ecological interactions between rhipidomella - Semantic Scholar

PALAIOS, 2010, v. 25, p. 196–208 Research Article DOI: 10.2110/palo.2009.p09-119r

ECOLOGICAL INTERACTIONS BETWEEN RHIPIDOMELLA (ORTHIDES, BRACHIOPODA) AND ITS ENDOSKELETOBIONTS AND PREDATORS FROM THE MIDDLE DEVONIAN DUNDEE FORMATION OF OHIO, UNITED STATES RITUPARNA BOSE,1* CHRIS SCHNEIDER,2 P. DAVID POLLY,3 and MARGARET M. YACOBUCCI 4 1Department

of Geological Sciences, Indiana University, 1001 East 10th Street, Bloomington, Indiana 47405, USA; 2Department of Geology, Energy Resources Conservation Board, Alberta Geological Survey, 402, Twin Atria Building, Edmonton, Alberta T6B 2X3, Canada; 3Department of Geological Sciences, Indiana University, 1001 East 10th Street, Bloomington, Indiana 47405, USA; 4Department of Geology, Bowling Green State University, 190 Overman Hall, Bowling Green, Ohio 43403, USA e-mail: [email protected]

* Corresponding author.

dead shells. Although such boring organisms as clionid sponges are well known from modern shells, fossil examples are encountered less frequently or, as Lescinsky (1996) has stated for encrusting organisms in general, are overlooked. Taylor and Wilson (2002) used the term endoskeletobiont for organisms boring into and infesting the surfaces of organic media (5substrates) whether living or dead; the commonly used terms encruster, encrusting organism, and epizoan generally refer to episkeletobionts. Although highly useful for descriptive precision, Taylor and Wilson’s (2002) terminology for categories of encrusting organisms has not yet gained acceptance in the modern epibiosis literature. We use their terminology herein. Episkeletobionts are useful as autecological or postmortem environmental indicators for their hosts—whether the host was living at the time of encrustation (Rodriguez and Gutschick, 1975; Pitrat and Rogers, 1978; Kesling et al., 1980; Watkins, 1981; Anderson and Megivern, 1982; Brezinski, 1984; Spjeldnaes, 1984; Alvarez and Taylor, 1987; Harland and Pickerill, 1987; Morris and Felton, 1993; Peters, 1995; Lescinsky, 1995; Sandy, 1996; Sumrall, 2000; Morris and Felton, 2003; Schneider, 2003, 2009a; Zhan and Vinn, 2007; Rodrigues et al., 2008) or was dead (Thayer, 1974; Rodriguez and Gutschick, 1975; Watkins, 1981; Anderson and Megivern, 1982; Brezinski, 1984; Gibson, 1992). Episkeletobionts have been studied with regard to the orientation of brachiopod valves (Rudwick, 1962; Hurst, 1974; Pitrat and Rogers, 1978; Kesling et al., 1980; Spjeldnaes, 1984), to the preferred orientation in marine currents (Kesling et al., 1980), to the potential as camouflaging agents for hosts (Schneider, 2003, 2009a), to the attracting or antifouling nature of ornamentation (Richards and Shabica, 1969; Richards, 1972; Carrera, 2000; Schneider, 2003, 2009a; Schneider and Leighton, 2007), and in regard to the purpose of the valve punctae (Thayer, 1974; Curry, 1983; Bordeaux and Brett, 1990). In addition, the presence of encrusting organisms in marine paleocommunities signifies a level of trophic and three-dimensional (3D) complexity in the benthos, particularly in environments where inorganic hard media are lacking. Encrusting organisms respond to and, in turn, create a level of 3D complexity in an ecosystem through secondary tiering on other organisms (Walker and Alberstadt, 1975; Bottjer and Ausich, 1986, Peters and Bork, 1998; Schneider, 2003, 2009a). Encrusting organisms in a community increase its biodiversity and may additionally feed back into community biodiversity and biomass by attracting other encrusters. Fossil episkeletobionts have even been suggested to be good indicators of primary productivity levels in paleoecosystems (Lescinsky and Vermeij, 1995). Often, brachiopods are the host of preference for studying encrustation in the Paleozoic fossil record, usually because of the abundance of specimens and, in the case of many siliciclastic units, ease of removal from the matrix. Indeed, because of the abundance of data collected about brachiopods and their encrusters, episkeletobiont ecological relationships are fairly well understood for the Paleozoic record.

Copyright

0883-1351/10/0025-0196/$3.00

ABSTRACT Only the small (2.0 6 0.1 cm) orthide brachiopod Rhipidomella in the Middle Devonian Dundee Formation exposed at Whitehouse Quarry, Ohio, preserves evidence of interactions with endoskeletobionts and predators (39.6%, n 5 48), as opposed to the slightly larger atrypides, spiriferides, and stropheodonts (n 5 245). All traces of predation and boring by other organisms are lacking on such larger brachiopods as strophomenides and spiriferides, which are more often encrusted in Devonian localities of North America—the Silica Shale of Ohio, Hamilton Group of New York, and Cedar Valley Limestone of Iowa. Punctate shells of Rhipidomella preserve interactions with endoskeletobionts and predators, phenomena less common for punctate brachiopods. All traces on Rhipidomella were preserved as endoskeletobionts; if calcified encrusters were present, they likely were lost postmortem. Several Rhipidomella individuals bear partially repaired grooves from parasitic interactions with sinuous, boring organisms, attributed to ctenostome bryozoans. These parasites bored into the shell along the commissure, likely benefiting from the inhalant and exhalant currents produced by the brachiopod, and in some cases, expanded away from the commissure following the host’s death. The straight, U-shaped borings in one Rhipidomella specimen with boreholes are similar in morphology to Caulostrepsis traces that occurred with their tubes opening along the margins of a modern brachiopod. Other endoskeletobiont traces likely formed on the postmortem shells of Rhipidomella. Predation repair scars are present on two specimens, indicating the presence of predators and the survival of some Rhipidomella individuals from durophagy. No specimens contain evidence of drilling predators. INTRODUCTION The Middle Devonian Dundee Formation, located in the Michigan Basin of North America, has long been recognized as a highly fossiliferous unit. The sedimentology and stratigraphy of the Dundee Formation has been well studied (Bassett, 1935; Sparling, 1988), but paleoecological studies are lacking for major portions of the unit. In general, fossils are well preserved, including encrusting and predation traces that are direct evidence of paleoecological interactions. The nonpredatory boring traces on brachiopod shells from the Dundee Formation are described here for the first time, providing a rare example of paleoecological interactions in a carbonate environment. Episkeletobionts—organisms that adhere to the surface of a shell, also known as encrusting organisms (Taylor and Wilson, 2002)—are well known from the paleoecological literature. Less understood are attached and encrusting organisms that bore into shell material, usually considered to have parasitized live hosts and often found colonizing

g

2010, SEPM (Society for Sedimentary Geology)

PALAIOS

197

ENDOSKELETOBIONTS ON RHIPIDOMELLA

FIGURE 1—Study area. A) Regional geologic structures of Ohio and adjacent states. Black star symbol marks the location of Whitehouse Quarry. Inset illustration of the United States map indicates Ohio shaded in black. B) Stratigraphic section of Whitehouse Quarry. Numbers mark the units. Samples collected from units 10, 11, and 12 (adapted from Wright, 2006).

Brachiopod hosts are useful for investigating influences on such episkeletobiont preferences as media texture (Schneider and Webb, 2004; Rodland et al., 2004; Schneider and Leighton, 2007), size of host (Ager, 1961; Kesling et al., 1980), antifouling strategies (Schneider and Leighton, 2007), and host switching during taxon loss across mass extinctions (Schneider and Webb, 2004). Although other taxa have been utilized for similar investigations, brachiopods remain one of, if not the most, well-understood hosts for Paleozoic episkeletobionts. Endoskeletobionts have rarely been studied in Devonian brachiopods of carbonate environments in North America. Brachiopods from the Middle Pennsylvanian Naco Formation, central Arizona, display a wide diversity of endoskeletobionts that colonized their hosts both during life and after death (Sumrall, 1992; Lescinsky, 1997; Dyer and Elliott, 2003; Irmis and Elliott, 2006). These endoskeletobiont traces on brachiopod hosts included ramifying networks of borings (Dyer and Elliott, 2003). The endoskeletobiont Chaetosalpinx sp. (class Polychaeta; order Serpulimorpha) has been reported to have occurred within the skeletons of the tabulate coral Yavorskia sp. (Favositida, Cleistoporidae) from the latest Famennian (‘‘Strunian’’) in the Etroeungt area (Northern France), and their interaction with the host coral has been described as commensalism (Tapanila, 2004). In the case of Chaetosalpinx, parasitism seems to be more probable (Zapalski 2004; Zapalski et al., 2008). Tapanila (2005) described diversity in boring that occurs in various skeletal marine invertebrates, including tabulate and rugose corals, calcareous sponges, bryozoans, brachiopods, and crinoids. Predatory interactions are those in which the predator benefits by consuming all or part of the prey organism. The interaction is always detrimental for the prey organism, even if it survives, because any damage incurred by the predator must be repaired. Such predation traces as durophagous repair scars indicate not only the presence of a predator and the interaction of said predator with its prey organism, but also trophic complexity in fossil ecosystems. Such data are, therefore, useful for understanding the paleoecology of an ancient system. Repair scars, however, represent failed attempts of the predator to successfully kill and consume its prey; instead, the prey organism has survived and repaired its shell. This evidence, therefore, records the

number of unsuccessful predation attempts, rather than the intensity of durophagous predation pressure (Alexander, 1981; Kowalewski et al., 1997; Alexander and Dietl, 2000; Leighton, 2003). Conversely, the lack of repair scars does not indicate lack of predation pressure; rather, successful durophagous predation destroys the shell, leaving no record of predation (Vermeij, 1982). Hoffmeister et al. (2003) reported a sample of the small-sized brachiopod Cardiarina cordata collected from a late Pennsylvanian limestone unit in Grapevine Canyon, New Mexico, that were frequently drilled (drilling intensity 5 32.7%) by predators or parasites of unknown origin. They suggest that relatively high frequencies of drilling may be found on smaller bodied Paleozoic animals relatively similar to the observed drilling frequency in the Late Mesozoic and Cenozoic, perhaps indicating the presence of a distinct small-bodied predator guild. Although the benefits of studying epi- and endoskeletobionts and predation traces for understanding paleoecological interactions are great, such studies are rare for Devonian carbonate regimes, and lacking entirely for the Dundee Formation. In this paper, we present the results of an intensive survey for synecological interactions within the Dundee Formation brachiopod fauna. GEOLOGICAL SETTING The study area within the Dundee Formation is located at Whitehouse, Lucas County, Ohio, on the northwestern flank of the Findlay Arch, which was a topographic high separating and isolating the Michigan Basin from the Appalachian Basin (Fig. 1A). Specimens for this study were collected from the Whitehouse Quarry, which is east of Whitehouse, Ohio, and was actively worked by the Whitehouse Stone Company in the early twentieth century. The Dundee Formation, which is Eifelian in age (397.5 6 2.7 Ma to 391.8 6 2.7 Ma) (Gradstein et al., 2004), is a predominantly carbonate unit deposited along the southeastern margin of the Michigan Basin (Fig. 1A). The name Dundee Limestone was first used for the rocks exposed in the abandoned Pulver Quarry in Dundee, Monroe County, Michigan (Ehlers et al., 1951).

198

PALAIOS

BOSE ET AL.

FIGURE 2—Trace fossils. A) T-shaped branching network of burrows in the Dundee Formation identified as Thalassinoides. B) Criss-cross pattern of the trace fossils in the same bed of the Dundee Formation in units 10–12. Scale bar 4 cm.

The paleoenvironment of the Dundee Formation was a shallow, tropical, high-energy shelf with intervals of decreased energy and occasional storm events. Most of the Dundee Formation was deposited below normal wave base but above storm wave base; there is evidence that the Middle Devonian sea in this region was shallow, but may have extended occasionally below storm wave base (Hansen, 1999). The Dundee Formation in the Whitehouse Quarry studied herein contains a rich faunal assemblage of brachiopods, corals, bryozoans, mollusks, rare crinoids, and rare tentaculitids originally described by Stauffer (1909), Bassett (1935), and Ehlers et al. (1951). Beds sampled for this study included the fossiliferous units 10, 11, and part of unit 12 from Whitehouse Quarry. The lower unit 10 is a packstone, 0.19 m thick, grading upward into the wackestone of unit 11, about 0.07 m thick, transitioning into the nodular-bedded unit 12 grainstone, 0.06 m thick (Fig. 1B). These units were chosen for study because of their high fossil abundance and diversity. These three fossiliferous units also contained the ichnofossil Thalassinoides (Fig. 2). These burrows either formed in softgrounds within the Cruziana ichnofacies, or in firmgrounds of the Glossifungites ichnofacies (Ekdale et al., 1984). Syndepositional hardgrounds are lacking in these units, and the intensity of burrowing suggests that the seafloor remained unconsolidated during biotic activity. Presence of abundant intraclasts in the sampled units are the remnants of prior firmgrounds or of buried, early lithified sediments that were exhumed, broken, and redistributed as clasts during high-energy events, suggesting intermittent storm activity during sediment deposition. Background sedimentation rates are considered herein to be slow: low enough for soft-sediment burrowing networks to be maintained, but not slow enough for hardground cementation. Thus, trace-fossil associations indicate a shallow marine environment and slow sediment deposition below fair weather wave base. The presence of intraclasts suggests deposition above storm wave base, with tempestite grains reworked into sediment by burrowing organisms. The sediments likely were deposited in a well-oxygenated environment based on the moderate size (0.5–1.0 cm in diameter) of the burrows. MATERIALS AND METHODS Sampling A total of 245 brachiopod specimens were collected in 72 limestone slabs from fossiliferous units of the Dundee Formation. Brachiopod shells were collected throughout the thickness of each sampled bed, although most were recovered from the packstone of unit 10 and the lower part of the unit 12 grainstone. In addition, the taphonomic, sedimentological, and ichnological character of the formation was also

recorded in order to interpret aspects of the depositional environment of this unit relevant to the study. For consistency, the ventral valves of all brachiopod specimens collected were investigated for evidence of activity by episkeletobionts, endoskeletobionts, and predators. Each brachiopod was identified to genus level. Brachiopod specimens were examined under a stereomicroscope using 403 magnification. Under 403 magnification, only Rhipidomella shells showed some evidence of biotic interaction (Figs. 3– 4). The ventral valves of all Rhipidomella were further examined under 1003 magnification. Those with evidence of borings were photographed using a high-resolution digital camera attached to the microscope. Branching grooves, scars, and borehole traces on the external surface of Rhipidomella were observed under the scanning electron microscope (SEM) to determine possible qualitative differences in type and intensity of boring. Fourteen Rhipidomella specimens of comparable size (2.0–2.1 cm) were selected for SEM analysis. These specimens were first dried in an oven at 80 uC for 12 hours; no additional coating was used. Under 30–2003 magnification, the trace types were determined, measured, and photographed. Borehole traces (0.1–0.3 mm in diameter), and sinuous traces formed by an unknown network of borings (0.04–0.2 mm in diameter) were recorded. The specimens are deposited at the Indiana University Paleontology Collections for further investigations. Where possible, endoskeletobiont traces were identified to the most precise taxonomic identification of the trace-makers. The shape, size, and position of these traces were noted. Diameters of boreholes and dimensions of branching grooves were determined using digital photographs and image analysis software. The presence of new shell material partially filling in grooves and scars was considered as evidence of shell repair and its absence as evidence for the lack of shell repair. Quantifying Epizoan Occurrences Following the methods previously established by Kesling et al. (1980), each valve of the 17 best-preserved Rhipidomella hosts was divided into six regions (Fig. 5). Area ratios were used to standardize each region of the shell to the others following this equation (Fig. 5): R~

AR , AT

ð1Þ

where R is the area ratio, AR is the area of each region, and AT is the total area. The frequency of endoskeletobiont activity in each region was recorded by counting individual colonies as one occurrence and then summing for all Rhipidomella hosts. For comparison with the actual frequency of endobiosis, expected endobiosis for each region was calculated by: E~NRi ,

ð2Þ

in which the expected number of endoskeletobiont traces (E) is calculated by multiplying the total number of endoskeletobiont traces (N) for all Rhipidomella specimens by the area ratio (R) for a given region on the Rhipidomella shell (i). Chi-square tests were performed for the total observed and expected endoskeletobionts of the six regions to test for endoskeletobiont preference for location on the Rhipidomella valve. Boring traces—straight and branching—posed a problem for these calculations because these specimens often crossed borders into adjacent regions. For these specimens, frequency of colonization of an endoskeletobiont in an observed region, which also extended into another region, was divided by the total number of regions it inhabited. For example, branching grooves that were observed in all the three anterior regions were counted as 1/3 for each region.

PALAIOS

ENDOSKELETOBIONTS ON RHIPIDOMELLA

199

FIGURE 3—Trace types on Rhipidomella shells (Endoskeletobiont type A). Inset illustrations indicate view of endoskeletobiont traces relative to the entire valve. A) (IU#19011) Y-shaped branching grooves (BG) close to the hinge in the posteromedial region (103). Black arrow points to BG perpendicular to the direction of rib direction. B) (IU#19012-1) Deep BG close to the hinge of the shell in the posteromedial region (163). Black arrow points to BG perpendicular to the direction of rib direction. C) (IU#19012-2) BG along the commissure of the shell covering the antero–left lateral, anteromedial, and antero–right lateral regions (203). Dashed square indicates the healing region of the bifurcating grooves. D) (IU#19012-2) Higher magnification of the dashed square region in C showing the bifurcating BG (613). E) (IU#19012-2) Healing of the bifurcating grooves along the anterior margin magnified in SEM image (803). F) (IU#19012-3) Deep wide BG in a partially recrystallized shell, covering the entire posterior and anterior regions of the shell except the antero–right lateral margin (83). Dashed square points to the BG perpendicular to the direction of rib direction. G) (IU#19012-3) SEM image magnifies the direction of the BG perpendicular to the rib in the postero–left lateral region (403). H) (IU#19014) Deep, wide, BGs close to the hinge along the posteromedial region (163). Dashed square represents some grooves perpendicular to rib direction similar to Figure 3D. I) (IU#19012-3) SEM figure magnifies the crosscutting ribs in the posteromedial region (383). J) (IU#19015) Distinct BG along the anteromedial region of the shell bifurcating toward the commissure. White arrow shows the point where the branching groove ends, probably indicating signs of healing (83). K) (IU#19015) Very deep, BG well captured in SEM (2023). L) (IU#19016) Clear BGs all over the shell surface, branching out from the hinge toward the shell margin and along the lateral sides (83). SEM figure magnifies the grooves in M) (IU#19014) posteromedial and postero–right lateral region (403), N) (IU#19014) anteromedial (503) and O) (IU#19012-3) antero–right lateral region (803). Note: Scale bar 5 mm in views A–C, F, H–J, L; scale bar 0.5 mm in views D–E, G, I, K, M–O.

RESULTS Frequency of Traces Out of 245 brachiopod specimens counted from the samples, 163 were identified as Strophodonta, 48 as Rhipidomella, 17 as unidentified rhynchonellids, 9 as Atrypa, and 8 as Mucrospirifer (Table 1). Of all of the brachiopod taxa encountered, only Rhipidomella bore endoskeletobiont traces and predation scars. Nineteen specimens out of 48 Rhipidomella, that is, 39.6 percent of the shells, bear evidence of endoskeletobionts on their examined ventral valves (Table 2). SEM analysis on fourteen specimens closely examined the depth and direction of sinuous traces and gave evidence for how these cross cut the Rhipidomella ribs (Figs. 3C–D, F–G). Some shells had branching grooves that could be produced by borers: branching endoskeletobiont type A (Figs. 3A–O). Some shells bore scars with disrupted growth lines (Figs. 4A–E) and one had numerous, shallow, large boreholes with elongated straight grooves: straight endoskeletobiont type B (Figs. 4F– G). Preservation of Shells Specimens of Strophodonta, Mucrospirifer, atrypides, and unidentified rhynchonellides were moderately well preserved, meaning largely uncrushed and unbroken with little or no abrasion and minimal postdepositional distortion. Some Strophodonta, Mucrospirifer, and

rhynchonellide shells were recrystallized and lacked surface detail, whereas others were abraded. Many stropheodontides occurred as single valves, evident from preserved cardinal processes. Atrypides could not be identified to genus because of the similarity between genera and poor preservation of key identifying details. Most Rhipidomella shells were well preserved, containing all external features. A few Rhipidomella shells were recrystallized, and others show evidence of a little abrasion and fragmentation, including exposure of punctae. Microscopic examination of nonrecrystallized Rhipidomella specimens reveals endoskeletobiont traces on hosts in contrast to other brachiopod taxa; even the large nonrecrystallized stropheodontids bore no evidence of endobiosis. Types and Nature of Traces on Rhipidomella Shells Endoskeletobiont Type A (Branching Grooves).—(13 specimens). Endoskeletobiont traces of curved, deep or shallow, narrow or wide, bifurcating grooves range from 0.01 to 0.20 mm in diameter on the external surface of the ventral valve. Traces of Y-shaped, wide, branching grooves (0.03 mm deep, 0.08 mm in diameter) were observed near the hinge on some Rhipidomella hosts (Figs. 3A–B). On one specimen, wide, branching groove traces were parallel to the shell margin and showed signs of active repair by the brachiopod (Figs. 3C– E), whereas on other specimens these traces covered much of the shell surface, originating at the hinge and branching outward toward the commissure (Figs. 3F–G, L–O). Another specimen also exhibited deep,

200

PALAIOS

BOSE ET AL.

FIGURE 4—Trace types on Rhipidomella shells (Endoskeletobiont type B 5 straight grooves, type C 5 boreholes, and type D 5 scars). A) (IU#19017) Two deep scars, one in the posteromedial region close to the hinge, the other on the antero–right lateral region marked in dashed square; punctae exposed near margin (83). Black circle outlines the scar showing evidence of repair. Inset illustrates the scar of the antero–right lateral region in SEM (1013). B) (IU#19018) Two parallel deep scars (in dashed square) along the margin of the antero–left lateral region (12.53) similar to the scars on Paraspirifer collected from Silica (Sparks et al., 1980, pl. 9, fig. 4). White arrows indicate the evidence for shell repair. C) (IU#19018) SEM illustrates the scars at 703 magnification. D) (IU#19018) Deep scar on the right lateral margin (within white dashed square marked as a black line), compressing one side of the shell surface with respect to the other side (12.53). E) (IU#19018) Magnified cleft is illustrated in SEM (533). F) (IU#19019) Central straight elongated grooves branching out from the hinge with shallow larger boreholes along the postero–left lateral, postero- and anteromedial regions of the shell (83). One center borehole marked by a white line. G) (IU#19019) SEM illustrates the cylindrical (longer than width) shallow boreholes along the postero–left lateral region (713). H) (IU#19020) Numerous branching and V-shaped grooves on the shell surface and a large scar along the commissure (83). Note: Scale bar 5 mm in views A, B, D, F, H; scale bar 0.5 mm in views C, E, G.

wide, branching groove traces only near the brachiopod hinge (Figs. 3H–I). One partially recrystallized specimen had a deep, wide, slightly branching groove located centrally that shows signs of shell repair along its margins (Figs. 3J–K). Traces of deep, but much narrower (0.01 mm), dendritic grooves extended from the commissure

toward the hinge of some Rhipidomella shells, crosscutting the ribs, and had no evidence of shell repair. Endoskeletobiont Type B (Straight Grooves).—(1 specimen). Elongated straight grooves ranging from 0.9 to 1.2 mm in diameter were observed in one Rhipidomella valve in the central portion of the shell, extending toward the hinge (Fig. 4F). Endoskeletobiont Type C.—(1 specimen). Circular borehole traces, ranging from 0.16 to 0.28 mm in diameter on the external valve surface, were found positioned between the ribs of Rhipidomella specimens. One host contained ten large circular holes, 0.16–0.28 mm in diameter, that were located centrally, with grooves terminating in, or extending from, holes. This aggregation of boreholes covered over half of the surface area of the valve (Figs. 4F–G). Durophagous Repair Scars.—(4 specimens). Crushing scars indicate damage caused to the brachiopod while alive by either abiotic (mechanical breakage) or biotic factors (a predator or parasite) that must be repaired, affecting the growth of the shell and leading to irregular, deformed growth lines. Two deep tapered scars were observed in one specimen; one located in the central region and the other in the lower right lateral region (Fig. 4A). In this specimen, there is a clear edge to the scar that is not parallel to the growth lines, and the associated ribs are discontinuous and deflected, indicating a repair scar. Another Rhipidomella specimen bore two deep parallel and geometric tapered scars along the lower left lateral margin (Figs. 4B–C). The same host specimen contains a deeply cleaved scar, almost parallel to the shell TABLE 1—Taxa and abundance of sampled brachiopods.

FIGURE 5—Rhipidomella valve divided into six regions for endoskeletobiont frequency study; PLL 5 postero–left lateral, PM 5 posteromedial, PRL 5 postero–right lateral, ALL 5 antero–left lateral, AM 5 anteromedial, and ARL 5 antero–right lateral. Numbers represent the area ratios of each grid across the Rhipidomella host valve. Scale bar 1 mm.

Brachiopod genus

Number of shells

Percentage (%)

Strophodonta Rhipidomella Unidentified rhynchonellids Atrypa Mucrospirifer Total

163 48 17 9 8 245

66.53 19.59 6.93 3.67 3.26 100.00

PALAIOS

201

ENDOSKELETOBIONTS ON RHIPIDOMELLA

TABLE 2—Morphology and abundance of biological traces on Rhipidomella shells. Note that only 4 specimens show live-live interactions for endoskeletobiont type A and 2 specimens for durophagous scars. Overall 19 out of 48 potential Rhipidomella hosts (39.6%) show these endoskeletobiont and predation traces. Types of biological traces

Number of shells

Percent of traces (%)

Percent of shells (%)

Branching grooves (Endoskeletobiont type A) Straight grooves (Endoskeletobiont type B) Large boreholes (Endoskeletobiont type C) Durophagous scars No traces Total shells with traces Total shells

13 (4 live interactions) 1 1 4 (2 live interactions) 29 19 48

68.42 5.26 5.26 21.05

27.08 2.08 2.08 8.33 60.40

costae along the lower right lateral margin, with one side of the shell margin compressed (Figs. 4D–E). Another specimen bears a central deep scar, containing punctures, near the commissure of one specimen. A large scar near the commissure of another specimen appears abraded, with the shell material fragmenting in the scar area (Fig. 4H). Overlapping occurrences of endoskeletobiont type B and type C— two types of endoskeletobiont borings, or predation plus environmental impedence—could result in complex traces on the shell. One specimen shows evidence of continuous, shallow, large, perpendicular holes, ranging in size from 0.1 to 1.0 mm in diameter along with centrally located elongated straight grooves extending toward the hinge (Fig. 4F). Another specimen had a crushing scar, ,6 mm wide, with several V-shaped branching grooves along the valve surface (Fig. 4H). Location of Traces The frequency of biotic interactions varies among the six regions on the ventral valves of the Rhipidomella hosts (Fig. 6). The posteromedial region contains the most frequent occurrence of endoskeletobiont type A (branching grooves) traces, although the postero–left lateral and anteromedial areas also have a high abundance of these traces. Durophagous scars are most frequent in the antero–left lateral region. The postero–left lateral and anteromedial regions have the highest frequency of boreholes—endoskeletobiont type C—in the single bored specimen. Combining all endoskeletobiont data, the observed frequency of endobiosis across all regions of the Rhipidomella shell is significantly different than expected if the endoskeletobionts randomly bored any portion of the shell (chi-square; p , 0.01) (Table 3, Fig. 7). Specifically, the postero–right lateral region was bored at a lower rate than expected

100.00 100.00

rate (chi-square; p 5 0.008). Conversely, the opposite region, postero– left lateral, was bored at a frequency higher than expected (chi-square, p 5 0.017). The remaining regions do not show any significant difference between expected and observed endoskeletobiont frequency (Table 4, Fig. 7). DISCUSSION Despite the highly diverse brachiopod community in the Dundee Formation and higher relative abundance of stropheodontides as compared to other brachiopod taxa, only the smallest brachiopod genus Rhipidomella (diameter roughly 2.0 6 0.1 cm) of the orthide group retains some evidence of endoskeletobionts and furthermore, only when viewed at very high magnification. Concavo-convex stropheodontides generally are thought to be oriented with their concave valve upright during life (e.g., Rudwick, 1970; Richards, 1972; however, see Lescinsky, 1995 for an alternate hypothesis). All of the stropheodontids were oriented with their ventral valves up, suggesting overturning. The somewhat larger stropheodontids (3.0–4.0 cm) from the locality bore no evidence of any biologic activity. Of all of the brachiopod taxa encountered, only Rhipidomella bore endoskeletobiont traces and predation scars. Due to the uniform size of all Rhipidomella valves investigated (2.0 6 0.1 cm), however, it was not possible to discern a relationship between valve size and degree of shell damage within the genus. Is There Any Taphonomic Influence on Brachiopods? While convincing evidence for endoskeletobiont activity has been documented for Dundee Formation Rhipidomella specimens, no evidence for episkeletobiont activity has been found. Two possible explanations for this discrepancy exist. Episkeletobionts were either not preserved or were nonexistent on the Rhipidomella specimens in this study. Given that the sedimentology of these beds suggests a highenergy environment and likely storm events, it is possible that calcified episkeletobionts were present, but lost through taphonomic processes. TABLE 3—Sum of the observed and expected number of endoskeletobiont and predation trace occurrences for three types of endoskeletobionts (type A 5 branching grooves, type B 5 straight grooves, and type C 5 boreholes) and durophagous scars across six shell regions. The six grids are as follows: PLL 5 postero–left lateral region; PM 5 posteromedial region; PRL 5 postero–right lateral region; ALL 5 antero–left lateral region; AM 5 anteromedial region; ARL 5 antero–right lateral region. Endoskeletobiont and predation activity Regions

FIGURE 6—Total standardized frequency of endoskeletobiont activity across each region for 17 Rhipidomella hosts. (Note: Standardized frequency 5 Frequency of colonized endoskeletobionts on host Rhipidomella) The six grids are as follows: PLL 5 postero–left lateral region; PM 5 posteromedial region; PRL 5 postero–right lateral region; ALL 5 antero–left lateral region; AM 5 anteromedial region; and ARL 5 antero–right lateral region.

Postero–left lateral Posteromedial Postero–right lateral Antero-left lateral Anteromedial Antero–right lateral

PLL PM PRL ALL AM ARL SUM

Observed

Expected

27.5 32.5 13 21.5 25 8 127.5

24.735 24.225 24.735 13.26 26.775 13.26 126.99

202

PALAIOS

BOSE ET AL.

TABLE 4—Results of chi-square test of observed versus expected endoskeletobiont and predation trace occurrences. A p-value , 0.05 indicates either more or less biological activity in that shell region than expected, as indicated.

PLL vs. sum of other areas PM vs. sum of other areas PRL vs. sum of other areas ALL vs. sum of other areas AM vs. sum of other areas ARL vs. sum of other areas

FIGURE 7—Observed versus expected endoskeletobiont activity across the six regions of Rhipidomella hosts. Observed values are the actual frequency of encrustation for each region of the shell; expected values are calculated as described in the text. PLL 5 postero–left lateral region; PM 5 posteromedial region; PRL 5 postero–right lateral region; ALL 5 antero–left lateral region; AM 5 anteromedial region; and ARL 5 antero–right lateral region.

Packstones and grainstones, common in the Dundee Formation of the Whitehouse Quarry, represent high-energy environments; in this study, units 10 and 12—from which the majority of specimens came—were a packstone and a nodular-bedded, intraclast-bearing grainstone, respectively. Fragmented and abraded shells suggest that some of the horizons from which the fossils were collected were storm beds; burrows mixing the fossils occurred after storm events. These data would further support the loss of encrusting organisms to abrasion, but other reasons need to be evaluated for reporting episkeletobiont loss, e.g., shells in motion in the seafloor, prior to drawing this conclusion. Local shifting of shells on the seafloor can affect encrustation. Encrustation rates were very low on modern mollusk shells derived from the carbonate reef and lagoon systems in the northeastern Caribbean and transported into nearby, high-energy beach environments, likely because of shells being in constant motion (ParsonsHubbard, 2005). In addition to constant motion, which results in abrasion, a shell that does not rest cannot attract episkeletobionts; a moving seafloor is difficult to encrust (Parsons-Hubbard, 2005). McKinney (1996) demonstrated the rapidity with which epizoans were removed from shells during experimental abrasion; natural abrasion during storm events and in other high-energy environments would produce similar results. Furthermore, one of us (Schneider) noted that, among Terebratalia and Laqueus brachiopods dredged from Puget Sound, Washington, only live specimens bore episkeletobionts; all dead specimens from both rocky and muddy environments were devoid of skeletal and soft-bodied encrusters. Rodland et al. (2006), however, have shown that taphonomic processes need not erase all evidence of biotic interaction through time, a fact evident from the rich endoskeletobiont activity preserved on Rhipidomella in our study. All the brachiopods from this study appear to have undergone type III shell alteration (Emig, 1990), where the large anterior part of the valves is frequently mechanically fragmented into small pieces. Dissolution of calcium carbonate or cryptocrystallization, according to local environmental conditions, likely were other processes that affected shell layers (Emig, 1990). These processes could readily erode any trace of episkeletobionts; however, endoskeletobionts and predation traces are preserved on Rhipidomella and, therefore, should have been preserved, if present, on larger taxa affected by the same taphonomic processes resulting in similar shell-surface alteration. In other words, traces of endoskeletobionts and durophagous predation scars should be found on the larger brachiopod taxa as well as Rhipidomella if such traces were originally present, as all brachiopod taxa were equally affected by shell breakage and fine-scale cryptocrystallization. Although we cannot rule out differential taphonomic effects

p values

Observed is — than expected

0.549 0.219 0.008 0.017 0.17 0.123

LESS MORE

on large and small brachiopods, this sort of taphonomic study is beyond the scope of this paper. Unit 11 was a wackestone, representing a lower energy environment than the bounding lower and upper beds. Likely, brachiopod specimens underwent fewer episodes of transport than in preceding or subsequent environments. Fossils in unit 11, therefore, were likely autochthonous rather than transported into the locality. Brachiopods in the higher energy environments of units 10 and 12, however, were very similar in diversity to those preserved in the wackestone of unit 11. These diversity similarities between the sampled lithological facies suggest that the brachiopods of units 10 and 12 underwent only local transport or reworking, and that the brachiopod biofacies were essentially the same from unit 10 through unit 12. Exposure history can control the surface area available for colonization and the temporal window in which encrustation may occur, so this factor should be taken into account (Best and Kidwell, 2000; Rodland et al., 2004). Higher levels of damage observed on some subsets of Rhipidomella valves could be a consequence of greater exposure time and such postmortem degradation events as storm transport, reworking, or postburial diagenesis. Particular types or higher intensities of endoskeletobiont activity on some highly abraded Rhipidomella valves could be a consequence of relatively higher postmortem residence time of these shells on the seafloor as compared to the other brachiopod taxa. All brachiopod taxa occurred in all three units. The presence of bioturbation in the sediments, however, increases the likelihood that the entire brachiopod assemblage is time averaged to some extent and that high-resolution facies changes are lost. We cannot rule out the possibility that Rhipidomella specimens that appear to be contemporaneous with other brachiopods mixed from other time-horizons or bioturbated, and hence lost, microfacies, because of the small size of Rhipidomella compared to other taxa and the high degree of bioturbation. Of all brachiopod taxa sampled, only Rhipidomella specimens bear signs of endoskeletobionts and durophagous predation scars regardless of preservation. The lack of observed episkeletobionts and endoskeletobionts in all the host brachiopods from the Dundee Formation could be due to abrasion, but this is unlikely. Endoskeletobionts leave deep traces within the shell material that would readily be preserved despite abrasion of the shell surface, particularly on hosts with ribs or other external ornament capable of providing shelter. Furthermore, durophagous predation scars, which would be preserved even in recrystallized or abraded shells, are absent on larger host brachiopods. Whether the larger brachiopods and Rhipidomella existed contemporaneously or were redistributed and mixed from time averaging is unknown; therefore, either (1) Rhipidomella was encrusted in communities containing only the one brachiopod taxon and then later timeaveraged with other communities containing large brachiopods, or (2) endoskeletobionts selectively colonized only Rhipidomella shells in multitaxa brachiopod communities. Identification of Tracemakers Many of the endoskeletobiont traces can be attributed to uncalcified, boring bryozoans. Traces of deep, wide branching grooves on

PALAIOS

ENDOSKELETOBIONTS ON RHIPIDOMELLA

Rhipidomella resemble the traces made by ctenostome bryozoans on late Eifelian crinoid columnals collected from Devonian strata in the Skaly Beds in the Holy Cross Mountains, Poland (Gluchowski, 2005). The branching traces on Rhipidomella in this study resemble the grooves of an Ordovician through recent trace made by a ctenostome bryozoan with a similar diameter, 0.15–0.40 mm. Moreover, ctenostome bryozoan zooids form a characteristic network of delicate, uniserial, branching lines and are uncalcified forms boring into calcareous media (McKinney and Jackson, 1989; Wood and Marsh, 1996), traits seen in the morphology of the branching grooves on Rhipidomella from the Dundee Formation. Botquelen and Mayoral (2005) have reported irregular, narrow, sinuous branching bioerosion traces made by tunnels of ctenostomate bryozoans on Lower Devonian internal molds of brachiopods from the Rade De Brest, Armorican Massif, France. These are oriented parallel to the host medium and range from 10 to 100 mm (0.01–0.1 mm) in diameter. Both the diameter and shape of the traces found in the Dundee Formation brachiopods vary within this range, with only a few traces wider than the range for boring ctenostomate bryozoans from France. The branching endoskeletobiont type A traces on the Dundee Formation Rhipidomella likely were the product of ctenostome bryozoans. Boreholes in shelly benthos are often made by parasites or by predators of live hosts, or organisms boring into dead shells. Nonsinuous, perpendicular borings in Paleozoic brachiopods have been attributed to many organisms. The morphologies of these multiple boreholes in Rhipidomella lacked the tapering at the margins with concentric or aconcentric openings caused by stratigraphically younger predatory gastropods studied by Arua and Hoque (1989). Most nonpredatory borers excavate multiple holes (e.g., sponges, Tapanila, 2006; polychaetes, Kern, 1979; worms, Cameron, 1969; Smith et al., 1985). Multiple borings on the single Rhipidomella specimen (Fig. 4F) could be attributed to Paleozoic worms, which ranged from 0.05 to 1 mm in diameter, because other evidence for worms exists in the form of worm tubes (Coleolus sp.) from the fossiliferous units of the Dundee Formation (Wright, 2006). Borings attributed to worms are roughly cylindrical, generally longer than their width, and usually oblique to the host’s surface (Cameron, 1969; Smith et al., 1985), very similar in morphology to those found on the Rhipidomella of this study from the Dundee Formation (Figs. 4F–G). Thus, the endoskeletobiont type B traces (straight grooves) are identified as worm borings. On the same specimen with the multiple borings, the straight Ushaped borings are similar in morphology to Caulostrepsis traces attributed to the polychaete Polydora (Rodland et al., 2006; Rodrigues et al., 2008). The U-shaped boring in this specimen, therefore, were likely caused by a polychaete worm. The identification of the organism that caused the large boreholes in the Dundee Formation Rhipidomella is uncertain, except that a predator likely did not cause the holes, as they do not meet the criteria for predatory drilling (Carriker and Yochelson, 1968). Borings produced by drilling predators are recognizable in that they have a distinct morphology (e.g., type A or type B holes of Ausich and Gurrola, 1979), they often are singular, circular, complete drills perpendicular to the shell (Carriker and Yochelson, 1968), and they typically occur in such optimal areas as the region of the shell protecting the muscle field versus the area that covers the mantle cavity (Leighton, 2001; Kelley and Hansen, 2003). Furthermore, given that the borings show no sign of repair or growth response to parasitism, these borings likely caused, or occurred after, the death of the brachiopod, such that the host was unable to initiate repairs. Predation scars resulted from damage by an attacking durophagous predator while the Rhipidomella brachiopods were alive; if the brachiopod survived, it repaired the damage by adding new shell material, resulting in a characteristic pattern of deflected growth lines (Leighton, 2001) (Figs. 4A–C). One specimen bears two nonlethal predation scars, one located medially and the other on the right

203

anterolateral region (Fig. 4A). Another Rhipidomella specimen with two parallel scars on the lower left anterolateral margin (Fig. 4B) shows evidence of a crushing attempt by a predator, as suggested by distortion in the ribbing and intense, and likely rapid, repair in the area of the scar (L. Leighton, personal communication, 2008). SEM imaging at a higher magnification (703) shows the depth and intensity of the scarred region (Figs. 4C–E), and Fig. 4B shows repair of the scar in the next lower successive layer as compared to the deep scar in the upper layer. Other specimens contain anterior and lateral scars that resemble those attributed to shell-crushing sharks in Chesterian brachiopods from the Confusion Range of west-central Utah, with scars positioned anteriorly or laterally (Figs. 4A–B; Alexander, 1981). Nautiloids also were efficient predators, equipped with crushing beaks and were major predators from the Ordovician through the Devonian (Alexander, 1986). One orthoconic nautiloid specimen was noted from the sampled units, which may have contributed to local predator activity. Other such durophages as placoderm and chondrichthyan fishes and phyllocarid and eumalacostracan arthropods, however, were also active predators during the deposition of the Dundee Formation carbonates (Signor and Brett, 1984; Leighton, 1999). The Rhipidomella predation scars could have been produced by any one of these shell-crushing predators. Although the scars on Rhipidomella described above appear to be characteristic of predator activity and subsequent damage repair, repair of abiotic breakage during the life of the brachiopod cannot be ruled out (Sarycheva, 1949; Alexander, 1981). Scars in brachiopod shells can be produced by phenomena other than predators; mechanical damage and impedance by objects (e.g., other organisms) also can deflect brachiopod growth (Ferguson, 1962; Alexander, 1981). Other shell disruptions and scars have been interpreted as being the result of parasites (e.g., Cornulites, Hoare and Steller, 1967), but the extent and morphology of injuries present on brachiopods of this study suggest that these scars were the result of durophagous predators, not parasites. One specimen that bears a predation scar also has a cleft-like indentation along the commissure, a feature of growth deflection that might not be attributed to a predator (Figs. 4D–E). Cleft-like damage was common in such Late Ordovician brachiopods as Rafinesquina from the Cincinnatian Series of the tristate area of Indiana-Kentucky-Ohio (Alexander, 1986, fig. 5.1, p. 276). This Rhipidomella specimen may have experienced impedance to growth along the margin at the area of deflection, although whether the impedance was another organism or an abiotic object cannot be discerned. Are the Endoskeletobiont and Durophagous Predation Traces Random? Trace occurrence varies among the six regions sampled on the ventral valves of the Rhipidomella hosts (Fig. 6). Durophagous predation scars are most frequent in the antero–left lateral region, suggesting a preferred attack scheme by an unknown predator, an example of a frequently failed attack direction, or a repeated mishandling orientation by the predator. Ctenostome bryozoans most heavily bored the posteromedial region of the valve with branching grooves, although the postero–left lateral and anteromedial also have a high frequency of these traces (endoskeletobiont type A; Fig. 6). The postero–left lateral and anteromedial regions have the highest frequency of straight grooves and boreholes (endoskeletobiont type B and type C) in the single bored specimen. These location preferences of the endoskeletobionts could imply a preference for settling and growing near their hosts’ posterolateral inhalant currents. Posteromedial, postero–left lateral, and anteromedial regions that show higher frequency of endoskeletobiont traces, suggest that bryozoans were utilizing both the hosts’ inhalant and exhalant currents (Figs. 3C, E, G–H, M). Thus, these endoskeletobiont traces are not random and likely the result of a live

204

BOSE ET AL.

endoskeletobiont-host Rhipidomella interaction. This nonrandom distribution of borings on Rhipidomella host could possibly be a real, biological signal with the antero–left lateral region having greater exposure to colonization by endoskeletobionts, possibly due to the life orientation of their hosts relative to water current direction, similar to the distribution of episkeletobionts on Paraspirifer brachiopods from the Silica Shale (Kesling et al., 1980). As there was no discernable difference in frequency of endosymbiosis on the anterior or posterior regions of the shell, it is assumed that settling larvae did not prefer one region over the other, preferring only the left side of the shell surface. Is Endoskeletobiont Activity Pre- or Postmortem? In most cases, direct evidence of live episkeletobiont–host relationships is rare. In the present study, a live host–live endoskeletobiont relationship can be inferred from evidence of shell repair in a few Rhipidomella specimens and the location of these endoskeletobionts with respect to host inhalant currents (Figs. 3A–J). Four Rhipidomella specimens show probable evidence of healing of endoskeletobiont (type A) traces along the anterior commissure (Figs. 3A–J). In several instances (Figs. 3K, M–O), it is difficult to determine whether endoskeletobiosis occurred on a live or dead Rhipidomella host. One host specimen bore a series of branching grooves radiating from the hinge toward the commissure (Fig. 3L). These traces covered the entire external valve surface with no evidence of utilization of host inhalant or exhalant current, and no evidence of shell repair. Given the nature of these traces when compared to those considered parasitic, it is possible that this ctenostome bryozoan colony encrusted a dead brachiopod, but whether the host was originally alive or dead at the time of settlement of the metamorphosing larva cannot be determined precisely. Another Rhipidomella specimen bore numerous branching grooves originating from the commissure and progressing toward the hinge. In this case, the larva clearly settled near the commissure, but the colony expanded hingeward, rather than following the commissure for best advantage of brachiopod-generated currents. Additional groovelike borings along the central region of the same host indicate additional bryozoan activity in these areas, away from the commissure and beneficial feeding currents (Fig. 3F). In another specimen, the ctenostome borings extend from the commissure and lateral margins and converge toward the hinge (Fig. 3G). By not following the commissure and instead deflecting away from the commissure, this mature bryozoan colony did not obviously utilize host inhalant and exhalant currents. In the Devonian episkeletobiont fossil record, auloporids and cornulitids frequently display preferential growth along or toward the commissure, particularly on alate spiriferides and large atrypides, in order to take advantage of feeding and respiration currents actively generated by the host’s lophophore (e.g., Ager, 1961; Hoare and Steller, 1967). Such other Devonian episkeletobiotic organisms as bryozoans, hederellids, and spirorbids do not consistently exhibit preference for the commissural area or host-generated inhalant and exhalant currents. In the current study, lack of preference for commissure-oriented or commissure-parallel growth of the entire colony, therefore, does not indicate either live or dead status of the host, but rather implies a nondependence of the mature ctenostome colony on any actively host-generated feeding and respiration currents. Thus, live or dead status of some Rhipidomella hosts, particularly those with no evidence of repair of ctenostome borings, cannot be readily discerned. There is some suggestion, however, of preferential settlement of ctenostome larvae near the commissure and potential expansion of the colony away from the commissure after the host’s death, but instances are too few for broad extrapolation. Endoskeletobiont and Predatory Interactions with Rhipidomella In the fossil record, relationships between fossil organisms can sometimes be recognized, especially in the case of epibiosis and

PALAIOS

predation. Predatory relationships, always detrimental to the prey, are recognized by distinctive repair scars (failed predatory attempts) and drillholes (successful predatory attempts). Episkeletobiotic and endoskeletobiotic relationships can be mutualistic, commensal, or parasitic (see review in Taylor and Wilson, 2003). Endoskeletobiont Type A (Branching Grooves).—Branching grooves observed in Rhipidomella shells cut across ribs in repetitive patterns, suggesting that an organism actively bored through the rib (Figs. 3A– O). One specimen bore a deep, straight groove produced by a ctenostome bryozoan, branching near and deflecting toward the commissure (Fig. 3F). Shell repair is evident in some of these borings, supporting a relationship between a live endoskeletobiont and a live host, at least during a portion of the endoskeletobiont’s lifespan. It is possible that some of the ctenostome colonies creating these branching grooves may have benefitted from host-generated currents by settling near commissures as larvae (Figs. 3E, J–K) or by concentrating in postero–left lateral and anteromedial regions (Figs. 3C, E, G). Other studies have shown that borers located close to the brachiopod’s commissure are known to have taken advantage of the increased water flow created by the host’s inhalant current system to maximize feeding efficiency (Pickerill 1976; Daley, 2008). The relationship between these boring ctenostome bryozoans and their Rhipidomella hosts is considered to be parasitic. Feeding from the exhalant current of the host would cause no detriment to the brachiopod, and consequences of utilizing the host’s inhalant current are undetermined. Boring into and, therefore, destroying shell material was harmful to the living host. Given that some of the host Rhipidomella brachiopods exhibit repair attempts to these borings, it is evident that these borings were not tolerated by the brachiopod. Endoskeletobiont Type B (Straight Grooves).—One Rhipidomella specimen bears wide, straight grooves that end in large boreholes, originating within the central region and progressing hingeward (Fig. 4F). Close to the hinge, the groove bifurcates, becoming a boring within the shell that continues until it truncates at the hinge. Unlike other groovelike borings, an enigmatic wormlike organism likely created these types of borings, perhaps similar to those attributed to Devonian polychaete worms (Cameron, 1969). This interpretation is further supported by a recent study by Rodrigues et al. (2008) who have shown that live polychaetes resided inside Caulostrepsis borings in live brachiopods. The straight U-shaped borings present on the same specimen of Rhipidomella are similar in morphology to these Caulostrepsis traces attributed to the polychaete Polydora (Rodrigues et al., 2008). Endoskeletobiosis may have occurred during the life of the Rhipidomella, based on the location of the traces. It is more likely, however, that the brachiopod was dead at the time of boring, suggested by the extent of the boring within the shell and lack of repair to the damaged valve. Endoskeletobiont Type C (Large Boreholes).—Boring is accomplished by secretion of acids or mechanical rasping (Taylor and Wilson, 2003). In some groups (e.g., clionid sponges, ctenostome bryozoans), the borer remains stationary within this hole, such that the external shape of the organism is maintained in the shape of the hole and, as such, is treated as a body fossil (Pohowsky, 1978). In the single specimen with multiple, nonpredatory bore holes, the anteromedial and postero–left lateral regions were most frequently bored (Fig. 6). Since there was only one specimen with bore holes, it is impossible to determine if there was stereotypy for boring position. It is also difficult to conclude any potential cause for these regions being bored. The unrepaired nature of the holes indicates boring either of a live host, just before its death, or on a dead host, suggesting a postmortem domicile or feeding trace (Fig. 4F). Durophagous Predation Scars.—Repair scars and drill holes are common in Devonian brachiopods, such as frequently drilled brachiopods from the Ludlowville and Moscow Formations of New York, including several specimens that showed signs of repair (Smith et al.,

PALAIOS

ENDOSKELETOBIONTS ON RHIPIDOMELLA

1985). In another example, several taxa of brachiopods from the Silica Formation, which overlies the Dundee Formation, also were the prey of crushing and drilling predators (Alexander, 1986; Leighton, 2003). It is interesting to note, however, that modern rhynchonelliform brachiopods are rarely attacked by predators due to their unpalatability while drilling rates for coexisting bivalves in the same locality are significantly higher (Thayer and Allmon, 1990; D.L. Rodland, personal communication, 2009). Biconvex brachiopods were more likely to explode when crushed than to suffer breakage that could easily be repaired; the fossil record for predatory scars on brachiopods is, thus, generally higher in concavo-convex brachiopods than most biconvex brachiopods, including orthides (Alexander, 1986). In addition to potential explosion of the shell, apparent predation frequency on orthide brachiopods may have been further reduced from actual, higher predation pressure. Predation frequencies on such orthides as those of Schizophoria of the Silica Formation may appear lower because predators may have had easier access through an open delthyrium in the orthide, thereby bypassing the protective shell and dispensing with the need for energetically and temporally costly drilling (Leighton, 2002). In a study of brachiopods from the Upper Pennsylvanian Magdalena Group, a limestone unit exposed in Grapevine Canyon, New Mexico, Hoffmeister et al. (2003) demonstrated that the small brachiopod Cardiarina cordata was selectively drilled over larger Olygothyrina alleni and Coledium bowsheri by predators or parasites of unknown origin (C. cordata drilling intensity 5 32.7%). Similarly, Ahmed and Leighton (2009) found that, in the Upper Devonian Lime Creek Formation of Iowa, the small brachiopods Devonoproductus and Douvellina were frequently prey for crushing predators. The Middle Pennsylvanian Vanport Shale of Ohio also displays evidence of predatory damage, including apertural breakage and boreholes, on small (,15 mm) components of the fauna, including the brachiopod Composita, the nautiloid Pseudorthoceras knoxense, and the microgastropod Microdoma conicum (Koy and Yacobucci, 2001; Yacobucci, personal observation, 2008) These occurrences support the idea that the smaller Rhipidomella hosts could be preferentially selected in lieu of larger brachiopods by Paleozoic predators, and bolster the assertion of Hoffmeister et al. (2003) that a distinct small-bodied predator guild may have been an important component of Paleozoic ecosystems. Rhipidomella, although punctate, certainly was the focus of at least a few predation attempts, as revealed in several repaired attacks. On one specimen, the evidence of a repaired durophagous predation attempt in the central region and subsequent shell repair by the secretion of younger skeletal material suggests that the brachiopod shell was alive during the time of damage (Fig. 4A). This scar shows a period of cessation of growth, followed by shell production and resulting in deformation of growth lines. On the same specimen, a second scar on the right lateral region obscures and distorts growth lines (Fig. 4A). Another specimen bears two deep, parallel, and tapered scars located near the lower left lateral margin (Fig. 4B). This specimen also shows repair along the anterior-most commissure through a minor distortion of ribs. Furthermore, these scars are oriented in a manner that strongly suggests a biting attempt, similar to other scars noted on brachiopods in the Middle Devonian Silica Formation (Sparks et al., 1980, pl. 9, fig. 4). In the current study, Rhipidomella brachiopods in the Dundee Formation certainly were the target of Devonian durophagous predators, but not drilling predators, and were bored by endoskeletobionts. There is no evidence of endoskeletobiosis or predation of other brachiopods in fossiliferous units 10 through 12 of the Dundee Formation, but Rhipidomella was obviously both attacked and fouled. We suggest four potential hypotheses for these phenomena: 1. In these paleoecosystems, crushing predators could likely only take small prey and, thus, larger brachiopods were within a large size refuge, too large for predators to attack.

205

2. Conversely, it is possible that predators of large brachiopods were more capable than the smaller predators of Rhipidomella, providing the Rhipidomella more opportunities to survive a predatory attempt. 3. In the case of epibiosis, punctate Rhipidomella is likely the preferred host over other brachiopods. 4. Burrowing by Thalassinoides-producing organisms likely produced a time-averaged record of the community and obscured microfacies favoring Rhipidomella over other brachiopods, thus, biasing the sample toward a seemingly preferred Rhipidomella host and more frequent predation attempt survivor. One other factor should be noted in the encrusting and predatory interactions with Rhipidomella, and that is the potential epibiotic nature of Rhipidomella itself. It was a small, pedunculate brachiopod, and an obligate attacher to other media. Therefore, Rhipidomella specimens may be episkeletobionts themselves, secondarily tiering themselves higher into the water column on larger brachiopod shells themselves or attached to intraclasts. Modern, heavily encrusted Terebratalia and Laqueus brachiopods in mixed muddy and rocky and shelly environments of benthic Puget Sound frequently are attached to rocks and shells, and bear an epibiont community distinct from their rocky or shelly media (C. Schneider, personal observation, 2009). (5) Rhipidomella, therefore, may be preferentially encrusted by ctenostome bryozoans taking advantage of secondarily tiered individuals, and may attempt to avoid small biting predators through secondary tiering. The Effects of Shell Texture and Punctae In a study of modern brachiopods, encrustation frequency was recorded to be higher for taxa with pronounced radial ribs as compared to finer ribbed taxa; the lowest frequencies were attained by the smooth-shelled Bouchardia and the spinule-bearing Platidia (Rodland et al., 2004). Endoskeletobiosis of various types of brachiopod shell ornament is poorly understood; however, this study reveals a unique preference of endoskeletobionts within the Dundee Formation brachiopods. Studies of Devonian encrusting organisms across entire brachiopod assemblages indicates the preference of most encrusting organisms for ribbed hosts and the avoidance of smooth hosts (Hurst, 1974; Thayer, 1974; Anderson and Megivern, 1982; Schneider and Webb, 2004; Zhan and Vinn, 2007; Schneider, 2009b). The more coarsely ribbed, costate brachiopods Mucrospirifer and rhynchonellids do not show any evidence of endoskeletobiosis. Additionally, the two finely ribbed taxa, Strophodonta and the atrypides, also were lacking in obvious shell borings. Only the small, radially and finely ribbed, punctate Rhipidomella retained evidence of interactions with fouling organisms. Brachiopod punctae have been suggested to be antifouling and antipredatory defenses. Caeca within the punctae apparently produce chemical deterrents that deter borers (Clarkson, 1986). Spicules supporting the caecae may also have an antipredatory function (Peck, 1993). Punctate brachiopods are less frequently encrusted than those that were impunctate (Curry, 1983; Thayer, 1986). The impunctate spiriferide Meristella bore higher rates of encrustation compared to the punctate terebratulides (except the orthide Tropidoleptus) collected from the Kashong Shale of the Middle Devonian Hamilton Group of New York State (Bordeaux and Brett, 1990). Modern punctate brachiopods, however, can be encrusted heavily by episkeletobionts during life (C. Schneider, personal observation, 2008). Orthide brachiopods can be punctate or impunctate, depending on the taxon (Rowell and Grant, 1987). Rhipidomella was a punctate brachiopod that experienced predation. It is certain that punctae did not entirely deter predators from attacking Rhipidomella in Dundee Formation environments. Given that these durophagous scars represent failed predation attempts, however, it is difficult to determine whether predation frequency on Rhipidomella was

206

PALAIOS

BOSE ET AL.

higher or lower than that of other brachiopods. Predators may have successfully attacked impunctate brachiopods, or may have been deterred from finishing a predation attempt on Rhipidomella because of antipredatory function by the punctae or caecae. Impunctate brachiopod taxa from the Dundee Formation lack evidence of endoskeletobionts. Although Rhipidomella was punctate, the brachiopod was a medium for boring organisms. Evidence provided above suggests the association between live hosts and endoskeletobionts, at least in some individual brachiopods. Thus, the Rhipidomella specimens herein are a rare example of preferential Paleozoic endobiosis of a punctate host. Paleoecological Interactions in Paleozoic Carbonate Regimes While there has been evidence of immense biotic activity in carbonate sedimentary regimes, studies undertaken in shaly units far outnumber those in limestones. Limestones likely are less studied for encrusting organisms because of the difficulty of removing host specimens from the matrix with the encrusters intact. Preparing limestone samples is much more time consuming than the collection of brachiopods from shales popularly studied, such as those of the Silica Formation of Michigan and Ohio and the Lime Creek Group of Iowa. The present study, although affected by taphonomic alteration of the brachiopods, suggests that Paleozoic carbonate regimes have untapped potential to contribute to our understanding of paleoecological interactions. CONCLUSIONS The detailed study of brachiopods from units 10 through 12 of the Middle Devonian Dundee Formation in the Whitehouse Quarry reveals that larger (3.0–4.0 cm) brachiopod genera that were typically encrusted or preyed upon elsewhere bore no episkeletobionts, endoskeletobionts, or predation traces, but valves of the somewhat smaller (2.0 6 0.1 cm), punctate brachiopod genus Rhipidomella bore traces left by endoskeletobionts and predators. These Rhipidomella shells preserve trace fossils of branching grooves made by ctenostome bryozoans (Fig. 3), U-shaped straight borings (Fig. 4F) made by worms (most likely by a polychaete), and large boreholes (Figs. 4F–G) made by enigmatic parasites, in addition to predatory repair scars (Figs. 4A–E) from crushing attempts. Thus, our study suggests selective bioerosion by endoskeletobionts and predation on Rhipidomella over other contemporaneous brachiopods, although time averaging of assemblages cannot be completely ruled out. Several Rhipidomella specimens bore evidence of parasitic relationships with ctenostome bryozoans, in that bryozoan larvae bored into areas along the commissure, perhaps taking advantage of inhalant and exhalant currents, while brachiopods attempted to repair the damage caused by the colonies. Later growth of these colonies over the brachiopod shell moved the colony away from the commissure, suggesting that mature ctenostome bryozoan colonies did not require assistance from host-generated currents. Two instances of unsuccessful, durophagous predation attempts on Rhipidomella were clearly evident from the repair scars. The sampled beds in the Dundee Formation include both softgrounds of muddy matrix and shelly beds formed by storm events, later burrowed with Thalassinoides. Moderate transport of shell material coupled with their local shifting in the seafloor could be the likely cause for abrasion of some Rhipidomella specimens, potentially resulting in loss of episkeletobionts. Burrowing further increases time averaging and abrasion and, therefore, episkeletobiont loss. It is still noteworthy that endobiosis was preserved on Rhipidomella shells regardless of their taphonomic status. Other brachiopod genera bore no evidence of endobiosis or predation. This differential effect in endobiosis could be due to (1) preference for Rhipidomella by endoskeletobionts, (2) time averaging of short-lived communities, or (3) potential effects of

secondary tiering or cryptic behavior by attached Rhipidomella on encrusting organisms. Predation differences between larger brachiopods and Rhipidomella may have resulted from (1) smaller predators being excluded from the large size refugia of other brachiopods, (2) more successful destructive predation on larger brachiopods, (3) the effects of cryptic behavior or secondary tiering on predation attempts, or (4) punctae causing predation deterrence. Paleoecological interactions are more frequently documented from Paleozoic siliciclastic units than from carbonate units. The present study of the Dundee Formation documents a relatively rare incidence of endobiosis from a carbonate sedimentological regime, which significantly contributes to our understanding of endoskeletobiont occurrences in Paleozoic limestones. We anticipate future work to compare epi- and endobiosis in brachiopods from various settings in order to pursue the question of why such interactions appear to be more common in siliciclastic versus carbonate environments. ACKNOWLEDGMENTS We express our appreciation to the Village of Whitehouse, as well as the Toledo Metroparks, for granting access to the Whitehouse Quarry. We thank Dr. Don C. Steinker and Dr. Richard Hoare for their support and suggestions. A special thanks to Dr. Lindsey Leighton for his helpful comments on scarred specimens and literature recommendations. Thanks to Christopher Wright for providing stratigraphic information about the Dundee Formation, and his assistance in fieldwork. RB thanks Juergen Schieber, Stephaney Puchalski and Monalisa Mishra for help with SEM and Claudia Johnson for help with housing these field collections in the Indiana University Paleontology repository. Finally, a special thanks to D. L. Rodland, the two anonymous reviewers, and Editor Stephen T. Hasiotis for their careful reviews. Last but not least, we express our deep gratitude to Sheila J. Roberts, James E. Evans, and other faculty and staff members of BGSU for the opportunity to carry out this research. REFERENCES AGER, D.V., 1961, The epifauna of a Devonian spiriferid: The Quarterly Journal of the Geological Society of London, v. 117, p. 1–10. AHMED, M., and LEIGHTON, L.R., 2009, Anti-predatory spines on brachiopods? Examining little brachiopods with dents: Geological Society of America Abstracts with Programs, v. 41, No. 7, p. 389. ALEXANDER, R.R., 1981, Predation scars preserved in Chesterian brachiopods: Probable culprits and evolutionary consequences for the articulates: Journal of Paleontology, v. 55, p. 192–203. ALEXANDER, R.R., 1986, Resistance to and repair of shell breakage induced by durophages in Late Ordovician brachiopods: Journal of Paleontology, v. 60, p. 273–285. ALEXANDER, R.R., and DIETL, G., 2000, Frequency of shell repairs in common clams from New Jersey: Journal of Shellfish Research, v. 19(1), p. 658–659. ALVAREZ, F., and TAYLOR, P.D., 1987, Bryozoan ecology and interactions in the Devonian of Spain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 61, p. 17–31. ANDERSON, W.I., and MEGIVERN, K.D., 1982, Epizoans from the Cerro Gordo Member of the Lime Creek Formation (Upper Devonian), Rockford, Iowa: Proceedings of the Iowa Academy of Science, v. 89, p. 71–80. ARUA, H., and HOQUE, M., 1989, Predatory gastropod boreholes in an Eocene molluscan assemblage from Nigeria: Lethaia, v. 22, p. 49–59. AUSICH, W.I., and GURROLA, R.A., 1979, Two boring organisms in a Lower Mississippian community of southern Indiana: Journal of Paleontology, v. 53, p. 335–344. BASSETT, C.F., 1935, Stratigraphy and paleontology of the Dundee Limestone of southeastern Michigan: Geological Society of America Bulletin, v. 46, p. 425–462. BEST, M.M.R., and KIDWELL, S.M., 2000, Bivalve taphonomy in tropical mixed siliciclastic-carbonate settings; II, Effect of bivalve life habits and shell types: Paleobiology, v. 26, p.103–115. BORDEAUX, Y.L., and BRETT, C.E., 1990, Substrate specific associations of epibionts on middle Devonian Brachiopods: Implications for Paleoecology: Historical Biology, v. 4, p. 203–220.

PALAIOS

ENDOSKELETOBIONTS ON RHIPIDOMELLA

BOTQUELEN, A., and MAYORAL, E., 2005, Early Devonian bioerosion in the Rade De Brest, Armorican Massif, France: Palaeontology, v. 48, p. 1057–1064. BOTTJER, D., and AUSICH, W., 1986, Phanerozoic development of tiering in soft substrata suspension-feeding communities: Paleobiology, v. 12, p. 400–420. BREZINSKI, D.K., 1984, Upper Mississippian epizoans and hosts from southwestern Pennsylvania: Proceedings of the Pennsylvania Academy of Science, v. 58, p. 223– 226. CAMERON, B., 1969, Paleozoic shell boring annelids and their trace fossils: Zoologist, v. 9, p. 689–703. CARRERA, M.G., 2000, Epizoan-sponge interactions in the Early Ordovician of the Argentine Precordillera: PALAIOS, v. 15, p. 261–272. CARRIKER, M.R., and YOCHELSON, E.L., 1968, Recent gastropod boreholes and Ordovician cylindrical borings: Contributions to Paleontology, Geological Survey Professional Paper 593B, p. B1–B26. CLARKSON, E.N.K., 1986, Invertebrate Palaeontology and Evolution, 2nd ed.: Allen and Unwin, London, 382 p. CURRY, G.B., 1983, Brachiopod caeca—A respiratory role?: Lethaia, v. 16, p. 311– 312. DALEY, A.C., 2008, Statistical analysis of mixed-motive shell borings in Ordovician, Silurian, and Devonian brachiopods from northern and eastern Canada: Canadian Journal of Earth Science, v. 45, p. 213–229. DYER, H.C., and ELLIOTT, D.K., 2003, Endoskeletozoan trace fossil from the Pennsylvanian Naco Formation at Kohl Ranch, central Arizona (abstract): Proceedings of the Arizona-Nevada Academy of Science, v. 38, p. 37. EHLERS, G.M., STUMM, E.C., and KESLING, R.V., 1951, Devonian rocks of southeastern Michigan and northwestern Ohio: Fieldguide for Geological Society of America, Detroit meeting: Edwards Brothers, Inc., Ann Arbor, Michigan, 40 p. EKDALE, A.A., BROMLEY, R.G., and PEMBERTON, S.G., 1984, Ichnology: Trace Fossils in Sedimentology and Stratigraphy: SEPM Short Course No. 15: Society for Sedimentary Geology, Tulsa, Oklahoma, 317 p. EMIG, C.C., 1990, Examples of post-mortality alteration in Recent brachiopod shells and (paleo)ecological consequences: Marine Biology, v. 104, p. 233–238. FERGUSON, L., 1962, Distortion of Crurithyris urei (Fleming) from the Visean Rocks of Fife, Scotland, by Compaction of the Containing Sediment: Journal of Paleontology, v. 36, p. 115–119. GIBSON, M.A., 1992, Some epibiont-host and epibiont-epibiont relationships from the Birdsong Shale Member of the Lower Devonian Ross Formation (west-central Tennessee, U.S.A.): Historical Biology, v. 6, p. 113–132. GLUCHOWSKI, E., 2005, Epizoans on upper Eifelian crinoid columnals from the Holy Cross Mountains, Poland: Acta Palaeontologica Polonica, v. 50, p. 315–328. GRADSTEIN, F.M., OGG, J.G., and SMITH, A.G., 2004, A Geologic Time Scale 2004: Cambridge University Press, Cambridge, UK, 589 p. HANSEN, M.C., 1999, The Geology of Ohio—The Devonian: Ohio Geology—A Quaterly publication of the Division of Geological Survey, v. 1, p. 1–8. HARLAND, T.L., and PICKERILL, R.F., 1987, Epizoic Schizocrania sp. from the Ordovician Trenton Group of Quebec, with comments on mode of life of conulariids: Journal of Paleontology, v. 61, p. 844–849. HOARE, R.D., and STELLER, D.L., 1967, A Devonian brachiopod with epifauna: The Ohio Journal of Science, v. 67, p. 291. HOFFMEISTER, A.P., KOWALEWSKI, M., BAMBACH, R.K., and BAUMILLER, T.K., 2003, Intense drilling in the Carboniferous brachiopod Cardiarina cordata Cooper, 1956: Lethaia, v. 36, p. 107–118. HURST, J.M., 1974, Selective epizoan encrustation of some Silurian brachiopods from Gotland: Palaeontology, v. 17, p. 423–429. IRMIS, R.B., and ELLIOTT, D.K., 2006, Taphonomy of a middle Pennsylvanian marine vertebrate assemblage and an actualistic model for marine abrasion of teeth: PALAIOS, v. 21, p. 466–479. KELLEY, P.H., and HANSEN, T.A., 2003, The fossil record of drilling predation on bivalves and gastropods, in Kelley, P.H., Kowalewski, M., and Hansen, T.A., eds., Predator-Prey Interactions in the Fossil record: Topics in Geobiology, Kluwer/ Plenum, New York, v. 20, p. 113–140. KERN, J.P., 1979, The ichnofossil Helicotaphrichnus commensalis in the Korytnica basin (Middle Miocene; Holycross Mountains, central Poland): Acta Geologica Polonica, v. 29, p. 239–242. KESLING, R.V., HOARE, R.D., and SPARKS, D.K., 1980, Epizoans of the Middle Devonian brachiopod Paraspirifer bownockeri: Their relationships to one another and to their host: Journal of Paleontology, v. 54, p. 1141–1154. KOWALEWSKI, M., FLESSA, K.W., and MARCOT, J.D., 1997, Predatory scars in the shells of a Recent lingulid brachiopod: Paleontological and ecological implications: Acta Palaeontologica Polonica, v. 42, p. 497–532. KOY, K., and YACOBUCCI, M.M., 2001, Terror in miniature: Trophic interactions in a shelly fauna of the Pennsylvanian of southern Ohio: Geological Society of America Abstracts with Programs, v. 33, No. 6, p. 14. LEIGHTON, L.R., 1999, Possible latitudinal predation gradient in middle Paleozoic oceans: Geology, v. 27, p. 47–50.

207

LEIGHTON, L.R., 2001, New directions in the paleoecology of Paleozoic brachiopods, in Carlson, S.J., and Sandy, M.R. eds., Brachiopods Ancient and Modern: The Paleontological Society Papers, v. 7, p. 185–205. LEIGHTON, L.R., 2002, Inferring predation intensity from the marine fossil record: Paleobiology, v. 28, p. 328–342. LEIGHTON, L.R., 2003, Predation on brachiopods, in Kelley, P.H., Kowalewski, M., and Hansen, T.A., eds., Predator-Prey Interactions in the Fossil record: Topics in Geobiology, v. 20, Kluwer/Plenum, New York, p. 215–237. LESCINSKY, H.L., 1995, The life orientation of concavo-convex brachiopods: Overturning the paradigm: Paleobiology, v. 21, p. 520–551. LESCINSKY, H.L., 1996, Don’t overlook the epibionts!: PALAIOS, v. 11, p. 495–496. LESCINSKY, H.L., 1997, Epibiont communities: Recruitment and competition on North American Carboniferous brachiopods: Journal of Paleontology, v. 71, p. 34–53. LESCINSKY, H.L., and VERMEIJ, G.J., 1995, Estimating ancient productivity; shellencrusting organisms as a paleo bio-assay: Geological Society of America Abstracts with Programs v. 27, p. 167. MCKINNEY, F.K., 1996, Encrusting organisms on co-occurring disarticulated valves of two marine bivalves: Comparison of living assemblages and skeletal residues: Paleobiology, v. 22, p. 543–567. MCKINNEY, F.K., and JACKSON, J.B.C., 1989. Bryozoan Evolution: University of Chicago Press, Chicago, 252 p. MORRIS, R.W., and FELTON, S.H., 1993, Symbiotic association of crinoids, platyceratid gastropods, and Cornulites in the Upper Ordovician (Cincinnatian) of the Cincinnati, Ohio region: PALAIOS, v. 8, p. 465–476. MORRIS, R.W., and FELTON, S.H., 2003, Paleoecologic associations and secondary tiering of Cornulites on crinoids and bivalves in the Upper Ordovician (Cincinnatian) of southwestern Ohio, southeastern Indiana, and northern Kentucky: PALAIOS, v. 18, p. 546–558. PARSONS-HUBBARD, K., 2005, Molluscan taphofacies in Recent carbonate reef/lagoon systems and their application to sub-fossil samples from reef cores: PALAIOS, v. 20, p. 175–191. PECK, L.S., 1993, The tissues of articulate brachiopods and their value to predators: Philosophical Transactions of the Royal Society of London: Biological Sciences, v. 339, p. 17–32. PETERS, S.E., 1995, Cornulitid occurrences on Paraspirifer bownockeri: Implications for Spiriferacea current-systems: Geological Society of America Abstracts with Programs, v. 27, p. 445. PETERS, S. E., and BORK, K.B., 1998, Secondary tiering on crinoids from the Waldron Shale (Silurian, Wenlockian) of Indiana: Journal of Paleontology, v. 72, p. 887. PICKERILL, R.K., 1976, Vermiforichnus borings in the Ordovician of central Wales: Geological Magazine, v. 113, p. 159–164. PITRAT, C.W., and ROGERS, F.S., 1978, Spinocyrtia and its epizoans in the Traverse Group (Devonian) of Michigan: Journal of Paleontology, v. 52, p. 1315–1324. POHOWSKY, R.A., 1978, The boring ctenostomate Bryozoa: Taxonomy and paleobiology based on cavities in calcareous substrata: Bulletin of American Paleontology, v. 73, p. 1–192. RICHARDS, R.P., 1972, Autecology of Richmondian brachiopods (Late Ordovician of Indiana and Ohio): Journal of Paleontology, v. 46, p. 386–405. RICHARDS, R.P., and SHABICA, C.W., 1969, Cylindrical living burrows in Ordovician dalmanellid brachiopod beds: Journal of Paleontology, v. 43, p. 838–841. RODLAND, D.L., KOWALEWSKI, M., CARROLL, M., and SIMOES, M.G., 2004, Colonization of a ‘Lost World’: Encrustation patterns in modern subtropical brachiopod assemblages: PALAIOS, v. 19, p. 381–395. RODLAND, D.L., KOWALEWSKI, M., CARROLL, M., and SIMOES, M.G., 2006, The temporal resolution of epibiont assemblages: Are they ecological snapshots or overexposures?: The Journal of Geology, v. 114, p. 313–324. RODRIGUES, S.C., SIMOES, M.G., KOWALEWSKI, M., PETTI, M.A.V., NONATO, E.F., MARTINEZ, S., and RIO, C.D.J., 2008, Biotic interaction between spionid polychaetes and bouchardiid brachiopods: Paleoecological, taphonomic and evolutionary implications: Acta Palaeontologica Polonica, v. 53, p. 657–668. RODRIGUEZ, J., and GUTSCHICK, R.C., 1975, Epibiontic relationships on a Late Devonian algal bank: Journal of Paleontology, v. 49, p. 1112–1120. ROWELL, A.J., and GRANT, R.E., 1987, Phylum Brachiopoda, in Boardman, R.C., Cheetham, A.H., and Rowell, A.J., eds., Fossil Invertebrates: Blackwell Science, Cambridge, Massachusetts, p. 463–488. RUDWICK, M.J.S., 1962, Filter-feeding mechanisms in some brachiopods from New Zealand: Journal of the Linnean Society of London, Zoology, v. 44, p. 592–615. RUDWICK, M.J.S., 1970, Living and Fossil Brachiopods: Hutchinson University Library, London, 199 p. SANDY, M.R., 1996, Oldest record of peduncular attachment of brachiopods to crinoid stems, Upper Ordovician, Ohio, USA (Brachiopods; Atrypida: Echinodermata; Crinoidea): Journal of Paleontology, v. 70, p. 532–534. SARYCHEVA, T.G., 1949, The study of damage to Carboniferous productoid shells: Trudy Palaeontological Institute, v. 18, p. 280–292.

208

PALAIOS

BOSE ET AL.

SCHNEIDER, C.L., 2003. Hitchhiking on Pennsylvanian echinoids: Epibionts on Archaeocidaris: PALAIOS, v. 18, 435–444. SCHNEIDER, C.L., 2009a, Epibionts on Late Carboniferous through Early Permian echinoid spines from Texas, USA, Echinoderms 2006: Proceedings of the 12th International Echinoderm Conference, 7–11 August 2006, Durham, New Hampshire, U.S.A.: CRC Press, Boca Raton, 679 p. SCHNEIDER, C.L., 2009b, Substrate preferences of Middle and Late Devonian Hederella from the midcontinent USA, in Hageman, S.J., Key, M.M. Jr., and Winston, J.E., eds., Bryozoan Studies 2007, Proceedings of the 14th IBA Conference, Boone, North Carolina, July 1–8, 2007: Virginia Museum of Natural History Special Publication, Martinsville, Virginia, v. 15, p. 295–300. SCHNEIDER, C.L., and LEIGHTON, L.R., 2007, The influence of spiriferide microornament on Devonian epizoans: Geological Society of America Abstracts with Programs, v. 39, p. 531. SCHNEIDER, C.L., and WEBB, A., 2004, Where have all the encrusters gone? Encrusting organisms on Devonian versus Mississippian brachiopods: Geological Society of America Abstracts with Programs, v. 36, p. 111. SIGNOR III, P.W., and BRETT, C.E., 1984, The Mid-Paleozoic Precursor to the Mesozoic Marine Revolution: Paleobiology, v. 10, p. 229–245. SMITH, S.A., THAYER, C.W., and BRETT, C.E., 1985, Predation in the Paleozoic: Gastropod-like drillholes in Devonian brachiopods: Science, v. 230, p. 1033–1035. SPARKS, D.K., HOARE, R.D., and KESLING, R.V., 1980, Epizoans on the brachiopod Paraspirifer bowneckeri (Stewart) from the Middle Devonian of Ohio: University of Michigan Museum of Paleontology, Papers on Paleontology, v. 23, p. 1–50. SPARLING, D.R., 1988, Middle Devonian stratigraphy and conodont biostratigraphy, north-central Ohio: Ohio Journal of Science, v. 88, p. 2–18. SPJELDNAES, N., 1984, Epifauna as a tool in autecological analysis of Silurian brachiopods: Special Papers in Paleontology, v. 32, p. 225–235. STAUFFER, C.R., 1909, The Middle Devonian of Ohio: Ohio Geological Survey Bulletin 10, 204 p. SUMRALL, C.D., 1992, Spiraclavus nacoensis, a new species of clavate agelacrinitid edrioasteroid from central Arizona: Journal of Paleontology, v. 66, p. 90–98. SUMRALL, C.D., 2000, The biological implications of an edrioasteroid attached to a pleurocystitid rhombiferan: Journal of Paleontology, v. 74, p. 67–71. TAPANILA, L., 2004, The earliest Helicosalpinx from Canada and the global expansion of commensalism in Late Ordovician sarcinulid corals (Tabulata): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 215, p. 99–110. TAPANILA, L., 2005, Palaeoecology and diversity of endosymbionts in Palaeozoic marine invertebrates: Trace fossil evidence: Lethaia, v. 38, p. 89–99.

TAPANILA, L., 2006, Devonian Entobia borings from Nevada, with a revision of Topsentopsis: Journal of Paleontology, v. 80, p. 760–767. TAYLOR, P.D., and WILSON, M.A., 2002, A new terminology for marine organisms inhabiting hard substrates: PALAIOS, v. 17, p. 522–525. TAYLOR, P.D., and WILSON, M.A., 2003, Palaeoecology and evolution of marine hard substrate communities: Earth Science Reviews, v. 62, p. 1–103. THAYER, C.W., 1974, Substrate specificity of Devonian epizoa: Journal of Paleontology, v. 48, p. 881–894. THAYER, C.W., 1986, Respiration and the function of brachiopod punctae: Lethaia, v. 19, p. 23–31. THAYER, C.W., and ALLMON, R. 1990, Unpalatable thecideid brachiopods from Palau: ecological and evolutionary implications, in MacKinnon, D.I., Lee, D.E., and Campbell, J.D., eds., Brachiopods through Time, Balkema, Rotterdam, p. 253– 260. VERMEIJ, G.J., 1982, Gastropod shell form, breakage, and repair in relation to predation by the crab Calappa: Malacologia, v. 23, p. 1–12. WALKER, K.R., and ALBERSTADT, L.P., 1975, Ecological succession as an aspect of structure in fossil communities: Paleobiology, v. 1, p. 238–257. WATKINS, R., 1981, Epizoan ecology of the type Ludlow Series (Upper Silurian), England: Journal of Paleontology, v. 55, p. 29–32. WOOD, T.S. and MARSH, T.G., 1996, The sinking floatoblasts of Lophopodella carteri (Bryozoa: Phylactolaemata), in Gordon, D.P., Smith, A.M., and Grant-Mackie, J.A., eds., Bryozoans in Space and Time: Wellington: National Institute of Water & Atmospheric Research, Wellington, New Zealand, p. 383–389. WRIGHT, C.E., 2006, Paleoecological and paleoenvironmental analysis of the Middle Devonian Dundee Formation at Whitehouse, Lucas County, Ohio: Unpublished M.S. Thesis, Bowling Green State University, Bowling Green, Ohio, 195 p. ZAPALSKI, M.K., 2004, Parasitism on favositids (Tabulata): Palaeontology Newsletter, v. 57, p. 194. ZAPALSKI, M.K., PINTE, E., and MISTIAEN, B., 2008, Late Famennian ?Chaetosalpinx in Yavorskia (Tabulata): The youngest record of tabulate endobionts: Acta Geologica Polonica, v. 58, p. 321–324. ZHAN, R., and VINN, O., 2007, Cornulitid epibionts on brachiopod shells from the Late Ordovician (middle Ashgill) of East China: Estonian Journal of Earth Sciences v. 56,2, p. 101–108.

ACCEPTED DECEMBER 2, 2009