New Ediacaran fossils from the Ukraine, some with a putative tunicate relationship

Sack-like body fossils Finkoella ukrainica gen. et sp. nov. and F. oblonga sp. nov., and reticulate fossil Pharyngomorpha reticulata gen. et sp. nov. are described from the upper Ediacaran shallow-marine deposits of Ukraine, which are no younger than 557 Ma. The first two resemble the flattened bodies of tunicates showing mainly the outline of tunica, while the third is considered as a fragment of the pharyngeal basket of a tunicate. F. ukrainica is represented by smaller individuals interpreted as juveniles, which may occur in clusters together with less numerous larger individuals. The larger forms are interpreted as adults, some of which show the preserved oral/atrial syphons and possible traces of internal organs bulging through the tunica. Moreover, Burykhia sp. from the uppermost Ediacaran of the same region is presented. This is the second and younger occurrence of the genus Burykhia, which is preserved as a possible fragment of the pharyngeal basket. All the fossils are preserved as the “death masks” between microbial mats, and their appearance depends partly on the relation to the parting surface on which they are observed. The presented new taxa along with the literature data reinforce the possibility that tunicates originated already in late Ediacaran.


Introduction
In the recent marine environments, the Ascidiacean tunicates (Tunicata: Ascidiacea) are world-widespread and diverse (2815 species) filter-feeding, mostly sessile, colonial or solitary invertebrates having sack-like body and two syphons. They are important link in the phylogenetic tree because they are regarded as the ancestors of chordates (Lambert 2005;Shenkar and Swalla 2011;Holland 2016). Their unquestionable fossils are dated back to the early Cambrian, as exemplified by Shankouclava shankouense Chen et al., 2003 from China. However, three taxa considered as ascidiaceans are known from the upper Ediacaran, i.e. Ausia fenestrata Hahn and Pflug, 1985 from the Kuibis Quartzite (549-543 Ma) of the Nama Group in Namibia, poorly preserved Yarnemia acidiformis Nessov in Chistyakov et al., 1984 from the Ediacaran in the middle reaches of the Onega river (lower part of the Erginskaya Formation (550-552 Ma; see Grazhdankin 2011;Grazhdankin and Maslov 2015), and Burykhia hunti Fedonkin et al., 2012 from the Ediacaran beds (Verkhovka Formation, 555 Ma), both in the White Sea coast region in the north-western Russia. Ausia is known from one specimen, and Burykhia from two specimens.
In this paper, four new fossil taxa which could be ascidiacean tunicates are presented from the Ediacaran of Podolia in Ukraine (Fig. 1). One of them is represented by several specimens. These fossils add a new information to the discussion on the still poorly recognized early evolution of the class Ascidiacea. The rocks bearing them are no younger than 557 Ma. Therefore, the fossils are potentially the possible oldest representatives of tunicates if their systematic position is so. The Ediacaran of Podolia provided several taxa of macrofossils belonging to diverse taxonomic groups (e.g., Palij 1976;Fedonkin 1985;Velikanov 1985;Fedonkin Handling Editor: Joachim Reitner. and Vickers-Rich 2007;Martyshyn 2012;Gozhik 2013;Ivantsov et al. 2015;Martyshyn 2015, 2017;Nesterovsky et al. 2018;Soldatenko et al. 2019). Moreover, some ichnotaxa are described from these deposits (e.g., Palij 1974Palij , 1976Gureev 1983Gureev , 1984Gureev , 1986Ivantsov et al. 2015;Grytsenko 2020;Uchman and Martyshyn 2020).

Geological setting
The study area stretches on the southwest flank of the Ukrainian Shield, which is called the Volyn-Podillya-Moldavia Basin. During the Ediacaran Period, the basin a b c d  (Fig. 2). Fossils from the Mogilev Formation were collected from the Novodnistrovs'k Quarry (Fig. 1c), which was monitored since 20 years by the senior author (Martyshyn 2012). At the bottom of the quarry, Lower Proterozoic granites and migmatites of the Ukrainian Shield are exposed. They are blanked by thin (up to a metre), discontinuous layer of basal conglomerates, which underlie the Lomozov (Lomoziv) Member of the Mogilev Formation. This member is dominated by mudstones and siltstones, which represent the second sedimentary cycle in the Podolia Ediacaran (Korenchuk and Ishchenko 1981). Some specimens of Finkoella oblonga, Pharyngomorpha reticulata and F. ukrainica derive from this unit. The Lomozov Member contains rare dickinsoniids (Dzik and Martyshyn 2015) and relatively common disc-like representatives of the Petalonamae group (Dzik and Martyshyn 2017). A higher position is occupied by the Yampol (Yampil) Member, which is composed of thick and very thick, partly cross-bedded sandstones (up to 9 m thick) and a package of thin sandstone and siltstone beds (up to 3 m thick) at the top. The holotype and paratype of Finkoella ukrainica presented in this paper derive from the top part of the Yampol Member. The overlying Lyadova (Lyadov) Member crowns the Mogilev Formation. It consists up to 2 m-thick siltstone at the base followed by up to 14 m-thick mudstone with intercalations of thin siltstone or sandstone beds.
The base of the overlying Yaryshev (Yarishiv) Formation was previously dated to 553 Ma by Grazhdankin et al. (2011) and Grazhdankin (2014) but newer dating of bentonite layers from the Yampol Member top package (layer B1, c. 1 m above the thick-bedded sandstones) points to 556.78 ± 0.18 Ma (Soldatenko et al. 2019). The presented fossils from the upper part of the Yampol Member are below the bentonite. Therefore, all the presented fossils are no younger than roughly 557 Ma.
Burykhia was found in a small abandoned quarry in the Ternava river gorge near Kitaygorod in the Khmelnytsky region (Fig. 1d), where the Komarovo Member form the uppermost part of the Studenitsa Formation (60 m thick) which is located in the uppermost part top of the Kanilovka Group. The Komarovo Member are composed of rhythmic intercalations of laminated mudstones and siltstones, which are intercalated with occasional sandstone beds. This formation transits up to the lower part of the Okunets Formation of the Baltic Group, which is already considered as the lower Cambrian (Cambrian Series 1) containing the trace fossil Treptichnus pedum (Seilacher, 1955) (see Ivantsov et al. 2015). Beside this and other trace fossils, the Studenitsa Formation contains body fossils of algae, including Tyrasotaenia podolica Gnilovskaya, 1971, Vendotaenia antiqua Gnilovskaja, 1971(Ivantsov et al. 2015, Tymkivia stuzhuki Martyshyn in Nestrovsky et al., 2018, the protozoan or alga Harlaniella vermiformis Martyshyn in Nesterovsky et al., 2018, and the protozoan Palaeopascichnus delicatus Palij, 1976(Nesterovsky et al. 2018).

Materials
The presented specimens are collected by (1) A. Martyshyn in collaboration with a research group during 20 years of the field work and monitoring of the quarry near the Dniester hydroelectric dam (DPHS) and other outcrops ( Fig. 1), and by (2) S. Finko who provided the holotype of F. ukrainica and the large sandstone slab with the paratype of F. ukrainica material for research. Specimens of the described taxa are in the collection of the Institute of Geology of the Taras Shevchenko National University of Kyiv. They have the institutional abbreviation IGKNU.

Systematic description
Subphylum Tunicata Lamarck, 1816 Class Ascidiacea de Blainville, 1824 Remarks. Classification of the fossil Ascidiacea is based on (1) their pouch-like structure, which is similar to recent representatives of this class, (2) the presence of traces of siphons, and (3) imprints of internal organs on the surface of semi-decomposed remains, which are similar to the tunic, pharyngeal basket mantle (branchial sac), atrium, stomach, and the atrial/oral siphons of the present-day tunicates. These morphological features are typical of the family Ascidiidae Herdman, 1882 (Herdman 1882;Fedonkin et al. 2012) and cannot be misled for other group of organisms.

Genus Finkoella gen. nov. Martyshyn
Etymology. After S.V. Finko, an amateur palaeontologist, who provided the holotype and paratype of the type species.

Diagnosis.
A convex or concave, ovate, oval, or rarely oblong, cucumber-shaped, lenticular structure on the lower or upper surface of sandstone/siltstone/mudstone layers. In most specimens, one end is more elongated, and the opposite one is more rounded or almost even. One of the ends may show a protruded, short neck and/or a shallow sinus. A low/ shallow circular protrusion/depression can be present in the convex or concave structure, respectively. The protrusion/ depression is usually located eccentrically towards one of the ends. Moreover, some shallow/low concavities/convexities may occur on the surface. The edges are usually slightly elevated in convex specimens or depressed in concave forms.
Species. Finkoella ukrainica and F. oblonga described in this paper. Etymology. After the country Ukraine.
Diagnosis. Finkoella showing oval to elliptical, rarely oblong shape, which may show a protruded, short neck and/or a shallow sinus. A low/shallow circular protrusion/ depression can be present in the convex or concave structure, respectively. Smaller specimens can be without neck and protrusions/depressions. Description. The holotype (Fig. 3a-c) is a convex oval structure on the lower surface of a thin bed of micaceous, clayey, very fine-grained sandstone. It is 95 mm long and up to 55 mm wide. A round protrusion, 5 mm in diameter, is located slightly eccentrically towards one of the ends. One of the ends is more elongated, and the opposite one is more rounded. The more rounded end shows a protruded short neck, which is wider at its termination. The termination is unevenly lobate. The margins are sharp and even. The surface of the structure is generally smooth, except for local grain-size corrugations, thin and short grooves and a longer, narrow, shallow groove which runs obliquely through three quarters of the width close to the wider end. The second fossil is located on the same side of the slab bearing the holotype (Fig. 3a, b). This is a concave, smooth structure in the shape of a simple cucumber. One end of the structure shows a sinus, which is 4 mm deep and 7 mm wide. The opposite end is broken on the edge of the slab and therefore not available. The preserved margins are sharp and locally elevated.
The paratype (Figs. 3d, 4a, b) is a concave, elongate oval structure on the upper surface of a sandstone slab. It is 56 mm long and up to 25 mm wide. The wider end is tapered and the opposite, narrower end shows a short, indistinct neck. The surface is covered with local bulges of oval outline and irregular, gentle depressions. The structure is mostly sharply bounded along its margin, but locally the boundary is less distinct. From one side, the margin is elevated and locally marked by a double ridge, which is 2 mm wide (Fig. 3d). The ridge is separated by an uneven narrow groove.
Other specimens (Figs. 4,5,6) are usually oval, convex or concave structures; some are obtuse from one end. They show variable sizes, with length ranging from 5 to 110 mm and width from 5 to 55 mm. The width versus length of all specimens show strong positive correlation (r = 0.84; Fig. 8).
Most of the smaller specimens are usually smooth, without any neck and protrusions, but some of them show slightly elevated/depressed edges and or bulges interpreted as atrial siphon. Only some of them show a short neck (Fig. 7c) interpreted as oral siphon and knob-like protrusions/depressions. One of them shows a narrow side neck, which can be atrial syphon (Fig. 7i).
The described structures may occur in clusters (Figs. 3a,  b, 4a, b). Individuals in the cluster may show different size.
Longer axis of the fossil shows distinct orientation (rosette in Fig. 4b).
Remarks. The small specimens, which are less than 25 mm wide (see Fig. 8), are considered as representing juvenile organisms (see "Discussion"). The holotype was selected from specimens representing large, fully developed and most completely preserved organism (Fig. 9).

Diagnosis. Elongated
Finkoella showing winding margins and longitudinal furrows/ridges, which tend to run parallel to the margin.
Holotype. Split slab of horizontally to wavy laminated siltstone showing positive, i.e. convex fossil (IGKNU17p202/1) and is counterpart showing negative, i.e. concave part (IGKNU17p202/2). This is the only specimens of the species. It derives from the Lomozov Member of the Mogilev Formation, Mogilev-Podolsky Group, in the quarry near the DHPS at Bernashivka village, Vinnytsia region.
Description. The holotype (Fig. 10a) is an elongate, cucumber-like structure, 120 mm long and up to 42 mm wide, which is visible as a positive (convex) part and its negative (concave) counterpart in a horizontally split slab. It shows six longitudinal, low groves and ridges, which show gentle slopes and are 3-4 mm wide. They run parallel to the margin of the structure and turn accordingly at the terminations, so they tend to form concentring rings. They are more continuous on one side than on the other side, where only their fragments are visible. Moreover, the structure shows small corrugations, but without any pattern. Edges of the structure are distinctly elevated/depressed and locally show a double ridge or groove, which is 1.5-3 mm wide. The specimen KSU17p215 and KSU17p215/1 (Fig. 10b) shows the same shape as the holotype. It is 95 mm long, up to 40 mm wide. Six, more or less complete and an additional one incomplete ridges/grooves are visible. Their course is the same as in the holotype. The edges are sharp and locally elevated. One of the termination is narrower. It forms a dome-like protrusion. The other end is obtuse and seems to partly broken.
The specimen IGKNU17p216 (Fig. 10c, d) is ovate, 38 mm long, up to 21 mm wide. It becomes gradually narrower toward its semi-circular distal. The opposite end is obtuse and uneven. The perimeter is located in one third of the length from the obtuse end. Six low ridges and grooves run parallel to the lateral margins of the structure. They disappear toward the ends. A curved tail of small corrugations emerges from the obtuse end (rm in Fig. 10c). Its margins are diffused. Similar corrugations cover larger part of the fossil body.
Remarks. In the holotype, both terminations of the structure touch the edge of the slab, so a possible neck-like protrusion is not visible, if it was present at all. The neck is also not obvious in KSU17p215. The specimen IGKN17p216 is determined as Finkoella cf. oblonga because its ovate shape differs from the elliptical, cucumber-like shape of the holotype. It is also distinctly smaller than the holotype. Etymology. After the pharyngeal basked, the characteristic anatomic part of tunicates, which is probably preserved in the type species of this genus, and after morpha-in relation to morphology the pharyngeal basked.

Diagnosis. Basket-like structures showing ribs and/or grooves arranged in a rectangular net.
Species. Only Pharyngomorpha recticulata-monotypic genus. Etymology. After reticulum (Latin)-net, in correspondence to the morphology.

Diagnosis. Elongated
Finkoella showing winding margins and longitudinal furrows/ridges, which tend to run parallel to the margins.
Description. The holotype is 37 mm long and up to 19 mm wide. It contains about ten, straight, perpendicular, densely packed parallel ribs separated by narrow furrows. There is 4 ribs per 10 mm. The furrows are less than 1 mm wide. The ribs are crossed by five longitudinal furrows (running along the structure) which are perpendicular to the ribs. The longitudinal furrows are c. 1 mm wide. The ribs in the place of intersection with the longitudinal furrows are still visible but they are much lower than in the inter-ribs stretches. The inter-ribs stretches form rectangles (5.0 × 2.5 mm), which are perpendicularly oriented to the structure, so they show the net-like pattern. The structure is elevated in the marginal part. In the narrower part, short bars occur on the margins. The smaller specimen IGKNU17p203/1 is located c. 20 mm from the holotype. It is 14 mm long and 12 mm wide and composed of five ribs separated by narrow furrows, which are similar to these in the holotype.
Remarks. It is not excluded that the smaller specimen is a disarticulated part of the holotype. In the vicinity of the holotype, poorly outlined ribs can be visible, which may also derive from fragmentation of the organisms represented by the holotype or other individuals. Both specimens co-occur with Finkoella ukrainica on the same surfaces.
Description. An incomplete (broken) convex structure having its concave counterpart and an outline of incomplete ellipse, up to 45 mm wide and at least 42 mm long (probably twice longer when would be complete). It shows diagonal, straight, narrow furrows, which are about 0.5 mm wide and 5-8 mm apart. The furrows show slit-like widenings, up to 4 mm long and up to 2 mm wide, 4-5 m apart, which are deeper than the remaining parts of the furrow. In some inter-furrow areas, small, oval depressions, 1-2 mm wide may occur. They have a concave counterpart. They are located in the middle between the furrows, in a line parallel to them. Fedonkin et al., 2012 (Fig. 11c, d), the only known species of Burykhia is also known from fragments of the pharyngeal basket. However, the pattern of the furrows and depressions and their counterpart is different: it does not show the slit-like depressions along the furrows, and the structures in Burykhia sp. are arranged diagonally, while in B. hunti they are transverse and longitudinal.

Discussion
The morphological features of the described fossils point to animals, which are similar to these of the class Ascidiacea (tunicates). The overall shape of Finkoella resembles sacklike soft body. The neck or sinus at the end of F. ukrainica can be interpreted as the oral (branchial, inhalant, incurrent) syphon, while the more regular, circular bulges can be the atrial (exhalent, excurrent) syphon. The elevated edge and the double marginal ridge/groove which are present in some specimens are an evidence of a compressed, elastic sack, which can be interpreted as the tunica. Other irregularities (bulges) on the surface can be manifestations of some internal anatomic elements (heart, stomach, gonads) of the tunicate bulging through the tunic (Fig. 13). Preservation of the syphons may depend on the position of the body during burial. If the syphons are underside the body, they can be visible only as bulges on the opposite of the body and cannot be clearly recognized. The smaller specimens of Finkoella ukrainica probably represent juvenile individuals (less than 25 mm wide by convention; Fig. 8). They prevail in number. Their occurrence in clusters in association with larger, well developed individuals suggest colonial style of life. Such style of life may result from the asexual reproduction by budding known from some recent ascidians (Holland 2016). This explains the occurrence of numerous juveniles close to small number of adults. However, accumulation by weak currents is not excluded. Distinct orientation of F. ukrainica (Fig. 4b) can result from such currents, which are related to tides, as the intertidal environment is proposed for the rocks bearing the fossils (Uchman and Martyshyn 2020).
Most of the modern tunicates have the standing life position, with their bag-shaped body attached to various hard substrates. However, some tunicates are passively lying on the bottom (Bonnet et al. 2013). It is possible that Finkoella was a tunicate of such type, which is adapted to soft substrates (Fig. 9). This is inferred from the fact that there are no sediment elements that would be hard and no evidence a part of the body which would be attached to such elements or would be anchored in the sediment. Similar mode of life is interpreted to the Ediacaran Yarnemia, which is also tentatively ascribed to tunicates (Chistyakov et al. 1984).
The net-like structure in Pharyngomorpha strongly suggests that it is an imprint or mold of the pharyngeal basket, which is partly similar to the other ascidian Burykhia hunti Fedonkin et al., 2012 from the late Ediacaran deposits (555 Ma) on the White Sea coast region (Fedonkin et al. 2012). The pharyngeal basket is a characteristic and unique anatomical element of tunicates, where the food particles are trapped in mucus and supplied into the stomach. It shows a sac-like morphology, which may occupy a large part if the body. Its wall is perforated with gill slits and reinforced with ribs. Morphological detail of the basket (arrangements and shape of the gill slits, ribs) may differ among taxa and develop through ontogenetic stages (Lambert 2005;Shenkar and Swalla 2011;Holland 2016). The pharyngeal basket is present in Shankouclava, the oldest reliable tunicate (Chen et al. 2003). Nevertheless, the reticulate anatomical elements, which may resemble the pharyngeal basket can occur in some Ediacaran fossils which ascidian affinity remains controversial, e.g. in Ausia (Hahn and Pflug 1985). Burykhia belongs also to them. The arrangement of slits and ribs in Burykhia points to a a b c Fig. 13 Preservational mode of preservation of Finkoella. a Formation of "death masks" in between microbial mats with indication of splitting surface. b Appearance on splitting surfaces. c Some morpho-logical features of Finkoella. The bulge can be formed over an internal organ of the tunicate more complex morphology than simple reticulate from in Pharyngomorpha.
The described Burykhia isp. represents the second occurrence of this genus, which extends its stratigraphic range by about 15 Myr to the latest Ediacaran. Finkoella preserved mostly external part of the body, i.e. tunica, while Burykhia and Pharyngomorpha are probably preserved fragments of the pharyngeal basket. Therefore, Finkoella can be considered as a more completely preserved body of a tunicate.
The changeability of morphology in Finkoella demonstrates dependence on taphonomic factors. These fossils are preserved as probable "death masks" (Gehling 1999) formed my microbes between microbial mats which enveloped the organism after its death (Fig. 13a). Local ferruginization of the surface of the fossils (Figs. 3d, 4a) can derived from the weathered pyrite, which mineralize the body and is common in this type of preservation. Surfaces of the mats and bedding surfaces are the parting surfaces on which the fossils are observed. The concave or convex appearance depends on whether the parting surface propagated below or above to the fossil (Fig. 13b). Therefore, both, convex or concave structures may occur on the same surface as e.g., in the case of slab bearing the holotype (Fig. 3a, b), with their counterparts on the opposite surface, including external molds if the surface runs on the surface of the fossil and the body of the fossil remains on the counterpart surface. If the surface runs through the fossil, its parts can be preserved on both sides. Such cases show that the broken fossil contains at least two layers (Figs. 6a, b, 7b), accordingly to the sack-like morphology.
The presented new taxa along with these which are described so far from the Ediacaran (Ausia, Yarnemia, Burykhia) reinforces the possibility that tunicates originated already in late Ediacaran. As Ausia is no older than 549 Ma, Yarnemia than 552 Ma and Burykhia than 555 Ma, Finkoella and Pharyngomorpha with their age no younger than 557 Ma can be regarded as the possible oldest representatives of tunicates. In this way, deep roots of vertebrates seem to be more and more probable.
Acknowledgements The authors would like to express their gratitude to S.S. Solodkyi, a devoted participant of field expeditions, who drew the reconstruction of Finkoella ukrainica, and to amateur palaeontologist Sergei Finko who provided the holotype of F. ukrainica and the large sandstone slab with the paratype of F. ukrainica material for research. A.M. thanks his research supervisor Victor Ogar for helpful discussion and he is also grateful to numerous unnamed assistants in the field research, participants of discussions and consultations. Visit of A.U. in the field was partly supported by the Jagiellonian University. Two anonymous reviewers and Mike Reich, Munich, are acknowledged for critical remarks which helped to improve the paper.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.