Opportunistic exploitation of dinosaur dung: fossil snails in coprolites from the Upper Cretaceous Two Medicine Formation of Montana
2008; Wiley; Volume: 42; Issue: 2 Linguagem: Inglês
10.1111/j.1502-3931.2008.00131.x
ISSN1502-3931
AutoresKaren Chin, Joseph H. Hartman, Barry Roth,
Tópico(s)Evolution and Paleontology Studies
ResumoLethaiaVolume 42, Issue 2 p. 185-198 Free Access Opportunistic exploitation of dinosaur dung: fossil snails in coprolites from the Upper Cretaceous Two Medicine Formation of Montana KAREN CHIN, KAREN CHINSearch for more papers by this authorJOSEPH H. HARTMAN, JOSEPH H. HARTMANSearch for more papers by this authorBARRY ROTH, BARRY ROTHSearch for more papers by this author KAREN CHIN, KAREN CHINSearch for more papers by this authorJOSEPH H. HARTMAN, JOSEPH H. HARTMANSearch for more papers by this authorBARRY ROTH, BARRY ROTHSearch for more papers by this author First published: 16 April 2009 https://doi.org/10.1111/j.1502-3931.2008.00131.xCitations: 20 Karen Chin[karen.chin@colorado.edu], Geological Sciences/Museum of Natural History, University of Colorado, UCB 265, Boulder, CO 80309, USA; Joseph H. Hartman [joseph_hartman@und.nodak.edu], Department of Geology and Geological Engineering, University of North Dakota, 81 Cornell Street, Stop 8358, Grand Forks, ND 58202, USA; and Barry Roth [barry_roth@yahoo.com], Museum of Paleontology, University of California, Berkeley, CA 94720, USA; manuscript received on 12/12/07; manuscript accepted on 15/05/08. AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract Multiple associations of fossil snails with dinosaur coprolites demonstrate that snails and dinosaurs not only shared ancient habitats but were trophically linked via dinosaur dung. Over 130 fossil snails representing at least seven different taxa have been found on or within herbivorous dinosaur coprolites from the Upper Cretaceous Two Medicine Formation of Montana. The terrestrial snail Megomphix is the most common taxon, but three other terrestrial taxa (Prograngerella, Hendersonia and Polygyrella) and three aquatic snails (Lioplacodes, ?Viviparus and a physid) also occur in coprolites. At least 46% of the shells in the faeces are whole or nearly so, indicating that most (if not all) of the snails were not ingested by dinosaurs, but were post-depositional visitors to the dung pats. The sizeable, moist and microbially enriched dinosaur faeces would have provided both food and roosting sites for the ancient snails, and the large number of snail–coprolite associations reflect recurring, opportunistic exploitation of dung. The terrestrial taxa in the coprolites suggest that this Late Cretaceous locality included sufficiently moist detrital or vegetative cover for snails when dinosaur dung was not present. The aquatic snails probably entered the faeces during flood events. Dinosaur dung would have provided an abundant but patchy influx of resources that was probably seasonally available in the ancient environment. The incidence of fossil snails in terrestrial Cretaceous deposits indicates that they often shared Cretaceous ecosystems with dinosaurs. Even so, it is not surprising that substantive interactions between these organisms have not been documented. Terrestrial snails are generally tied to crepuscular or nocturnal regimes (Cain 1983), and the disparate sizes and lifestyles of snails and dinosaurs suggest that they would typically have had only limited, incidental contact. Now, however, numerous snails in herbivorous dinosaur coprolites from the Upper Cretaceous Two Medicine Formation provide rare fossil evidence for recurring associations between incongruent organisms that would seem to have had little cause for interaction. No ecological interactions between dinosaurs and molluscs have been previously reported. The presence of land snails in ancient sediments can also provide useful palaeoenvironmental information. Freshwater aquatic taxa reflect the proximity of riparian and/or lacustrine habitats, and the wet and permeable integument of terrestrial snails (Cook 2001) makes them very sensitive to environmental conditions. As such, palaeontologists have utilized fossil snails to help interpret ancient terrestrial environments (e.g. Roth 1986, 1991; Roth & Megaw 1989; Varricchio 1995). This paper describes the fossil snails from this locality and analyses the nature of their relationship with the dinosaurian faecal producers. Geologic setting and methods The Two Medicine Formation is a 600-m-thick fluvial unit occurring in northwestern Montana. The formation is dominated by calcareous mudstones interspersed with channel sandstones (Lorenz & Gavin 1984), and represents a coastal plain on the western edge of the Cretaceous Western Interior Seaway (Rogers 1990). The snails in this study are from the Willow Creek Anticline locality, southwest of Choteau, Montana (Fig. 1). Sediments at this site have been dated at approximately 76.7 Ma (unpublished 40Ar/39Ar analysis; R.R. Rogers, personal communication, 2004). Palaeogeographic reconstructions and the occurrence of widespread caliche development suggest that the area had a temperate, semiarid palaeoclimate with monsoonal rainfall patterns (Lorenz & Gavin 1985). Numerous dinosaur fossils are known from this locality, including isolated skeletal discoveries, an extensive Maiasaura bone bed (Schmitt et al. 1998), eggshell and baby bones from dinosaur nest sites (e.g. Horner & Makela 1979; Hirsch & Quinn 1990; Horner 1999; Varricchio et al. 1999; Varricchio et al. 2002), and herbivorous dinosaur coprolites (Chin & Gill 1996; Hollocher et al. 2001; Chin, 2007). Figure 1Open in figure viewerPowerPoint Maps showing location of the Willow Creek Anticline locality in Montana and coprolite deposits at this site. The Willow Creek Anticline exposure of the Two Medicine Formation is unique in having multiple deposits of large coprolites produced by herbivorous dinosaurs. These irregular calcareous specimens (Fig. 2A) are distinguished by their fragmented wood content (Chin 2007) and distinctive backfilled dung beetle burrows (Chin & Gill 1996). The coprolites often break apart upon subaerial exposure and occur as clusters of weathered chunks in deflation lags of overbank sediments. Most clusters of coprolite pieces appear to represent individual faecal deposits, and 15 deposits have been mapped and are distinguished by 2-letter field designations (Chin 2007; Fig. 1). The coprolite deposits occur within a short stratigraphic interval, but it is not known whether they were deposited within one or multiple seasons. Figure 2Open in figure viewerPowerPoint Herbivorous dinosaur coprolites and fossil snail inclusions: A. Piece of coprolite largely composed of fragments of conifer wood (centimeter scale in photo; specimen MOR 771/BU-93-2). B. Two gastropods embedded in coprolite ground mass; coprolite MOR 1132/SS-93-2. Snail 2-2 (upper left) was subsequently prepared out of the ground mass and is shown in Figure 5A. Note that the coprolitic ground mass on the right side of the image shows partial preparation to expose snail 2-3 at the lower right; this snail was later removed and is shown in Figure 5D. C. Well-preserved Megomphix sp. (snail L4-1) in coprolite MOR 1619/LM-94-4. D. Poorly preserved snail showing acid dissolution after prolonged aerial exposure. Snail is probably Polygyrella sp. (snail 11-3 in coprolite MOR 1132/SS-93-11). E. Photomicrograph showing a Megomphix sp. shell (snail 13-2) embedded within a coprolite (MOR 1132/SS-93-13). Note that the shell is filled with drusy calcite spar. The dark masses in the coprolitic ground mass are pieces of conifer wood; the tiny linear flecks are disaggregated tracheids from the conifer wood. Although most of the coprolites are broken, three nearly-intact masses are 6 to 7 L in volume, indicating sizeable dinosaurian defecators. Of the 15 coprolite deposits examined, 11 are predominantly composed of fragmented conifer wood, while four contain scattered fragments of conifer wood along with plant tissues that cannot be identified by taxon. The high percentage of comminuted wood in the coprolites suggests that the faecal producers ingested rotting wood with its attendant microbial and detritivore fauna (Chin 2007). Surfaces and thin sections of coprolite blocks were examined for snail shell. Where possible, snails were identified by genus, and diameters were measured. Four snail shells were mechanically prepared out of three coprolite blocks by Charles Magovern of the Stone Company. Snails from coprolites were compared with specimens collected as isolated float from the Willow Creek Anticline sediments. Fossils are catalogued in the palaeontology collection of the Museum of the Rockies (MOR) in Bozeman, Montana. Fossil snails at the Willow Creek Anticline locality Fossil snails were found both within coprolites and in laterally equivalent sediments at the Willow Creek Anticline locality. Over 70 snails were recovered as float in overbank mudstone deposits. These specimens are often poorly preserved; some are distorted and fractured, and many are preserved as internal calcite moulds with absent or diagenetically altered shell. The isolated float specimens were collected by many different crews and were not individually mapped. More than 130 snails have also been observed on or within six of the 15 herbivorous dinosaur coprolite deposits (Fig. 2B–E). The number of specimens associated with different coprolite deposits is variable, ranging from zero to more than 65 (Fig. 3). Snails are usually evident as partial or whole shells embedded on the exposed surfaces of coprolite blocks (Fig. 2B–D), but thin sections and broken specimens also reveal shell fragments and at least five apparently intact shells in the interiors of coprolites (Fig. 2E). Many shells on small coprolite pieces may have actually occurred deep inside larger faecal masses before the coprolites broke apart during weathering. These serendipitous discoveries suggest that many more shells and shell fragments are concealed within the coprolites. Figure 3Open in figure viewerPowerPoint Pie chart showing relative numbers of snails in the six coprolite deposits in which snails were found. The 2-letter designations refer to the deposits shown in Figure 1. Inset histograms show the size distribution of Megomphix sp. within these deposits (only measurable snails are included in the histograms). Most of the fossil snail shells are filled with calcite spar (Fig. 2E). Some specimens are well-preserved and have good shell detail (Fig. 2C), whereas others are barely discernible (Fig. 2D). These differences in preservation often reflect extent of subaerial exposure; both the coprolitic ground mass and the fossil snails of highly weathered coprolites are extensively pitted. Of 132 snails in the coprolites, at least 30 (23%) appear to be whole. Another 62 (47%) are broken but are intact enough that they may have been originally complete before being damaged and/or partially dissolved during breakup and weathering of the coprolites (e.g. Fig. 7B). Nineteen (14%) appear to be fragments, and the state of completeness of 21 (16%) could not be determined. Whether originally entire or not, 60 (46%) of the snails are complete enough so that at least 75% of the original diameter could be measured. Figure 7Open in figure viewerPowerPoint Aquatic snail taxa associated with coprolites from the Willow Creek Anticline. A. Lioplacodes sp. (snail E1-7 from coprolite MOR 1614/ES-95-1). B. Viviparid (snail 1-15 in coprolite MOR 1132/SS-95-1). C. Probable physid (snail 19-8 from coprolite MOR 1132/SS-95-19). Description of material Approximately 60 (46%) of the snails on or within coprolites display enough distinctive characteristics to be assigned to various taxa with relative confidence. Analyses of the specimens reveal that at least seven different fossil snail taxa are associated with the Willow Creek Anticline coprolites, including four terrestrial and three aquatic taxa (Fig. 4). Five of these taxa were also found as float in the Willow Creek Anticline sediments, but two aquatic taxa were only found in coprolites. Table 1 shows the taxonomy of the snails recovered from the coprolites and their distribution among the coprolite deposits. Figure 4Open in figure viewerPowerPoint Pie chart showing numbers of recognizable snail taxa in coprolites at the Willow Creek Anticline locality. Table 1. Numbers and taxonomy of fossil snails recovered from different coprolite deposits at the Willow Creek Anticline locality. Taxon Coprolite deposits (indicated by 2-letter field designations, see Fig. 1) SS LM HN ES BU RP Totals Gastropoda Cuvier, 1797 Neritopsina Cox, 1960 Hendersoniidae Baker, 1925 Prograngerella sp., cf. P. sperata Russell, 1941 1 – – – – – 1 Hendersonia sp. – 1 1 1 – – 3 Architaenioglossa Haller, 1890 Viviparoidea Gray, 1847 Viviparidae Gray, 1847 ?Viviparus sp. 1 – – – – – 1 Pleuroceridae Fischer, 1885 Lioplacodes – – 1 1 – – 2 Pulmonata Ehrenberg, 1831 Stylommatophora Schmidt, 1855 Megomphicidae Baker, 1930 Megomphix sp. 25 13 7 3 – 48 Polygyrella sp. 2 1 1 – – – 4 Basommatophora Keferstein, 1864 Physidae Fitzinger, 1833 Physid, unidentified 1 – – – – – 1 Gastropoda, unidentified 36 15 8 8 3 2 72 Totals 66 30 18 13 3 2 132 Terrestrial snail taxa Megomphix sp. – The most common snails found both in coprolites and as isolated specimens are identified as a species of Megomphix. This taxon is recognized by a widely umbilicate, discoidal shell with a nearly flat spire of narrow, closely coiled whorls. The sutures are distinct. The shoulders of the whorls are crossed by low, rounded ribbing, convex in the direction of growth. The ribs fade out at the periphery but reappear towards the base. In cross-section, the whorls are ear-shaped, narrower towards the base (Fig. 2E). The umbilicus is about one third the diameter of the shell. No resorption of inner whorls is present, consistent with these snails being pulmonates rather than prosobranchs. There are no prior reports of Megomphix in the fossil record. Analyses of the two megomphicid snails prepared out of two different coprolites indicate that these specimens appear to be conspecific with the most common taxon recovered from the probably laterally equivalent, overbank deposits. Snail L6-1 (from coprolite MOR 1619/LM-94-96) consists of half of a shell of five whorls, broken vertically through the middle. It is 8.0 mm in maximum diameter and 2.9 mm in preserved height parallel to the coiling axis. The shell material is recrystallized, but surface detail is preserved quite well. Snail 2-2 (from coprolite MOR 1132/SS-93-2; 2, 5) consists of an internal mould (steinkern) and altered shell of approximately 4.2 whorls, 4.9 mm in maximum diameter and 2.3 mm in height. Traces of low ribbing on the shoulder are preserved, presumably tracking relief of the interior of the whorl concomitant with external costulae as in the previously described specimen. Figure 5Open in figure viewerPowerPoint Terrestrial snail taxa found within the herbivorous dinosaur coprolites. A. Megomphix sp. prepared out of a coprolite (MOR 1132/SS-93-2). Snail 2-2 is the same one shown in the upper left corner of Figure 2B. B. Polygyrella sp. (snail 3-2 in coprolite MOR 1132/SS-93-3). C. Hendersonia sp. (snail E3-1) prepared out of coprolite MOR 1614/ES-95-3. D. Prograngerella sp. (snail 2-3) prepared out of coprolite MOR 1132/SS-93-2. This snail is shown in situ in Figure 2B. Forty-eight of the snails in the coprolites (~37%) can be reasonably assigned to this species. Megomphix shells in the coprolites range from < 2 to 16 mm in diameter (3, 6), with a mean of 7 mm. The three tiniest specimens (≤ 2 mm) were difficult to see, suggesting that many more might be present but undetected. Figure 6Open in figure viewerPowerPoint Histogram showing size distribution of Megomphix sp. in coprolites. Only snails with measurable diameters are included. Inset photos show one of the smallest specimens (left: snail 4-4 in coprolite MOR 1132/SS-93-4) and the largest (right: snail H3-1 in coprolite MOR 1618/HN-94-3). Polygyrella sp. – Four specimens of a much less common megomphicid taxon were found in three coprolite deposits and in laterally equivalent overbank deposits. The specimens range from 3–5 mm in diameter and are assigned to Polygyrella on the basis of their numerous whorls and planispiral coiling (Fig. 5B). Hendersonia sp. – Specimen E3-1 (extracted from coprolite MOR 1614/ES-95-3; Fig. 5C) is a neritopsine prosobranch assignable to the genus Hendersonia. This and similar specimens from laterally equivalent overbank deposits lack internal whorl partitions and have sculpture of regular ribs with linear interspaces crossing the body whorl, as in the extant Hendersonia occulta (Say, 1831). Its low-trochoid shape, semilunate whorl cross-section, and imperforate base (the latter characters evident in snail L9-4 from coprolite MOR 1619/LM-94-9) are consistent with H. occulta and other species of the genus. The fossil record of Hendersonia includes Pleistocene occurrences of H. occulta in the USA Midwest; Hendersonia stillmani, Roth & Emberton, 1994, from the Deep River Formation, Montana (Arikareean); and 'H.'valida Pierce (in Pierce & Rasmussen 1989) from the Flint Creek Beds, Montana (early Barstovian). Assignments of other fossil neritopsine snails to Hendersonia are incorrect (Henderson 1935; Pierce & Rasmussen 1989). Prograngerella sp. – Snail 2-3 (removed from coprolite MOR 1132/SS-93-2; 2, 5) is tentatively identified as a species of Prograngerella. It is a pill-shaped shell 4.8 mm in maximum diameter, missing the spire, the outer lip, and part of the basal lip; what remains of the basal lip is thick and reflected. The aperture appears to have been slit-like and transverse, and there is no indication of apertural barriers. The base is imperforate, convex in the middle as though having an umbilical callus; recrystallization of the shell material has obscured details such as the borders of a possible callus. Fragments of preserved external sculpture show crowded, regular ribs with linear interspaces crossing the body whorl, as in the extant H. occulta from eastern North America. Internal whorl partitions are apparently absent, consistent with this specimen being a neritopsine prosobranch, in which the internal whorls of the spire are absorbed during growth (Solem 1983). Prograngerella was originally assigned to the family Grangerellidae Russell, 1931, and considered to be a pulmonate, but Roth (1992) showed that Grangerellidae is a junior synonym of the neritopsine taxon Hendersoniidae. Prograngerella includes one species, P. sperata, from the Campanian Oldman Formation in Alberta; the holotype is 4.9 mm in diameter (Russell 1941). Apertural barriers mentioned in the description of P. sperata are on a part of the shell that is missing in the present specimen. A firm identification is therefore not possible, but the preserved characters do not rule out assignment to P. sperata. Aquatic snail taxa Lioplacodes sp. – Parts of two high-spired snails are evident on two different coprolites (Fig. 7A). Although these specimens are incomplete, a more intact representative from laterally equivalent sediments has diagnostic features. This dextral snail (MOR IV 152) is 18.2 mm in height for parts of the last 3.7 whorls. The apical whorls are broken, but a reasonable estimate of the spire angle is 34.7°. The shell lacks surface detail except for a narrow vertical band on the penultimate whorl and just behind the top of the aperture; these surfaces suggest strong, straight growth lines and possibly revolving striae, the latter at least abapical to the suture. Sutures are only slightly impressed. The aperture appears relatively elongate ovate, but the outer edge is incomplete. The nearly complete inner lip indicates a closed umbilicus. The preserved features of this snail conform relatively well to the extinct genus Lioplacodes, and the specimen is similar in most respects to Lioplacodes praecursa Dyer, 1930. Species of Lioplacodes typically preserve nearly vertical growth lines with a slight sinusoidal pattern, but preservation of the Willow Creek Anticline specimen is too poor to confirm the true pattern and degree of growth line type. L. praecursa is known from both the Oldman and the Foremost Formations of the Campanian Belly River Group, but its morphological variability has not been described. Historic identifications of Campeloma (?Lioplax) producta (White, 1883) from the Two Medicine Formation may be referable to Lioplacodes (possibly, L. praecursa or the taxon from the Willow Creek Anticline locality) because C. producta is now referable to Lioplacodes nebrascensis producta (Dyer, 1930). A few incomplete specimens from the upper part of the Two Medicine Formation at Jack's Birthday site are also attributed to Lioplacodes (Varricchio, 1995). ?Viviparus sp. – The dextral snail 1-15 (on coprolite MOR 1132/SS-93-1; Fig. 7B) has a relatively high-spired and multiwhorled turbinate shell that is 6.3 mm in height for five whorls. Although the specimen lacks apical whorls and apertural features and has suffered plastic deformation, it appears to be a viviparid caenogastropod. It is difficult to identify Viviparus taxa to species using shell only, but this specimen is similar in whorl conformation to Viviparus prudentius White, 1878 (see Tozer 1956; upper Campanian through Maastrichtian). Whorls are well-rounded, with sutures deeply impressed. Although small areas of the specimen appear to preserve an external shell surface, no sculpture or growth lines are apparent. Historically, more trochiform snails (such as Viviparus) than snails of other shapes have been found in the Two Medicine Formation (data from Reeside 1916). Physid taxon. – One tiny Physa-like specimen (snail 19-8) was found on coprolite MOR 1132/SS-95-19 (Fig. 7C). This sinistral snail is 3.5 mm in oblique length, and the slightly laterally compressed steinkern preserves about 3.5 whorls. The apex is rounded, with the first whorl slightly raised. As preserved, the sutures are only moderately impressed; the spire is relatively broad conic at an angle of about 66.3°. The snail's placement in the coprolite prohibits more detailed comments, because the aperture is poorly defined. An apical angle suggests a comparison to the apical whorls of Physa subelongata Meek & Hayden, 1856, which is known from the Judith River Formation. Other sinistral specimens from the Two Medicine Formation have been assigned to Physa, including specimens from several localities in Glacier County (Reeside 1916; Varricchio 1995). Discussion The snail assemblage at the Willow Creek Anticline locality is intriguing because so many specimens are associated with herbivorous dinosaur coprolites. The occurrence of over 130 snails on or within six of 15 (40%) different coprolite deposits reveals recurring, if indirect interactions between gastropods and the dinosaurian defecators. The actual number of snails that were associated with dinosaur faeces is undoubtedly higher than that documented because shells can only be observed on fortuitously exposed coprolite surfaces that are not overly weathered. Approximately 46% of the snails can be reasonably classified by taxon, but over half of the specimens are too incomplete or diagenetically altered to identify. Because meteoric precipitation is undersaturated in calcium carbonate (Tucker 2001), dissolution-induced pitting of the calcareous coprolites and snail shells probably occurred after prolonged subaerial exposure. Some snail taxa may have been more susceptible to such diagenetic alteration, making it difficult to determine the original proportions of different taxa in the coprolites. Nevertheless, Megomphix appears to have been the most common snails in both dinosaur faeces and sediments in the ancient Willow Creek Anticline ecosystem. The distribution of Megomphix sizes in the coprolites may be representative of natural populations, or might reflect age-related movement patterns; for example, juveniles of the extant land snail Helicella virgata generally travel farther than the adults (Pomeroy 1968). Small numbers of three other terrestrial taxa (Polygyrella, Hendersonia and Prograngerella) occurred along with Megomphix. Surprisingly, however, three different aquatic snail taxa were also found in the coprolite deposits. The prevalence of the terrestrial snails in the coprolites suggests that the faeces were initially deposited subaerially and were subsequently flooded. This sequence of events is consistent with the presence of back-filled dung beetle burrows in all but two of the faecal deposits (SS and RM; Chin 2007) and the occurrence of the coprolites in floodplain sediments. Eaten or eating? Two scenarios can explain the presence of land snails in the dinosaur faeces: (1) the dinosaurian faecal producers ingested the snails, or (2) the snails visited fresh dung pats. The observation of some shell fragments inside the coprolites suggests that dinosaurs inadvertently or intentionally ingested snails that were feeding on or sheltering within dinosaur browse. This is a logical interpretation because the coprolite contents suggest that the faecal producers fed on rotting wood and other plant tissues (Chin 2007), and snails are known to frequent both vegetation and rotting logs (Smith 1943; Pearce & Örstan 2006). On the other hand, most of the shells do not appear to have passed through a digestive tract. A large proportion (at least ~46% and more likely as much as 70%) of snails in the coprolites are mostly whole. Furthermore, the coprolite producers were probably Maiasaura hadrosaurs (Chin 2007) which bore densely packed dental batteries with efficient grinding surfaces. As such, most ingested snails would have very likely been crushed during mastication and digestion. Thus, the high percentage of intact shells in the coprolites suggests that most came from snails that occupied the dung pats after deposition. Although the evidence indicates that most snails in the faeces were not ingested, it is nonetheless possible that some individuals were indeed eaten by dinosaurs feeding in snail habitats. Isolated shell fragments in the coprolites might thus have been broken either during dinosaur ingestion, or through trampling of live snails within dung pats. Herbivore dung deposits may attract gastropods for several reasons. Land snails are susceptible to evaporative water loss through the skin, and actively seek sites that provide protection from desiccation and temperature extremes, as well as proximity to food (Burch & Pearce 1990; Cook 2001). Herbivore faeces provide a moist environment and also present a nutritious source of food for snails. Terrestrial snails tend to have generalized diets (Cain 1983; Speiser 2001). This feeding strategy helps conserve energy by reducing the necessity for excessive travel; mucus-lubricated locomotion is energetically expensive (Speiser 2001). The repertoire of terrestrial snail food selections includes rotting vegetative detritus and fungi (e.g. Graham 1955; Chatfield 1976; Speiser 2001). Such dietary choices reflect the fact that most terrestrial gastropods appear to be microphagous (Speiser 2001). Faeces provide concentrated sources of microorganisms, so it is not surprising that some extant gastropods have been observed to feed on dung (Cain 1983; Harper 1988; Speiser 2001; Garvon & Bird 2005). In the Late Cretaceous, large, moist dinosaur faecal masses would have presented humid, food-rich microenvironments for Two Medicine Formation snails, and the numerous shells on exterior surfaces of the coprolites reflect the appeal of these dung pats. Whole shells in the interiors of faecal masses may reflect ingress through crevices or active burrowing within the dung; a number of extant land snails are known to burrow during adverse conditions (e.g. Livshits 1985) or to excavate nests (e.g. Gugler 1963; Baur 1994). Recent Megomphix californicus Smith, 1960, are sometimes found deep within rotting coniferous deadfalls (B. Roth, personal observation). The dinosaur dung at the ancient Willow Creek Anticline locality may have also provided good sites for land snails to mate and lay eggs. This possibility is supported by the discovery of two tiny (1.5 and 1.7 mm in diameter) Megomphix in the coprolites. After flooding, the submerged faeces probably also provided an atypical but nutritious substrate for aquatic snails. Pleurocerids (Dillon 2000), physids (Brown 1991) and viviparids (Allison 1942) can be detritivorous or microphagous, and extant Campeloma viviparids have been successfully baited with terrestrial vertebrate dung (Allison 1942). Alternatively, empty shells of the aquatic species might have been deposited as drift, but this seems less likely because other drift materials were not observed on the coprolites. Nature of the snail–dinosaur relationship The snail–coprolite associations suggest that patchily distributed dinosaur faeces provided a windfall for the Willow Creek Anticline gastropods in terms of food and roosting sites. Numerous back-filled burrow traces in the coprolites indicate that dung beetles exploited the same dinosaur faeces (Chin & Gill 1996), but there are differences between these types of dung utilization. The dung beetles at the ancient Willow Creek Anticline locality were probably similar to extant representatives in being o
Referência(s)