Evidence for cryptic molting behavior in the trilobite Toxochasmops vormsiensis from the Upper Ordovician Katian Kõrgessaare Formation, Estonia

Documentation of cryptic trilobite behavior has presented important insights into the paleoecology of this fully extinct arthropod group. One such example is the preservation of trilobites inside the remains of larger animals. To date, evidence for trilobites within cephalopods, gastropods, hyoliths, and other trilobites has been presented. Importantly, most of these interactions show trilobite molts, suggesting that trilobites used larger animals for protection during molting. To expand the record of molted trilobites within cephalopods, we present a unique case of a Toxochasmops vormsiensis trilobite within the body chamber of a Gorbyoceras textumaraneum nautiloid from the Upper Ordovician Kõrgessaare Formation of Estonia. By examining this material, we present new insights into the ecology of pterygometopid trilobites, highlighting how these forms used large cephalopods as areas to successfully molt. Supplementary information The online version contains supplementary material available at 10.1007/s00114-024-01906-8.


Table 1
Summary of trilobites within host groups.Ordered temporally and then taxonomically by genus.Note that there has been additional documentation of trilobites within larger animals (see Mikulic 1994;Feist 2001).However, these lacked geological information so could not be included within this

Geological context
Vormsi is a low and flat island west of mainland Estonia that emerged from the Baltic Sea ~ 3000 years ago and rises to a maximum height of 11 m above sea level.Deposits below the thin Quaternary material are historically called the Lyckholm Layer and consist of the Upper Ordovician (middle to upper Katian shallow sea sediments mainly of bioclastic limestones that correlate upwards with the Nabala, Vormsi, and Pirgu regional stages; Fig. 1a; Schmidt 1858; Jaanusson 1944Jaanusson , 1956)).The first detailed fossil collection on Saxby seashore, on the northwestern section of Vormsi, was by Sauramo (1929) who recorded a rich fauna over a 2-km-long section, proximal to the beach.A diverse assemblage of trilobites, brachiopods, cephalopods, gastropods, and corals totaling 46 taxa has been recorded.Furthermore, the fauna from the northern coastal region differs from the southern section.This was confirmed in Jaanusson (1944;1956) who elected the substages based on this lateral faunal change-the older Kõrgessaare (in the northern region) and the younger Nõmmküla (in the southern region).This is due to a slight, 4-7° dip in the beds, which parallels the Paleozoic sedimentological belts observed in Estonia (Fig. 1).Despite this, both zones are considered part of the Kõrgessaare Formation, with the younger coral and calcitic vermiporellid algae-rich beds gradually transitioning into the overlying Moe Formation (see Kröger et al. 2017).The Upper Ordovician Kõrgessaare Formation is a 10-20-m-thick limestone intercalated with marls and clay interlayers, becoming more argillaceous up-section.There is a ~ 1.7 m exposure along Saxby Beach at a cliff, but this is commonly covered by rubble and erosion.In the northern beach region, a discontinuity surface marks the upper Vormsi Stage (Einasto 2012).This is the most accessible area, as there is a road to the beach.This situation may have resulted in selective sampling bias in more recent collections, including our specimen.The fauna in the formation consists of brachiopods, gastropods, cephalopods, bryozoans, tabulate and rugose corals, and trilobites (Schmidt 1858(Schmidt , 1881;;Sauramo 1929;Jaanusson 1944Jaanusson , 1956;;Rõõmusoks 1998Rõõmusoks , 2000)), mostly from the Vormsi Regional Stage.Documented trilobites include illaenids (Parillaenus roemeri (Volborth 1864) and Parillaenus angustifrons (Holm 1886)), proetids (e.g., Ascetopeltis kertelensis (Schmidt 1894), and Cyphaspis sp.), rare calymenids, enrinurids (Erratencrinurus  (Suzuki et al. 2009) and during the Hirnantian trilobites with schizochroal eyes and better vision expanded into the pterygometopid niche (Hints et al. 2012;Ebbestad et al. 2015).The examined T. vormsiensis therefore likely lived relatively nearshore, in a warm, low-latitude epicontinental sea environment, rich in corals and shelly fauna.

Methods and materials
The examined specimen is housed in the Geological collections of the University of Tartu (TUG) and assigned the specimen number TUG 1355-193.The specimen was photographed under LED lighting using an Olympus E-M1 MarkIII camera with a 12-45-mm lens.Images were stacked using OM Capture.Measurements of specimens were gathered using ImageJ (Schneider et al. 2012) and 3D Slicer 4.11 (Fedorov et al. 2012).
The specimen was micro-CT scanned on a GE PHOENIX v|tome|x scanner with a 240 kV X-ray tube at the Microscopy and Imaging Facility at the American Museum of Natural History (AMNH).The scan was run at 200 kV and 230 μA.Scan data were reconstructed using the GE software datos|x 2.1 and segmented in the software 3D Slicer 4.11 using the SlicerMorph toolkit (Rolfe et al. 2021).The reconstruction of the scan was exported as a.PLY file (Supplemental Document 1).

Results
The specimen shows a molted exoskeleton of an individual of the trilobite Toxochasmops vormsiensis within the steinkern of a Gorbyoceras textumaraneum nautiloid (Fig. 2).The G. textumaraneum specimen measures 95.1 mm in length and 36.6 mm in width (across the aperture) tapering to 22.4 mm posteriorly.The trilobite is situated within the body chamber of the nautiloid.The chamber is entirely filled with sediment.
The Toxochasmops vormsiensis is partially preserved.The cephalon is disarticulated, missing the left eye (possibly due to erosion), and is partially covered by matrix on the right side; the cephalon sagittal length is 19.8 mm.The pygidium is disarticulated and partially covered by matrix on the left side; the pygidium sagittal length is 21.7 mm.The pygidial sagittal axis lies at an angle from the cephalic sagittal axis: ~ 10° to the right on a horizontal plane and ~ 60° declining posteriorly on a vertical plane.The cephalon and pygidium are separated by 11.6 mm, with the pygidium resting behind the cephalon in a telescoped configuration.The counterpart of a thoracic segment is located between the cephalon and pygidium.Upon examining the micro-CT scans, five to six articulated thoracic segments are identified in the matrix, under the cephalon.

Molting in shelter
Trilobites required quiet environments to complete ecdysis (Henningsmoen 1975;Brett 1977;Brandt 1993;Zong et al. 2016) and inhabiting the remains of other animals, such as empty cephalopod shells, would have been ideal for this purpose (Chatterton et al. 2003).The trilobite considered here is preserved within a molting configuration, demonstrating that the individual had entered the dead cephalopod conch to molt and excludes the possibility of the carcass having been washed into the nautiloid by passive fluid flow.While cephalopod shells may have been an ideal location for successful molting in paleoenvironments that lacked other safer areas to molt (Ladd 1929;Chatterton et al. 2003), the association of a trilobite within a nautiloid from the Kõrgessaare Formation is unique.This indicates that the environment likely had locations for successful molting, and, in this case, the trilobite was simply fortunate enough to locate a shell to molt within.

Hiding from predation
Large, empty shells would have provided ideal shelter from predation, as well as molting (Brett 2003;Chatterton et al. 2003;Fatka and Kozak 2014).While there is no evidence for predation on the observed specimen, records of failed predation are known from Ordovician trilobites (see Owen 1985;Rudkin 1985;Zong 2021;Bicknell et al. 2022a, b;Fatka et al. 2022;Bicknell and Kimmig 2023).Although this specimen may not have been subject to predation, this interaction may indicate a preference for molting within cavities to avoid possible attacks (Brett 1990(Brett , 2003;;Brett and Walker 2002;Fatka and Budil 2014;Fatka et al. 2021).

Sea floor disturbance
Large shells represent ideal localities to survive periods of disturbances from storm events or rapid sediment inundation (Fatka and Kozak 2014).We are unable to determine if this situation can be excluded.As the paleoenvironment associated with this fossil is considered relatively nearshore, the trilobite may have entered the cephalopod to avoid disturbances and then may subsequently have molted.

Scavenging on food
The cephalopod may have had decaying soft material ideal for benthic scavengers (Fatka and Kozak 2014) and would have attracted trilobites and other smaller animals to the shell (Fatka et al. 2009).There is no evidence for any other feeding activity preserved in the specimen.As such, we can exclude this option here as a primary justification for the trilobite entering the cephalopod.Taken together, the evidence suggests that this interaction represents molting within a sheltered condition.However, this condition would also have allowed the trilobite to be protected from predators and from poor environmental conditions.
The molt configuration observed in this specimen-the pygidium wedged directly behind the cephalon in a telescoped manner-is common in Toxochasmops (McNamara and Rudkin 1984, figs. 1, 3;Rõõmusoks 1998, pl 1, fig. 8; pl 2, fig.13).During exuviation, the cephalothoracic joint disarticulated, similar to Salter's molting mode (Henningsmoen 1975).The cephalon was therefore molted in a way that allowed the trilobite to move anteriorly on an angle and wedge the pygidium against the cephalon, facilitating its removal (McNamara and Rudkin 1984;Budil and Bruthansová, 2005).Previously records of this molt configuration are of specimens in non-sheltered conditions (McNamara and Rudkin 1984).In these examples, few to no thoracic segments are observed, indicating the thorax is shed elsewhere (McNamara and Rudkin 1984).In our specimen, most of the thorax is preserved under the cephalon, as would be expected from molting in shelter.Further, TUG 1355-193 demonstrates that this complex exuviation technique was possible within enclosed spaces.
The limited space within the shell would have presented some complications, such as escaping from the shell after molting.Passive sediment in-filling of an open structure on the sea floor likely resulted in a partly filled shell when the trilobite entered to molt (Hewitt 1988).This would have further limited the space for movement.Given the molt size, the trilobite likely filled up over half of the remaining space and would have had to move farther into the conch after molting.If the individual did leave the shell after molting, it likely rotated itself and moved over its molt.Such flexibility would have been possible in its soft-shelled condition (Drage et al. 2019) as the trilobite may have been more dorso-ventrally compressible, permitting movement through the limited space (Drage and Daley 2016).This exit from the conch could have taken minutes to days (Zong et al. 2016).Alternatively, the individual may have been trapped behind its molt, unable to escape.The Kõrgessaare Formation does not preserve soft-bodied fossils and, as such, we cannot test this possibility.Due to this limitation of the fossil record, we present both possible outcomes here.
By considering the history of trilobite molting within larger organisms, we can explore when in time and where in the group this cryptic behavior arose (Table 1).Although it was more common within agnostids (see Chatterton et al. 2003;Valent et al. 2008;Fatka et al. 2009;Fatka and Kozak 2014), hiding during molting had an origin in Cambrian agraulid trilobites (Valent et al. 2008).There is an increase in the prevalence of these interactions in the Ordovician asaphids, calymenids, harpetids, odontopleurids, and pliomerids (Table 1).This increase may reflect a biological signal, or it may be sampling bias.Additional examination of trilobite interactions during the Cambrian would help resolve this question.The long record of trilobites concealing themselves within larger animals may also reflect the rise of larger, more effective durophagous predators that targeted smaller trilobites (Brett 1990;2003;Bicknell and Paterson 2018).Additionally, the increased abundance of this behavior likely gave trilobite species a protective advantage, adding to their success.
Habitation of animals in cephalopod shells is also observed in the Mesozoic (Fraaye and Jäger 1995a, b;Davis et al. 2001;Fraaije and Pennings 2006;Vullo et al. 2009;Klompmaker and Fraaije 2012;Landman et al. 2014;Nyborg et al. 2014;Smith and Holland 2016;Fraaije et al. 2020;Bicknell et al. 2021).The apparent increase in the diversity of animals showing this association reflects the use of cavities as a means of shelter, a food source, or for reproduction (Fraaye and Jäger 1995a, b;Nyborg et al. 2014).Despite the Mesozoic record, there are large gaps in information regarding these behaviors and associations, especially in the later Paleozoic trilobites (Table 1).We therefore propose the ongoing examination of cephalopod shells with the aim of documenting the use of these remains by benthic animals across the Phanerozoic.

Fig. 1
Fig. 1 Geological, stratigraphic, and geographical information for specimen location.a Map showing local geology and specimen location (red star) in Estonia.b Paleogeography showing Baltica, 444 mya.Red star indicates the specimen location.Reconstruction constructed using BugPlates (Torsvik and Cocks 2009).c Stratigraphic section of local geology, showing position of the Kõrgessaare Formation ◂

Fig. 2
Fig. 2 Examined Toxochasmops vormsiensis trilobite molt in Gorbyoceras textumaraneum nautiloid shell.TUG 1355-193.a View showing cephalon and trunk in oblique orientation.b View showing trunk table.The symbol "-" indicates that the data was not presented in the manuscript Toxochasmops vormsiensis Rõõmusoks 1998 within the body chamber of a nautiloid cephalopod (Gorbyoceras textumaraneum (Roemer 1861), identified by Björn Kröger pers.comms.)from the Upper Ordovician Kõrgessaare Formation, Vormsi, Estonia.This record presents the first example of cryptic molting for pterygometopid trilobites.