed through the fusion of discrete basal broad, cap-shaped sclerites and laterally compressed triangular sclerites (Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011). The individual sclerites are ontogenetically fused into ring-shaped units that are sequentially stacked vertically, forming a tube (Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011). The scleritome of P. pyramidalis represents a slightly more complex system, with a small triangular sclerite (S2) seated within the notch of a large pyramidal sclerite (S1) with a currently unresolved number of laterally compressed sclerites (L) flanking the margins and posterior and anterior sinuses of the S sclerites, producing an open, tube-like construction (Skovsted et al., Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Pettersson Stolk and Paterson2009b; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). Both taxa are interpreted as sessile filter-feeding organisms, most likely attaching to hard substrates via organic structures emerging through the apical perforations (Skovsted et al., Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Pettersson Stolk and Paterson2009b, Reference Skovsted, Brock, Topper, Paterson and Holmer2011; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014).
Both Paterimitra and Eccentrotheca have been directly compared with the Brachiopoda, based on overall morphological and skeletal microstructural details (Skovsted et al., Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Pettersson Stolk and Paterson2009b, Reference Skovsted, Brock, Topper, Paterson and Holmer2011; Balthasar et al., Reference Balthasar, Skovsted, Holmer and Brock2009; Topper et al., Reference Topper, Holmer, Skovsted, Brock, Balthasar, Larsson, Pettersson Stolk and Harper2013; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). The pervasive shell-penetrating scaffolding of polygonal layers has only been detected in eccentrothecimorph tommotiids and the paterinid brachiopod Askepasma Laurie, Reference Laurie1986 (Balthasar et al., Reference Balthasar, Skovsted, Holmer and Brock2009; Topper et al., Reference Topper, Holmer, Skovsted, Brock, Balthasar, Larsson, Pettersson Stolk and Harper2013) and the possible ancestral chileate brachiopod Salanygolina Ushatinskaya, Reference Ushatinskaya1987 (Holmer et al., Reference Holmer, Pettersson Stolk, Skovsted, Balthasar and Popov2009), suggesting a similar mode of cyclic shell secretion in eccentrothecimorph tommotiids, paterinid brachiopods, and perhaps stem rhynchonelliform brachiopods (Holmer et al., Reference Holmer, Pettersson Stolk, Skovsted, Balthasar and Popov2009, Reference Holmer, Skovsted, Larsson, Brock and Zhang2011). Kulparina rostrata is reported to share the same skeletal architecture as Eccentrotheca, Paterimitra, and Askepasma, firmly rooting the taxa within the eccentrothecimorphs and in the stem of the Brachiopoda (Murdock et al., Reference Murdock, Bengtson, Marone, Greenwood and Donoghue2014). The overall scleritome configuration of Kulparina is not as clear as in Paterimitra and Eccentrotheca due to the lack of more complete articulated specimens, although the intimate association of S and L sclerites is clear (Fig. 5).
Kulparina, like Paterimitra, possesses a combination of Eccentrotheca-like sclerites organized in a tubular scleritome displaying specialized S and L sclerite morphs. However, the individual sclerites of Kulparina, when compared to those of Paterimitra, display a much more irregular and variable morphology without the distinctive polygonal micro-ornament and the scleritome seems to lack the opposing symmetrical S2 sclerite morph. This implies that Kulparina most probably represents an intermediate form on the stem of the brachiopod clade between Eccentrotheca and Paterimitra (see Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011, text-fig. 21), introducing bilateral symmetry to the scleritome, coupled with a reduction in number and the specialization of basal sclerites, however not yet reaching the scleritome complexity and sclerite specialization seen in Paterimitra.
Systematic paleontology
Order Tommotiida Missarzhevsky, Reference Missarzhevsky1970
Family uncertain
Genus Kulparina Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990
Type species
Type and only species Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990.
Emended diagnosis
Tommotiid with scleritome comprising two distinctive sclerite types: symmetrical pyramidal S sclerites and high, laterally compressed, asymmetrical L sclerites. S sclerites with elongated, narrow lateral plates, strongly flexed toward posterior. Anterior plate subtriangular to slightly U-shaped. L sclerites high, irregular, extremely variable, but generally with strong lateral compression and tendency toward ontogenetic fusion.
Differing from Sunnaginia and Eccentrotheca by presence of bilaterally symmetrical sclerites. Differing from Paterimitra by triangular cross section and concave anterior plate in S sclerites, lack of differentiated S2 sclerites, highly irregular morphology of L sclerites, and absence of polygonal surface micro-ornament.
Occurrence
Lower Cambrian of South Australia (Terreneuvian, Stage 2).
Discussion
The sclerites of Kulparina exhibit many morphological similarities to sclerites of Paterimitra, such as the subapical flange, lateral plates with anterior boundaries, and the anterior plate of the high, pyramidal S sclerites. The manner in which L sclerites of Kulparina are fused to S sclerites (Fig. 5) is also very similar to the way L and S1 sclerites of Paterimitra are fused (Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). However, there are numerous obvious differences between Paterimitra and Kulparina. The S sclerite of Kulparina is generally angled toward the posterior and the anterior plate is concave, resulting in a triangular cross section (Fig. 2.1), whereas the S1 sclerite of Paterimitra has more anteriorly angled lateral plates and a convex anterior plate, resulting in a rectangular to trapezoidal outline in cross section (Fig. 2.2, 2.6). The anterior boundaries of Kulparina are generally developed as ridges (Fig. 2.5, 2.7), whereas the anterior boundaries of Paterimitra are commonly developed as furrows (Fig. 2.8). Paterimitra also possesses a distinct triangular or saddle-shaped S2 sclerite, a sclerite morph absent in Kulparina. Of particular note is the absence of the unmistakable Paterimitra polygonal surface micro-ornament in Kulparina sclerites. The morphology, plasticity, and external surface of L sclerites of Kulparina are more reminiscent of the laterally compressed sclerites of Eccentrotheca. L sclerites of Kulparina also lack the characteristic apical twist possessed by L sclerites of Paterimitra and the laterally compressed sclerites of Eccentrotheca.
Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990
Figures 2.1, 2.3, 2.5, 2.7, 3–6
1990 Eccentrotheca guano Bengtson in Bengtson et al., p. 120, figs. 71–73.
1990 Kulparina rostrata Conway Morris and Bengtson in Bengtson et al., p. 136–142, figs. 86–91.
2001 Eccentrotheca guano Bengtson in Bengtson et al.; Demidenko in Gravestock et al., p. 230–231, pl. VII, fig. 11.
2011 Kulparina sp.; Skovsted et al., p. 275–281, text-figs. 18–19.
?2011 Eccentrotheca helenia; Skovsted et al., text-fig. 5a, c.
2014 Kulparina rostrata; Murdock et al., fig. 4A–D.
2014 ‘Eccentrotheca guano’; Murdock et al., fig. 4A–D.
Diagnosis
As for genus by monotypy.
Holotype
SAMP 30698 (Bengtson et al., Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990, figs. 88F, 90A–H) from UNEL1858, Horse Gully, Kulpara Limestone.
Occurrence
Lower Cambrian pre-trilobitic strata (Terreneuvian, Stage 2) of Wilkawillina and Wirrapowie limestones, Flinders Ranges, Arrowie Basin, and upper Kulpara, and lowermost Parara limestones, Yorke Peninsula, Stansbury Basin, South Australia.
Discussion
Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011) previously noted the mutually exclusive stratigraphic ranges of Kulparina rostrata (including sclerites originally described as Eccentrotheca guano) and E. helenia in all investigated lower Cambrian sections in South Australia. Kulparina rostrata has a range which is entirely subtrilobitic in South Australia and likely correlates with Terreneuvian, Stage 2. Betts et al. (Reference Betts, Topper, Valentine, Skovsted, Paterson and Brock2014) reported that whereas the first appearance of E. helenia occurs slightly below the incoming of Abadiella huoi Chang, Reference Chang1966 (eponym of the oldest trilobite zone in East Gondwana), the majority of its range overlaps with the A. huoi Trilobite Zone. The only sample that apparently contains sclerites of both K. rostrata and E. helenia is sample Bun 12 from Bunyeroo Gorge in the Heysen Range, originally collected by Brian Daily in the 1960s. However, Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011) pointed out that the specimens from this sample identified as E. helenia are distinctly higher and more columnar than the majority of specimens from other localities and are very similar to columnar, laterally compressed L sclerites of K. rostrata. Consequently, it seems most likely that these sclerites actually represent specimens of K. rostrata. The exact stratigraphical position of sample Bun 12 remains difficult to pinpoint, because Daily’s original field notebook covering this section is missing (presumed lost). It is thus impossible to resample this level to obtain more sclerites that would likely resolve this question. Subsequent sampling along new stratigraphic sections intersecting the Wilkawillina Limestone (and equivalents) in northern and central Flinders Ranges (Betts et al., in review) provides clear evidence of a substantial stratigraphical separation between K. rostrata and E. helenia.
Specimens from Todd River Dolomite of central Australia originally described as Eccentrotheca kanesia (see Laurie, Reference Laurie1986) and later referred to E. guano by Bengtson in Bengtson et al. (Reference Bengtson, Conway Morris, Cooper, Jell and Runnegar1990) probably represent L sclerites of Paterimitra pyramidalis (see Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014). Similarly, specimens from the Kulpara and Parara limestones at Horse Gully (HG0), described as E. guano by Demidenko in Gravestock et al. (Reference Gravestock, Alexander, Demidenko, Esakova, Holmer, Jago, Lin, Melnikova, Parkhaev, Rozanov, Ushatinskaya, Zang, Zhegallo and Zhuravlev2001), were considered to belong to E. helenia by Skovsted et al. (Reference Skovsted, Brock, Topper, Paterson and Holmer2011).
Although Kulparina rostrata precedes Eccentrotheca helenia in all known stratigraphic sections in South Australia, Eccentrotheca sclerites (for discussion, see Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011) have been documented from probably older Terreneuvian (Fortunian or basal Cambrian Stage 2) strata in Newfoundland (E. kanesia Landing et al., Reference Landing, Nowlan and Fletcher1980, Reference Landing, Myrow, Benus and Narbonne1989, Reference Landing, Geyer, Brasier and Bowring2013; Bengtson and Fletcher, Reference Bengtson and Fletcher1983; Hinz, Reference Hinz1987; Landing, Reference Landing1995), supporting its potential basal position on the brachiopod stem. The majority of sclerites of E. kanesia are high, triangular, laterally compressed sclerites that are similar to the L sclerites of K. rostrata in ornament and general morphology (compare, for example, the specimens in Fig. 4 with Hinz, Reference Hinz1987, pl. 15, figs. 1–7, 12). However, sclerites of E. kanesia differ from Kulparina L sclerites in the more regular triangular shape and the lower height/width ratio. In Avalonia, E. kanesia also typically occurs together with broader, cap-shaped sclerites [Landing et al., Reference Landing, Nowlan and Fletcher1980, pl. 1, fig. 13; Landing, Reference Landing1984, fig. 4B; Reference Landing1995, fig. 7.23, fig. 6.1–6.13 (cap-shaped sclerites referred to the potentially synonymous genus Jaycea Landing, Reference Landing1995)]. These are closely comparable to the low, cap-shaped sclerites of E. helenia (see Skovsted et al., Reference Skovsted, Brock, Topper, Paterson and Holmer2011, text-fig. 7). No such sclerites have been recovered in association with K. rostrata in our collections.
The relatively poor preservation of the material in our collections hinders studies of the internal shell microstructure, but SRXTM data presented by Murdock et al. (Reference Murdock, Bengtson, Marone, Greenwood and Donoghue2014, p. 20, fig. 4) shows that the internal shell structure of K. rostrata and E. guano is identical, supporting the synonymy herein. The shell structure is closely comparable to the microstructure documented for Paterimitra, Eccentrotheca, and Askepasma (Balthasar et al., Reference Balthasar, Skovsted, Holmer and Brock2009; Topper et al., Reference Topper, Holmer, Skovsted, Brock, Balthasar, Larsson, Pettersson Stolk and Harper2013; Murdock et al., Reference Murdock, Bengtson, Marone, Greenwood and Donoghue2014; Larsson et al., Reference Larsson, Skovsted, Brock, Balthasar, Holmer and Topper2014), strengthening the suggested relationship between eccentrothecimorph tommotiids and paterinid brachiopods (Skovsted et al., Reference Skovsted, Balthasar, Brock and Paterson2009a, Reference Skovsted, Holmer, Larsson, Högström, Brock, Topper, Balthasar, Pettersson Stolk and Patersonb, Reference Skovsted, Brock, Topper, Paterson and Holmer2011).
Acknowledgments
The first draft of this manuscript was compiled as part of the Ph.D. thesis of CML and the manuscript was partly revised during visiting scholarships to the Swedish Museum of Natural History (Stockholm, Sweden) by TPT and the Institute of Advanced Studies, Durham University (Durham, Great Britain) by GAB. The work was financed by grants from the Swedish Research Council (to CBS, nr VR 2006-6039, LEH, nr VR 2009-4395, 2012-1658), the Australian Research Council Discovery Project #120104251 and Wenner Gren Foundation (to GAB), the European Community Research Infrastructure Action SYNTHESYS (to TPT, project number SE-TAF-2649), from Uppsala University and the Swedish Royal Academy of Science (to CML). We also express our thanks to J.R. Laurie (Canberra) for sharing the material from D. Gravestock’s section H, and to D. Rice (Adelaide) for making specimens from the B. Daily collections available. U. Balthasar (Glasgow) is thanked for advice and valuable discussions on tommotiid and early brachiopod evolution. S. Bengtson (Stockholm) and G. Budd (Uppsala) are thanked for advice regarding systematic procedures and zoological nomenclature. Thanks to E. Landing (Albany) and J.L. Moore (Santa Barbara) for astute and constructive reviews of the manuscript.
