Ichnofossils associated with lingulide shells
from the Lower Permian of Brazil

The Lingulidae Family is part of the Lingulida Order (Holmer & Popov, 2000). The lingulide brachiopods have chitino phosphatic shells (Holmer & Popov, 2000). They have a stratigraphic range from the early Cambrian to Recent (Holmer & Popov, 2000). The data on soft body anatomy of Early Cambrian lingulides from the Chengjiang Fauna (Zhang et al., 2003) include preservation of a pedicle, schizolophous lophophore, as well as a recurved, U-shaped digestive tract with a functional anus showing that major features of soft body anatomy characteristic of Recent lingulates were already present in early Cambrian. Sometimes fossil lingulide shells contain borings.

The earliest predatory borings have been described in Cloudina from the Ediacaran (Bengtson & Zhao, 1992). The Paleozoic record of predation is mostly based on drillings known from the shells of both lingulate and rhynchonellate brachiopods, and a lesser amount from mollusks (Brett & Walker, 2002; Brett, 2003; Harper, 2003; Huntley & Kowalewski, 2007). Vermeij (1977) has shown that predation was an important evolutionary factor since the early Cambrian. It is well known that predation intensities have increased from the Cambrian to the Recent (Vermeij, 1977, 1987; Vermeij et al., 1981). This increase in predation intensities is mirrored by the morphological evolution of the gastropod shell (Vermeij, 1977, 1987; Vermeij et al., 1981).

Often fossilized shells of various invertebrates contain burrows of the other invertebrates. The ichnofossil Arachnostega was first systematically described and named by Bertling (1992) accommodating irregular, branching burrows on the surface of internal moulds of various shelly fossils (e.g., Vinn et al., 2014). Arachnostega presumably belongs to the category fodichnia (Bertling, 1992), but it is equally possible that it in fact represents a dwelling burrow (domichnion) (Fatka et al., 2011). The exact zoological affinities of the tracemakers that produced Arachnostega and their paleoecology are not yet fully understood (Vinn et al., 2018).

The aims of the present note are threefold; firstly, to describe bioerosion of early Permian lingulide shells; secondly, to describe predatory borings in early Permian lingulides and, thirdly, to document the ichnogenus Arachnostega from Lower Permian strata in Brazil for the first time.

2. Geological background and outcrop

Lingulide shells were collected from the Campáleo outcrop, which is located in the city of Mafra in the State of Santa Catarina, Brazil (Fig. 1 ). The outcrop is positioned in the Lontras Shale (Fig. 2 ), upper part of Campo Mourão Formation and eastern border of Paraná Basin. The Campáleo outcrop (UTM 22J 0.618.473, 7.106.243) comprises the first 10 m of the Lontras Member (Mouro et al., 2017). At the base of outcrop there is a 1-m-thick silty bioturbated argillite bed (i.e., the Glossifungites suite) and is overlain by a fossiliferous, 1.10-m-thick black shale layer (i.e., the Lontras Shale), followed by a non-fossiliferous, 7-m-thick shelly rhythmite layer (Mouro et al., 2017). Lontras Shale can be informally divided into four levels, including sublevels, distinguished on the bases of rock features, fossil content, and pyrite concentration (Mouro et al., 2016, 2017). Based on the quantity and preservation of the fossils that have been recovered in the Lontras Shale and are exposed at the Campáleo outcrop, this site has been identified as an important lower Permian (Cisuralian, 299 myr to 275 myr) Fossil-Lagerstätte (Ricetti et al., 2016; Mouro et al., 2016, 2017). Lontras Shale is interpreted as a deglaciation phase. The paleoenvironment is interpreted as a restricted marine area, very close to the coast (Ricetti et al., 2016; Mouro et al., 2016, 2017).

Click on thumbnail to enlarge the image.

Figure 1: Locality map (modified after Ricetti et al., 2016).

Click on thumbnail to enlarge the image.

Figure 2: Stratigraphic position of Lontras Shale in the Lower Permian of Paraná Basin (modified after Ricetti et al., 2016).

3. Material and methods

A small collection of 58 lingulide shells was searched for associated ichnofossils. Shells with signs of bioerosion, predatory borings and cryptic burrows were selected for photographing. Photographing was carried out using a Zeiss digital camera. Dimensions of ichnofossils were measured from calibrated photos.

4. Results

About 24% of 58 lingulides studied showed signs of bioerosion in form of multiple small shallow pits (Fig. 3 ). This type of pits resemble incomplete Oichnus paraboloides borings. Their diameters of aperture have a range from 0.03 mm to 0.17 mm. The diameter of pits in a single lingulid valve is largely variable. Also, some shells are completely covered with small pits; borings do not show preference for special shell regions.

Click on thumbnail to enlarge the image.

Figure 3: Small rounded pits in the ligulid shell (arrows) indicating the bioerosion.

A single lingulide valve shows a centrally located large circular Oichnus simplex boring with 0.67 mm wide (Fig. 4 ). Same valve also exhibits some traces of bioerosion.

Click on thumbnail to enlarge the image.

Figure 4: Lingulide valve shows a centrally located large circular Oichnus simplex boring (Oich.).

Almost 21% of the lingulide shells show small Arachnostega traces in their interior (Fig. 5 ). Arachnostega traces are visible through somewhat translucent shell material as a colour contrast. The diameter of burrows is relatively constant and ranges from 0.08 mm to 0.09 mm. These traces are more similar to juvenile stages of Arachnostega gastrochaena and do not form well-developed meshwork of tunnels. The Arachnostega burrows are filled with lighter colored sediment than the dark grey rock matrix.

Click on thumbnail to enlarge the image.

Figure 5: Small Arachnostega (Ar.) traces gleaming from the interior of lingulide shells.

5. Discussion and conclusions

Bioerosion

Dead or/and living lingulide shells were encrusted by phosphate boring organisms that needed a domicile. Considering the shallow depth of the pits in lingulides, it is likely that the soft body of organisms responsible for pit formation were only partially located within their borings. Thus, the function of these pits may have been rather enabling encrusters stronger contact with substrate than providing a shelter against predators for a true endobiont. It is interesting to notice that bioerosion in phosphatic substrates is rare when compared to calcareous ones (Vinn & Toom, 2016), which partially can be explained by study bias, as most of skeletons of marine organisms are calcareous. The earliest boring in phosphatic substrates are known from the early Cambrian (Vinn & Toom, 2016). The domiciles bored by worm-like organisms into calcareous brachiopod shells are common since the Middle Ordovician (Vinn, 2004, 2005). They have usually been classified as Trypanites or Vermiforichnus (Vinn, 2005). The latter two are much deeper than shallow pits in studied Permian lingulides and were created by worms that could fully retract to their borings. Thus, in calcareous brachiopods borings were usually created by endobiotic organisms, while in studied lingulides they were made by epibiotic organism. The variable diameter of pits in lingulides indicates that they were made by several generations of encrusters and were not a result of a single colonization event by multiple larvae of boring epibionts. Alternatively, different diameters of borings could result from growing organism which over time produce traces of larger diameter. The biological affinities of organisms responsible for creating the shallow pits in lingulides will remain unresolved without soft body preservation, but the architecture of these pits would fit with sponges as the makers. It is difficult to estimate bioerosion intensities in phosphatic lingulides, because of lack of other comparable records, but most of specimens with small pits show moderate bioerosion intensities.

Predation

Most likely large circular Oichnus simplex boring was made by some shell drilling predator, probably by a predatory gastropod, though the boring is slightly too small for common gastropod borings (Zonneveld & Gingras, 2014). The central location of the boring in a lingulide valve is well consistent with the predatory nature of the boring. The boring intensity was relatively low in the studied lingulide community as only one shell with O. simplex was found among 58 lingulide specimens. There are no reports on the boring frequency in Permian lingulides (Huntley & Kowalewski, 2007), but some data on boring frequencies on articulate brachiopods are available for the Carboniferous (see Harper, 2003, and Klompmaker et al., 2019, for review) and Permian (Hoffmeister et al., 2004). Carboniferous articulate brachiopods show boring frequencies ranging from 3.3 to 17.1 % (Harper, 2003), which is much higher than boring frequency of studied Permian lingulides. Therefore, it is very likely that lingulides were not among favorite prey groups in the early Permian of Brazil. In contrast, some Cambrian lingulides exhibit fairly similar boring intensities to the studied lingulides, being <2% of the population (Harper, 2003). The relatively low boring frequency of studied lingulides may reflect also their infaunal life mode, and their moderate bioerosion intensities could result from post mortem bioerosion.

Cryptic organisms

The earliest examples of Arachnostega known to date are of Cambrian age (e.g., Fatka et al., 2011), but the ichnogenus did not achieve global distribution and become common until the Ordovician (e.g., Aceñolaza et al., 2003).This is the first record of Arachnostega from the interior of lingulide shells. Dead lingulide shells offered shelter for cryptic organisms that left behind Arachnostega traces. Arachnostega traces were probably made by small polychaetes that inhabited soft mud filling of the dead lingulide shell. It is interesting to note that even very small lingulide shells were suitable substrates cryptic organisms such as Arachnostega trace makers. However, the Arachnostega trace makers in Permian of Brazil must have been much smaller than their representatives in the Ordovician of Baltica (Vinn et al., 2014) or Miocene of Egypt (Vinn et al., 2018).

Acknowledgments

Financial support to OV was provided by the Estonian Research Council, project IUT20-34. CZ thanks Fapesp for financial support, project 2017/10956-5. The authors are grateful to the reviewers for their constructive comments on the manuscript and to Mark A. Wilson for linguistic corrections.