A star-shaped trace fossil here assigned to Asteriacites quinquefolius (Quenstedt) was found in the Miocene Shirahama Formation, Wakayama Prefecture. This is the first report in Japan and stratigraphically youngest record for the ichnospecies. The fossil has five distinct arms and wide striations on both lateral sides of each arm, and the shape is bilaterally symmetrical. To clarify the fossil producing process, we conducted burial experiments of extant asteroids in aquaria and in situ for the first time. Asteroids buried themselves in the substratum using the tube-feet, and when asteroids were covered with thin sand, they escaped slantingly upward onto the sand tilting their body in a bilaterally symmetrical posture. As a result, the remaining trace was very similar to the fossil of A. quinquefolius from the Shirahama Formation. Accordingly, the results suggest the present fossil was formed by the behavior of escaping from thin sand cover by asteroids.
Introduction
Asteriacites quinquefolius (Quenstedt, 1876) is a well known star-shaped trace fossil, with records from the Carboniferous to Triassic of India, Europe and the USA (Seilacher, 1953; Hakes, 1977; Twitchett and Wignall, 1996; Röper and Rothgaenger, 1998; Patel et al., 2008; Ishida et al., 2013, 2017). The ichnospecies has wide arms with dense striations from an undifferentiated center, and was interpreted as a resting trace of asteroids on the basis of its morphology (Seilacher, 1953; Patel et al., 2008). Ishida et al. (2013, 2017) found an aberrant form of A. quinquefolius from the Jurassic Hienheim Formation in southern Germany. The specimen has one indistinct arm without striations and four distinct arms with wide and deep striations. Based on observations of the behavior of the extant asteroids, they demonstrated that this form of trace was produced when the asteroid moved horizontally on the trace (Ishida et al., 2013, 2017).
Some time ago, an example of a star-shaped trace fossil of Asteriacites quinquefolius was found in the Miocene Shirahama Formation in Japan (Tanabe Research Group, 1985a, b). The trace has the typical form of A. quinquefolius, and has five distinct arms with only wide striations. The form is different from that of the Hienheim specimen (Ishida et al., 2013, 2017) which suggests the trace was produced by a different process. In this study, to clarify the producing process, we compared the detailed morphology of the new fossil from Japan and that of the trace produced by an extant asteroid. The fossil has no indistinct arm suggesting the asteroid moved vertically upward after producing the trace. We obtained the traces by laboratory observation of the behavior of an extant asteroid after burial of the asteroid by sand. There are the first laboratory and in situ observations of the behavior of extant asteroids after burial by sand.
Material and methods
A star-shaped trace fossil exposed on the surface of a boulder (10 × 20 × 10 cm) was found on the coast in Tonda, Shirahama-cho, Wakayama Prefecture, about 500 m south of the mouth of the Takase River (Site A in Figure 1; 33.62°N, 135.39°E). The boulder was composed of massive sandstone partly with mudstone on the surface. The boulder was interpreted to be autochthonous, because of the similar lithofacies of the outcrop in the immediate vicinity. In this area, the Shirahama Formation of the Tanabe Group is exposed. The lithofacies is composed mainly of massive sandstone and alternations of mudstone and sandstone with hummocky cross-lamination or planar cross-lamination, interpreted as storm or tidal current sediments (Tanabe Research Group, 1985a, b, 1992). The Tanabe Group is of middle Miocene age on the basis of the planktonic foraminifera such as Globigerinoides subquadratus, G. sicanus, Praeorbulina glomerosa, Globorotalia birnageae, and others (Tanabe Research Group, 1984).
The star-shaped trace fossil was preserved as a convex hyporelief and was composed of fine sandstone. The size of the trace fossil was measured by a vernier caliper. The detailed morphology of the present fossil was analyzed by taking molds using a synthetic resin and tracing it onto transparent paper placed upon the mold. Two synthetic resin molds of the trace fossil from the Shirahama Formation were made and housed at the Wakayama Prefectural Museum of Natural History (WMNH-Ge-2120160001) and National Museum of Nature and Science, Tsukuba (NMNS PA18450). The actual fossil belongs to Manabi Manabe, one of the authors.
Specimens of an extant asteroid, Astropecten scoparius Müller and Troschel, 1842, were collected by hand on the coarse sand bottom in the intertidal zone of Moroiso Bay, Miura-shi, Kanagawa Prefecture, Japan from April in 2016 to June 2017 (Site B in Figure 1). The asteroids were kept alive for one to 62 days in the laboratories (National Museum of Nature and Science, Ibaraki Prefecture and Misaki Marine Biological Station, the University of Tokyo, Kanagawa Prefecture) before observation. Twenty specimens were used in a total of twenty-three laboratory observations (Table 1). Their major radius (R) and body thickness at the disc center were 28.5–55.5 mm and 5.2–9.0 mm, respectively, measured by a vernier caliper.
The behavior was observed in aquaria of small (170 mm in diameter, 70 mm in height) and large (230 mm in diameter, 75 mm in height) hemisphere-shaped plastic bowls or exceptionally of a square-shaped polystyrene foam box (250 × 160 × 140 mm) for thickest cover (100 mm) by sand. Aluminiferous abrasive (#2000 or #1500; equivalent to mud in grain size) or gypsum (equivalent to mud) was laid as a substratum about 15–20 mm in thickness and on-site or artificial sea water was filled in the aquaria (Figure 2.1, Table 1). Water temperature was kept at 14–22°C throughout the observation. One specimen of Astropecten scoparius was placed on the substratum in each aquarium. After the asteroids buried themselves shallowly in the substratum (Figure 2.2), they were quickly covered with sand (0.2–0.5 mm in grain size equivalent to fine to medium sand) in thickness of 5–100 mm (Figure 2.3). Asteroids were removed after they successfully escaped onto the sand cover (Figure 2.4). Asteroids did not appear on the sand cover in two observations (Table 1, Observation nos. 22, 23), and then were excavated by removing the sand cover. The substratum was dried for several days to a few weeks to allow the abrasive or gypsum to harden. The upper parts of the plastic bowls and polystyrene foam box were cut off above the sand by a heat iron thread (polystyrene foam cutter). Then, the cover sand was carefully removed with a spoon and paint brush (Figure 2.5). The morphology of the trace left on the hardened substratum was observed (Figure 2.6). During the process of escape, the movements of arms and tube-feet of asteroids were recorded by a video camera.
In situ observations by scuba diving were done for two separate specimens of Astropecten scoparius (about 40 mm in R) on the coarse sand bottom at a depth of 2 m in Moroiso Bay in November 2016 (Site B in Figure 1). In the experiments, when the asteroids slightly buried themselves in the substratum, sand on the sea floor in the immediate vicinity was scooped and dumped quickly on the two specimens about 20 mm thick by hand. Asteroid behavior was recorded using an underwater video camera.
Table 1.
Laboratory observation of Astropecten scoparius covered with sand in aquaria. The same individual was repeatedly used in observations 18, 21, 22 and 19, 20, respectively, shown by asterisks. Abbreviations: abr, aluminiferous abrasive; as, artificial sea water; ns, natural sea water.
The morphology of the traces left on the substratum was sketched on a transparent paper placed upon a photo of the traces.
Results
Morphology of Asteriacites quinquefolius from Japan
The star-shaped trace fossil has five clear radiating arms and an undifferentiated center (Figure 3). The arms are wide and deep rod-shaped with rounded or slightly pointed tips. The shape does not reveal complete radial symmetry but rather having five arms unevenly stretched in bilateral symmetry. The arms have striations on both lateral sides of each arm. The striations are parallel and perpendicular to the arm axis, and wide and in contact with each other. The length (from the estimated center of the trace to the tip of the arm) of five arms measures 63.8–94.2 mm (mean 75.3 mm), and the width of the arm base is 22.8–28.8 mm (mean 25.9 mm). The depth of the central depression of the specimen is 14.5 mm. The specimen had five triangular mounds on both sides of the arms (Figure 3). The morphology completely matches the diagnosis of Asteriacites quinquefolius (Seilacher, 1953), and agrees well with previous descriptions (Hakes, 1977; Patel et al., 2008). Ishida et al. (2013) reported an anomalous form of the ichnospecies from Germany: the heredescribed Japanese specimen differs in having five distinct arms and only wide striations on each arm.
Laboratory observations of the behavior of extant asteroids
In all laboratory observations, the asteroids buried themselves in the substratum shallowly at depths of about 20–90% of body thickness, and formed crescent-shaped mounds along the arms (Figure 4). The mounds finally remained in the trace on the substratum (Figure 5).
When asteroids were covered with sand, they showed three different behavior patterns depending on the thickness of the sand cover (Table 1). When asteroids were covered with thin sand equaling 11–74% of the major radius (R) in thickness (Table 1, Observation nos. 1–11), they slantingly crawled up tilting their body by cyclic movement of tube-feet onto the sand cover in 0.7–19 minutes (Figure 4A, B; “slanting” in Table 1). The escape posture was performed in bilateral symmetry: one arm preceded the body, two arms extended laterally on each side, and the other two arms extended obliquely backward (Figure 4A, B). All asteroids (n=11) left star-shaped traces on the substratum. They had five distinct arms with wide striations. Eight traces were not radially but bilaterally symmetrical similar to the posture of asteroids when escaping as described above (Figure 5B, D–J; Table 1, Observation nos. 2, 4–10). The other three traces, on the other hand, were radially symmetrical (Figure 5A, C, K; Table 1, Observation nos. 1, 3, 11). The striations were in contact with each other, arranged in parallel, perpendicular to the arm axis on both radial halves (Figure 5A–K).
On the other hand, when asteroids were covered with relatively thick sand equaling 75–136% of R in thickness (Table 1, Observation nos. 12–21), they escaped onto the sand cover in 1–27 minutes by a different behavior (Figure 4C–E; “vertical” in Table 1). The asteroids lifted one or two arms almost vertically on the sand cover bending themselves aborally (Figure 4C2, D2, E2). They then heaved most of the body above the sand (Figure 4C3) and sometimes were nearly upside down (Figure 4D3, E3). Finally they returned to their original postures with the aboral side up (Figure 4C4, D4, E4). The traces left on the substratum had five irregular arm depressions with wide striations on both radial halves. One or two arms of the traces corresponding to the arms lifted were shorter than the others (Figure 5L–U). The traces had various shapes of grooves, which could be sharply curved (Figure 5M, R), slightly curved (Figure 5N, U), or long and straight (Figure 5L, T).
When asteroids were covered with an even thicker layer of sand equaling 136% and 266% of R in thickness, they did not appear on the sand after 74 and 281 minutes, respectively (Table 1, Observation nos. 22, 23). When the sand was removed, the asteroids remained upside down in the sand (Figure 4F2, G2; “inverted” in Table 1). In both observations, the traces left on the substratum had five irregular indistinct arms with wide striations. The traces had grooves similar to those in relatively thick sand cover (Figure 5V, W).
In situ observations of Astropecten scoparius
When asteroids were covered with coarse sand of about 50% of major radius (R) in thickness, they escaped onto the sand slightly tilting their bodies in about 20 seconds (Figure 6). Their behavior is similar to that of asteroids covered with thin sand in laboratory observations (Figure 6).
Discussion
The ichnospecies Asteriacites quinquefolius from the Shirahama Formation is described for the first time from Japan in this study although another ichnospecies A. lumbricalis of the star-shaped trace fossil ichnogenus Asteriacites has been reported from the Triassic Hiraiso Formation, Miyagi Prefecture (Ishida et al., 2013, 2017) and from the Cretaceous Himenoura Group, Kumamoto Prefecture (Ishida et al., 2017). Asteriacites quinquefolius has been previously reported from older strata: the upper Carboniferous of the U.S.A. (Hakes, 1977), the Middle Jurassic of Germany (Seilacher, 1953), the Upper Jurassic of Germany (Röper and Rothgaenger, 1998; Ishida et al., 2013), the Lower Cretaceous of India (Patel et al., 2008) and the Lower Triassic of Italy (Twitchett and Wignall, 1996). This fossil from the Miocene Shirahama Formation is the youngest record. The detailed morphology of the fossil from the Shirahama Formation is similar to that reported in the previous papers on A. quinquefolius from the Middle Jurassic of Germany (Seilacher, 1953, Table 10, fig. 1a), and the Lower Cretaceous of India (Patel et al., 2008, Fig. 1c–e, g).
In this study, burial observations using an extant asteroid in aquaria and in situ (Figures 4, 6) revealed the producing process of the present fossil. This paper mentioned for the first time that Asteriacites quinquefolius is produced by asteroids' escaping behavior after being covered by sand. The ophiuroid-related trace fossil A. lumbricalis is also inferred to have been made by escaping behavior from sand cover on the basis of the mode of occurrence of fossils (Seilacher, 1953, 2007). Escape movements of extant ophiuroids from smothering sediments in aquaria and in situ were reported by Ishida (1999), Ishida and Fujita (2001) and Ishida et al. (2004).
In aquarium observations, if the thickness of sand cover does not exceed 136% of the major radius (R), the asteroids successfully escaped from the sand cover (Table 1). The asteroids moved obliquely upwards when the sand cover was thin (Figure 4A, B), while they moved mostly vertical (Figure 4C–E) when they were covered with a thick layer of sand. However, when the thickness exceeded 136% of R, they failed to escape from the sand cover and were found in the sand in an upside-down posture (Figure 4F, G). The behavior on the sea floor in in situ observations (Figure 6) was similar to that from thin cover sand in the laboratory observations (Figure 4A, B). The results of these observations suggest that escape behavior from the sand cover produces star-shaped traces. The star-shaped trace of the extant asteroids has five distinct arms with wide striations on both lateral sides when it escapes from the sand cover. The morphology of the traces, however, varied slightly depending on the thickness of sand. In thick sand, the arms are not uniform in length, the central disc has various irregularly shaped grooves and the shape was not bilateral (Figure 5L–W). Asteriacites quinquefolius from the Shirahama Formation had distinct five arms, bilaterally symmetrical body and wide striations on both sides of each arm (Figure 3). Their morphologies are strikingly similar to the traces when the asteroid is covered with thin sand (Figure 5B, D–J). Our results suggest that the Shirahama specimen represents the trace produced by an asteroid that escaped after being covered by thin sand (Figure 7). The sand cover may have been caused by a storm or tidal current event (Tanabe Research Group, 1992).
Acknowledgments
We would like to thank Masataka Okimoto of the Kiiminpo Press for providing information on the present fossil. Many thanks go to Mamoru Sekifuji, Akito Ogawa, Itaru Kobayashi of the University of Tokyo and Hiroshi Namikawa of the National Museum of Nature and Science for their help in collecting extant asteroid specimens. Thanks are due to Yasunari Shigeta of the National Museum of Nature and Science for his kind help in taking photos. Thanks are extended to Takumu Kato for his help for analysis of the movies. John W. M. Jagt of the Natuurhistorisch Museum, Maastricht, the Netherlands and Ben Thuy of the National Museum of Natural History, Luxembourg greatly improved an earlier draft of this manuscript.
Author contributions
We declare that none of the material in this manuscript has been published or is under consideration for publication elsewhere. Y. I. did experiments in laboratory, and analyzed the fossil. T. F. analyzed the experimental data. H. K. did the experiments, mainly in situ observation. M. M. collected the fossil and surveyed the site. M. O. identified the fossil and surveyed the fossil site. All authors contributed to the writing of the manuscript.