A New Eurypterid (Chelicerata) from the Upper Ordovician of Manitoulin Island, Ontario, Canada

J. Paleont., 79(6), 2005, pp. 1166±1174 Copyright ᭧ 2005, The Paleontological Society 0022-3360/05/0079-1166$03.00

A NEW EURYPTERID (CHELICERATA) FROM THE UPPER ORDOVICIAN OF MANITOULIN ISLAND, ONTARIO, CANADA

CHRISTOPHER A. STOTT,1 O. ERIK TETLIE,2 SIMON J. BRADDY,3 GODFREY S. NOWLAN,4 PAUL M. GLASSER,5 AND MATTHEW G. DEVEREUX6 1Department of Earth Sciences, The University of Western Ontario, London, Ontario N6A 5B7, Canada Ͻ[email protected]Ͼ, 2Department of Geology and Geophysics, Yale University, PO Box 208109, New Haven, Connecticut 06520-8109, Ͻ[email protected]Ͼ, 3Department of Earth Sciences, University of Bristol, Wills Memorial Building, Queen's Road, Bristol BS8 1RJ, United Kingdom, 4Geological Survey of Canada, 3303-33rd Street N.W., Calgary, Alberta T2L 2A7, Canada, 5School of Dentistry, Faculty of Medicine & Dentistry, The University of Western Ontario, London, Ontario N6A 5C1, Canada, and 6Faculty of Science, Of®ce of the Dean, The University of Western Ontario, London, Ontario N6A 5B7, Canada

ABSTRACTÐA new genus and species of eurypterid (Eurypterida: Chelicerata) is described as Orcanopterus manitoulinensis from the Upper Ordovician Kagawong Submember (Upper Member) of the Georgian Bay Formation, Manitoulin Island, Ontario, Canada. The material comprises several partial specimens in addition to disarticulated carapaces, appendages, metastomas, opisthosomal segments, and telsons. Associated fossils include rare bryozoans, a conularid, ostracodes, and conodonts. A restricted marine lagoon, or very shallow subtidal to intertidal environment is inferred. This assemblage, perhaps representing an accumulation of molted exuviae, was apparently preserved as the result of rapid burial by carbonate muds and silts during a storm event. O. manitoulinensis shares a number of traits with both the Hughmilleriidae and the Carcinosomatidae. Diagnostic features include curved preabdominal segments, a petaloid A metastoma with deep anterior emargination, spiniferous appendages of Carcinosoma type, paddle with enlarged, symmetrical podomere 9, and a xiphous telson. It is only the fourth (the ®rst Canadian) well-documented Ordovician eurypterid genus, and provides the oldest reliable record of the Hughmillerioidea to date.

INTRODUCTION form differs markedly from the new material described herein. URYPTERIDS ARE Paleozoic predatory chelicerates. They are Eurypterid appendages and telsons were also reported by Way E rare as fossils, due to their thin, unmineralized chitinous cu- (1936), from a 15 cm thick, recessive, shaly dolostone unit im- ticle, which was subject to rapid postmortem disarticulation and mediately below typical lithologies of the basal Manitoulin For- bacterial decay (Plotnick, 1999), although they are locally abun- mation in the Kagawong West Quarry (our locality 2; Fig. 1). dant in some strata of Late Silurian age. The evolutionary acme Copeland and Bolton (1960, 1985) assigned this eurypterid-bear- of the group occurred during the Late Silurian and Early Devo- ing stratum, occurring at a slightly higher stratigraphic level than nian, most of the known species originating in the Pridolian± the eurypterid horizon we report herein, to the Manitoulin For- Lochkovian interval (Plotnick, 1999). Reports of pre-Silurian eu- mation and consequently inferred an Early Silurian age for this rypterids are few; most species from the Ordovician of New York fauna. Subsequent biostratigraphic work (Barnes and Bolton, State are particularly dubious (see Tollerton and Landing, 1994; 1988; Nowlan, 2001) has revealed that this recessive unit contains Braddy et al., 2004; Tollerton, 2004). Thus, their early evolution- the stratigraphically highest Ordovician conodont fauna in local ary history is obscure, and any new material from this interval is boundary sections, suggesting continuity with immediately un- potentially signi®cant. derlying lithologies of the Kagawong Submember, Upper Mem- Canadian eurypterids are primarily known from Silurian strata ber, Georgian Bay Formation containing the eurypterid horizon of Ontario (see Copeland and Bolton, 1960, 1985). Rare reports reported herein, and the potential for additional, as yet undiscov- from other provinces include an equivocal specimen, apparently ered, eurypterid layers in the uppermost Ordovician strata of the an eurypterid opisthosomal fragment, from the Upper Ordovician region. (Richmondian) Stony Mountain Formation of southern Manitoba In this paper, we report on a new eurypterid fauna from two (Elias, 1980), an unidenti®ed stylonurid from the Battery Point localities on Manitoulin Island, Ontario, Canada. This fauna was Formation (Middle Devonian) of GaspeÂ, QueÂbec (Jeram, 1996), uncovered during excavation of a 2±3 cm thick dolostone bed in as well as fragmentary remains of probable Megalograptus Miller, the uppermost Kagawong Submember (Upper Member) of the 1874 from Richmondian strata of the Nicolet River Formation, Georgian Bay Formation, slightly below its contact with the Man- RivieÁre des Hurons, southern QueÂbec (Chartier et al., 2002) itoulin Formation, at the West Bay Indian Reserve road cut (lo- Recent reports from Arctic Canada include: Stylonurus sp. from cality 1; Fig. 1), central Manitoulin Island, as well as from a the Snowblind Bay Formation (Early Devonian) of Cornwallis similar, if not identical, stratigraphic level at the Kagawong West Island, Nunavut (Plotnick and Elliott, 1995), and a fauna com- Quarry section (locality 2), 13 km northwest of locality 1. The prising Erieopterus microphthalmus (Hall, 1859), Drepanopterus initial discovery was made at locality 1 by S. Donato, then of the sp., and Carcinosoma sp. from the Bear Rock Formation (Early Department of Earth Sciences, University of Western Ontario, in Devonian) of the Northwest Territories (Braddy and Dunlop, the company of CAS, PMG, and J. Jin. Signi®cant collections 2000). Additional discoveries from Ontario include a putative On- were made during the initial visit, additional material being un- ychopterella carapace from the Whirlpool Formation (early Llan- earthed on a subsequent collecting trip to locality 1 undertaken dovery) near Georgetown (J. Waddington, personal commun., by PMG and MGD. These strata are reliably dated as Late Or- 2003), and material from the Eramosa Member of the Guelph dovician, very close to the Ordovician±Silurian boundary, based Formation (late Wenlock) of the southern Bruce Peninsula which on conodonts (see below). has been tentatively assigned to the pterygotid genus Erettopterus Salter, 1859 (D. Rudkin, personal commun., 2002). Previously reported eurypterids from Manitoulin Island include GEOLOGICAL SETTING AND AGE OF THE FAUNA Carcinosoma libertyi Copeland and Bolton, 1960 from the Early The eurypterid-bearing lithology consists of thinly bedded, Silurian St. Edmund Formation (mid-Llandovery), although this ®nely crystalline hypidiotopic to xenotopic dolostone, probably 1166 STOTT ET AL.ÐNEW ORDOVICIAN EURYPTERID 1167

FIGURE 1ÐMap of eastern and central Manitoulin Island, showing locations of two Ordovician±Silurian boundary sections at which eurypterid remains have been recovered from the uppermost Kagawong Submember, Upper Member, Georgian Bay Formation.

representing recrystallization of primary micritic limestone, sug- et al., 1976; Sweet, 1979a, 1979b; Leatham, 1984; Norford et al., gestive of a relatively low-energy depositional environment. Rare, 1998; Armstrong and Owen, 2002). Low-energy, restricted envi- branching bryozoans and one conularid were the only other mac- ronments are likely to have been present landward of laterally rofossils found in this bed. Probable ostracodes were observed in extensive wave-baf¯ing stromatoporoid-tetradiid-bryozoan car- thin sections of the eurypterid-bearing lithology, as were rare bonate buildups represented by the Mudge Bay and Maple Point bryozoans. The sparse macrofauna suggests that this environment biostromes (see Copper, 1978; Copper and Grawbarger, 1978). was marine, although highly restrictive, possibly the result of the The latter biostrome occurs in the interval between 1.5 and 4 m development of hypersaline, variable salinity, or anaerobic/dys- below the Ordovician±Silurian boundary at locality 1 (Fig. 2). aerobic conditions. These structures are estimated to have risen up to 1 m above the Conodont faunas recovered from the eurypterid-bearing bed surrounding sea¯oor, their tops being only marginally submerged and immediately higher and lower strata (samples M1±4 through during the closing phase of local Ordovician sedimentation (Cop- M1±6; Fig. 2) are dominated by Rhipidognathus symmetricus per and Grawbarger, 1978). Low-energy lagoons in the lee of Branson, Mehl, and Branson, 1951 (Nowlan, 2001). The abun- these biostromes, characterized by poorly aerated waters, would dance of this species indicates the shallowest Late Ordovician provide ideal conditions for the preservation of uncalci®ed ar- conodont biofacies (Sweet and BergstroÈm, 1984), and very shal- thropod cuticle, given suf®ciently rapid burial and anoxia in the low subtidal to intertidal, and possibly hypersaline, conditions bottom waters (see Brett et al., 1997; Plotnick, 1999). Burial was (Kohut and Sweet, 1968; Barnes and FaÊhraeus, 1975; Le FeÁvre almost certainly the product of the deposition of carbonate muds 1168 JOURNAL OF PALEONTOLOGY, V. 79, NO. 6, 2005

FIGURE 2ÐChronostratigraphy and lithostratigraphy of the Upper Ordovician to lowermost Silurian sequence of Manitoulin Island. The bracket adjacent to the lithology column illustrates the portion of the local Ordovician and Silurian stratigraphy represented by the West Bay section. The precise stratigraphic level of the eurypterid horizon within the West Bay (locality 1) section is shown in relation to the Kagawong Submember/ Manitoulin Formation contact, as well as to conodont sample levels (M1 series) throughout the boundary interval. O ϭ sample yielded Ordovician conodonts; S ϭ sample yielded Silurian conodonts.

and silts during a single storm event. Despite pervasive dolomi- absence of large and deep burrows suggesting the prevalence of tization of the host lithology, thin-section examination of a 5 mm anoxic conditions immediately below the sediment-water inter- interval of sediment overlying visible eurypterid integument has face. revealed an upward-®ning trend characteristic of tempestite de- In the absence of escape traces or other indications of the post- posits and the presence of ®ne silt-sized detrital quartz grains in burial conditions of the animals, the true nature of the eurypterid a lithology apparently otherwise free of such grains. The latter assemblage remains uncertain. The partial specimens may repre- may be indicative of the short-term introduction of terrestrially sent animals overcome by local restrictive conditions and/or storm derived ®ne clastics into nearshore carbonate environments by deposition, or may merely represent exuviae. As suggested by returning ¯ows of strong storm waves (see Byerley and Coniglio, Plotnick (1999), this determination is not easily made on the basis 1991). Organic-rich carbonate intervals noted in the West Bay of the fossils alone as all portions of the preservable integument exposure (Fig. 2) and in other outcrops of uppermost Ordovician may potentially be suf®ciently well represented in molts. Evi- strata on Manitoulin Island may constitute evidence for periodic dence of incipient preburial disarticulation in the form of abun- sea¯oor anoxia, most probably occurring in stagnant zones land- dant dissociated skeletal elements favors the latter interpretation. ward of the wave-baf¯ing biostromes. Evidence of bioturbation The suggestion that lagoonal environments were not regularly in- within the eurypterid-bearing bed appears to be con®ned to nar- habited by eurypterids but were sought out as areas where molting row, predominantly horizontal, tubelike trace fossils, the general could occur under current- and predator-free conditions, and in STOTT ET AL.ÐNEW ORDOVICIAN EURYPTERID 1169 which preservation potential was relatively high (Caster and Kjel- were present in patchy distributions characterized by localized lesvig-Waering, 1964; Plotnick, 1999; Braddy, 2001), is a possi- high abundance. Several partial eurypterids were recovered in ad- bility that should be considered in future taphonomic interpreta- dition to dissociated carapaces, appendages, metastomas, opistho- tions of eurypterid faunas. somal segments, and telsons. Drawings were made using a camera Micropaleontological samples obtained from the eurypterid ho- lucida. All measurements are given in millimeters. Morphological rizon and immediately under- and overlying strata have yielded a terminology and systematics for the higher eurypterid taxa follow conodont fauna of Late Ordovician aspect comprised by the taxa Tollerton (1989). The material described is deposited in the col- Aphelognathus grandis Branson, Mehl, and Branson, 1951, Oul- lections of the Royal Ontario Museum (ROM), Toronto, under the odus robustus (Branson, Mehl, and Branson, 1951), Panderodus repository numbers ROM 56449±56463, 57017, 57018, and aff. P. panderi (Stauffer, 1940), Panderodus gibber Nowlan and 57026. Additionally, more poorly preserved and less informative Barnes, 1981, Panderodus gracilis (Branson and Mehl, 1933), material is stored in the ROM under the original label 03D1. Pseudobelodina? aff. Pseudobelodina? dispansa (Glenister, 1957), Pseudobelodina inclinata (Branson and Mehl, 1933), and SYSTEMATIC PALEONTOLOGY Rhipidognathus symmetricus. This conodont fauna is typical of Phylum CHELICERATA Heymons, 1901 Richmondian (approximately equivalent to the British Cautleyan± Order EURYPTERIDA Burmeister, 1843 lower Rawtheyan stages) assemblages of the eastern North Amer- Superfamily HUGHMILLERIOIDEA Kjellesvig-Waering, 1951 ican Midcontinent (Fauna 12). To date, no Atlantic Faunal Region Discussion.The new form shares a number of features with conodonts have been reported from this high in the Georgian Bay two of the constituent families of the Hughmillerioidea, namely Formation. Amorphognathus superbus (Rhodes, 1953), Periodon the Hughmilleriidae and the Carcinosomatidae (see below), but grandis (Ethington, 1959), and Icriodella superba Rhodes, 1953 does not ®t easily within the diagnosis of either family (see Tol- are known from the lowermost 40 m of the Georgian Bay For- lerton, 1989; Braddy et al., 2002), indicating that this form has mation on Manitoulin Island (Barnes et al., 1978; Goldman and an intermediate hughmilleriid-carcinosomatid af®nity, and is pos- BergstroÈm, 1997), suggesting assignment of the lower portion of sibly a basal hughmillerioid. We hesitate herein to propose a new the formation to the A. superbus Zone. On the basis of available family, pending cladistic analysis of the entire Eurypterida, which graptolite and shelly fossil biostratigraphies, the A. superbus±A. is beyond the scope of this study. ordovicicus Zonal boundary on Manitoulin Island is estimated to occur some 50±60 m above the base of the Georgian Bay For- Genus ORCANOPTERUS new genus mation within the claystone-dominated Lower Member; this zonal Type species.Orcanopterus manitoulinensis n. gen. and sp., boundary is regarded to occur in the lower portion of the Rich- by monotypy. mondian Stage within the upper Amplexograptus manitoulinensis Diagnosis.Hughmillerioid with parabolic carapace, antelater- Zone (Fig. 2; Goldman and BergstroÈm, 1997). al intramarginal eyes, curved preabdominal segments, petaloid A Lowermost strata of the Manitoulin Formation outcropping metastoma with deep anterior emargination, a unique paddle with ϳ 20 cm above the eurypterid horizon at locality 1 yield cono- an enlarged, symmetrical podomere 9, and a xiphous telson with donts suggestive of a Silurian age, including Icriodella discreta a median keel. Pollock, Rexroad, and Nicoll, 1970, Kockelella manitoulinensis Etymology.A composite of Ordovician, Canada, and ptero (Pollock, Rexroad, and Nicoll, 1970), Ozarkodina hassi (Pollock, (greek)±wing, nom. trans. opterus, the traditional suf®x for eu- Rexroad, and Nicoll, 1970), and Spathognathodus comptus Pol- rypterid genera. lock, Rexroad, and Nicoll, 1970 sensu formo (Nowlan, 2001). Observed occurrence of the atrypoid brachiopod Zygospiraella ORCANOPTERUS MANITOULINENSIS new species planoconvexa (Hall, 1852), an index of Llandovery A age (Cop- Figures 3±5 per, 1978, 1982) within 20 cm of the base of the Manitoulin Unnamed hughmilleriid; STOTT ET AL., 2001, p. 46±47, ®g. 3. Formation at this locality, con®rms the Rhuddanian age of the Eurypterida indet.? new genus; BRADDY ET AL., 2004, p. 256, table 25.1. Manitoulin Formation. Consequently, the eurypterid fauna occurs New genus and species (?); TOLLERTON, 2004, p. 238, table 2. just below the local Ordovician±Silurian boundary. As in much of the North American Midcontinent, this boundary appears to be Diagnosis.As for genus. de®ned by a disconformity, the result of local nondeposition and/ Description.The carapace is known from four undistorted or subaerial exposure during the latest Ordovician glacioeustatic specimens (only two ®gured: ROM 56450, Fig. 3.1; 56451, Fig. regression (see Dennison, 1976; Kobluk, 1984; Barnes and Bol- 3.2) with length to width ratios ranging from 0.88 to 0.94, aver- ton, 1988; Byerley and Coniglio, 1991). aging 0.89. The lateral angles vary between 90Њ and 105Њ, giving Overall, a mid- to late Richmondian (middle Ashgill) age is a parabolic carapace shape. The eyes are intermediate between indicated for this eurypterid assemblage. Although developed at reniform and oval (Fig. 5), positioned antelaterally and intramar- a similar stratigraphic level, this fauna does not appear to be re- ginally. They are separated from the margin by the relatively lated to the Upper Ordovician eurypterids of Ohio, characterized broad marginal rim, which is widest anteriorly and narrows to- by an extremely well-preserved Megalograptus fauna (Caster and wards the posterolateral angles, and is ornamented with three to Kjellesvig-Waering, 1964). four lines converging posteriorly. The ocelli are known from one specimen (ROM 56450, Fig. 3.1); they are situated slightly pos- MATERIALS AND METHODS terior to the posterior margin of the eyes, and three-eighths the Bedding-plane exposure of the eurypterid-bearing stratum near length of the carapace from the anterior margin. The carapace is the top of the West Bay (locality 1) section permitted a collection ornamented anteriorly with numerous terraced folds (Fig. 3.3) fol- over approximately 10 m2. The eurypterid fossils are preserved lowing the contours. Anterior to and between the eyes, these folds in positive relief in a dark gray, poorly laminated marl matrix, split up into large lunate scales. The suture of the doublure was although in places this matrix has weathered to a pale brown originally reported (Stott et al., 2001) to be of Hughmilleria type, color. Eurypterid cuticle is well preserved as black to maroon however the doublure could not be found and it would appear organic ®lms and accompanying underlying molds at locality 1. that the dorsal marginal rim was originally misinterpreted as the Preservation at locality 2 is comparatively poor. The eurypterid ventral doublure. The chelicerae are unknown, but the spiniferous remains, ranging from fragments to near-complete individuals, prosomal appendages II±V are known from one specimen (ROM 1170 JOURNAL OF PALEONTOLOGY, V. 79, NO. 6, 2005 STOTT ET AL.ÐNEW ORDOVICIAN EURYPTERID 1171

57018; Figs. 3.4, 4.3). We interpret the spines on the podomeres as directed anteriorly, giving the identity of the appendages as labeled in Figure 4.3. Appendage II is poorly preserved and only the outlines of three or four podomeres are distinguished. Ap- pendage III has four short and stout podomeres preserved, and traces of two or three anteriorly oriented spines are evident. Ap- pendage IV has three very stout podomeres and there is evidence for anteriorly oriented spines on the two distal podomeres. Ap- pendage V has the four distal podomeres preserved; there is a massive spine on the penultimate podomere, and the distal podomere is also a long spine. The spiniferous appendages are also known from two coxal morphotypes, here tentatively assigned to appendages IV and V. Coxa IV (ROM 56455; Fig. 3.5) bears at least eight very long and slender curving teeth. Coxa V (ROM 56454; Fig. 3.6) also has eight teeth, but these differ from those of coxa IV in having one blunt, large anterior tooth, and seven smaller posterior teeth. Appendage VI is known from a coxa (ROM 56453; Fig. 3.7) and two other specimens showing the distal podomeres of the paddle. The coxa has a continuous gnathobase of 22 small teeth, all of similar size (0.5 mm long), and an ornament of broad lunules ventrally. The rest of appendage VI is known from podomeres 6, 7, 7a, 8, and 9 (ROM 56452; Fig. 3.8) and 7, 7a, 8, and 9 (ROM 57017; Fig. 4.2). The most conspicuous feature of the paddle is the large size of podomere 9, which makes this a unique type of appendage VI (see below). Five dissociated metastomas are present in the collection. Four of these are morphologically similar and can be con®dently assigned to O. manitoulinensis as one of the four is associated with the holotype (ROM 56459), and appears to be the correct size to belong with it. The only complete metastoma of this type (ROM 56456; Fig. 3.9) is widest at one-third of the distance along its FIGURE 4ÐOrcanopterus manitoulinensis n. gen. and sp.; 1, genital ap- length from the anterior margin, and has a length-width ratio of pendage type B, ROM 56459; 2, distal podomeres (7, 7a, 8, and 9) of 1.91. The specimen also has rounded shoulders and a broad and appendage VI, ROM 57017; 3, appendages II±V, ROM 57018. Scale bars deep-rounded anterior emargination, with an angle of cordation ϭ 10 mm. of 88Њ. The convex sides converge towards the truncated posterior margin, with a lateral angle of 75Њ. The shape is most similar to petaloid A (Tollerton, 1989, ®g. 5.7), but the deep emargination is reduced, being approximately half the length of the other mes- is more similar to that of the obovate shape (Tollerton, 1989, ®g. asomal segments. Several specimens show the opisthosoma to 5.3). The metastoma has a doublure (Fig. 3.9, 3.10), but no dis- have a gradual transition between the pre- and postabdomen. cernible ornament on the dorsal side, and an extremely coarse However, the holotype has what appears to be epimerae (same ornament of broad angular scales on the ventral side (ROM 56456; type as those present in Hughmilleria socialis Sarle, 1903) on the Fig. 3.9, 3.11). A single small metastoma (ROM 57026) (22 mm ventral side of segment 7, and as such, a midsection second-order long, 10 mm wide at widest point; length-width ratio of 2.20) differentiation is present (Tollerton, 1989). The individual seg- exhibits an apparently different morphology in which the position ment boundaries of the mesosoma are highly convex (cf. carci- of greatest width occurs just past the midlength point on the pos- nosomatids) as compared to the straight boundaries of the meta- terior half of the element. It is similar to the larger morphotype, soma (Fig. 3.12), but there is no clear-cut third-order ®gured herein, in having a truncated posterior margin and a deep differentiation as in mixopterids. No fourth-order differentiation anterior notch with rounded shoulders anterolaterally. The simi- is present. larities in these key characters suggest that this could represent The genital operculum is composed of two segments. Although the metastoma of a juvenile O. manitoulinensis, however, we no discrete boundary can be seen, the ornament and coloration of choose to withhold judgment on the latter pending future discov- the two segments are different on the holotype. The genital ap- eries concerning eurypterid ontogeny. pendage of the holotype (ROM 56459; Figs. 3.13, 4.1) was orig- As in most eurypterids, the ®rst visible opisthosomal segment inally interpreted as being a type A, but is here reinterpreted as

←

FIGURE 3ÐOrcanopterus manitoulinensis n. gen. and sp. 1, Parabolic-shaped carapace with visible ocelli, ROM 56450; 2, carapace mold showing antelaterally positioned reniform eyes, ROM 56451; 3, dorsal anterior margin of carapace showing terraced fold ornamentation, ROM 56462; 4, prosomal appendages II±V, ROM 57018; 5, coxa of prosomal appendage IV?, ROM 56455; 6, coxa of prosomal appendage V?, ROM 56454; 7, coxa of prosomal appendage VI, ROM 56453; 8, prosomal appendage VI with swimming paddle, ROM 56452; 9, complete petaloid A metastoma with deep anterior cordation, ROM 56456; 10, incomplete petaloid A metastoma with well-developed dorsal doublure, ROM 56457; 11, ventral surface of metastoma exhibiting prominent crescentic scales, ROM 56456; 12, counterpart and part of partial specimen showing curving of preabdominal segments, ROM 56458 A, B; 13, ventral surface of partial eurypterid showing type B genital appendage with furca, ROM 56459; 14, ventral surface of partial xiphous telson, ROM 56461; 15, dorsal surface of incomplete telson exhibiting prominent central keel, ROM 56460; 16, distorted near- complete specimen, ROM 56449 (scale bar ϭ 50 mm). Scale bars ϭ 10 mm unless otherwise indicated. 1172 JOURNAL OF PALEONTOLOGY, V. 79, NO. 6, 2005 a type B, and is composed of two segments and the furca. There are no deltoid plates visible on the genital operculum, and spa- tulae are not present. Furthermore, the holotype (ROM 56459; Fig. 3.13) also shows that only the ®rst pair of sternites (posterior to the genital operculum) are unfused. The three following sternite pairs are apparently fused, although no sutures are visible. The ornament of the dorsal opisthosoma and the ventral part of the mesosoma is composed of overlapping lunate and chevron-shaped scales, and the ventral metasoma has an ornament of scattered, raised, narrow lunules. This ornament is most notably expressed in smaller specimens on the dorsal middle third of each segment, but in larger specimens, the entire segment has this type of ornament dorsally. Larger specimens also tend to show an ornament of three to four lines parallel to the anterior segment margins on the dorsal side of the mesosoma; laterally these lines curve posteriorly to become parallel to the lateral margins of the segments. Metasomal segments reveal dentate posterior margins. The telson is incomplete but appears to be xiphous-shaped (ROM 56461; Fig. 3.14). There is a narrow marginal rim present, but no marginal ornament. Another specimen (ROM 56460; Fig. 3.15) indicates that a median keel was present on the dorsal side of the telson. Etymology.After Manitoulin Island, Ontario, where the specimens were found. Type.Holotype, ROM 56459, Royal Ontario Museum, Toron- to, Canada. Other material examined.ROM 56449±56458; 56460±56463; 57017±57018; 57026 and many additional specimens in the ROM all retaining the original labeling (03D1), repository as above. Occurrence.Kagawong Submember (Upper Member) of the Georgian Bay Formation, Upper Ordovician (Richmondian). West Bay Indian Reserve road cut and Kagawong West Quarry section, Manitoulin Island, Ontario, Canada. Discussion.All well-preserved specimens can reliably be assigned to O. manitoulinensis, which on the basis of the largest partial specimen (Fig. 3.16) and proportions of dissociated elements is estimated to have had a total mature body length ap- proaching 60 cm. A reconstruction of O. manitoulinensis is pro- vided in Figure 5; the number of spines on each podomere of the spiniferous appendages is presumed to be two, as in hughmilleriids and most carcinosomatids. The variable opisthosomal ornament initially suggested that two forms were present in the fauna. However, one specimen (ROM 56458) shows that the ornament of the dorsal opisthosoma and the ventral mesosoma is composed of overlapping large lunate and chevron-shaped scales, and the ventral metasoma has an ornament of smaller, scattered, raised, narrow lunules more similar to that of Salteropterus abbreviatus (Salter, 1859). Fragments with an ornament of lunate and chevron-shaped scales mixed with fragments with other ornament do not necessarily imply that 1) a pterygotid is present in the fauna, nor that 2) there are two different forms present in the fauna. The spiniferous appendages are all apparently of Carcinosoma type (Tollerton, 1989), with much longer spines than those present in the Hughmilleria type. It also appears that the spines were anteriorly oriented, typical for carcinosomatids. The swimming appendage (VI) is unique. The large podomere 9 can only be compared to that of Dolichopterus Hall, 1859 and those of the carcinosomatids Carcinosoma Claypole, 1890 and Paracarcino- soma Caster and Kjellesvig-Waering, 1964, although this podomere is larger than the Carcinosoma type and more symmetrical FIGURE 5ÐReconstruction of Orcanopterus manitoulinensis n. gen. and than the Dolichopterus type. Podomeres 7 and 8 closely resemble ϭ the Mixopterus type. Coxa VI is interesting as it has a typical sp. with a type B genital appendage ( male). 1, Dorsal view; 2, ventral view. The chelicerae are hypothetical and reconstructed according to shape for hughmilleriids with a very bulbous distal end and a its closest relatives. proximal, long, continuous gnathobase with 22 teeth. Coxa VI in Hughmilleria socialis has around 18±20 teeth (Sarle, 1903), while mixopterids (16 teeth for Lanarkopterus dolichoschelus Ritchie, STOTT ET AL.ÐNEW ORDOVICIAN EURYPTERID 1173

1968) and pterygotids (12±13 teeth; Clarke and Ruedemann, the proposed nektobenthic habit of Ordovician conodontophorids. Le- 1912) have a reduced number of teeth on this element. Another thaia, 8(2):133±149. character shared between O. manitoulinensis and hughmilleriids BARNES, C. R., P. G. TELFORD, AND G. A. TARRANT. 1978. Ordovician are a series of lines parallel to the anterior segment margins on and Silurian conodont biostratigraphy, Manitoulin Island and Bruce Peninsula, Ontario. Michigan Basin Geological Society Special Paper, the mesosomal segments. This has been reported from, for ex- 3:63±71. ample, Parahughmilleria hefteri Stùrmer, 1974. Another putative BRADDY, S. J. 2001. Eurypterid palaeoecology: palaeobiological, ichnol- character shared with hughmilleriids is the dentate posterior meta- ogical and comparative evidence for a `mass-moult-mate' hypothesis. somal segment margins previously reported from P. hefteri (e.g., Palaeogeography, Palaeoclimatology, Palaeoecology, 172:115±132. Stùrmer, 1974). BRADDY,S.J.,AND J. A. DUNLOP. 2000. Early Devonian eurypterids The type B genital appendage in O. manitoulinensis is more from the Northwest Territories of Arctic Canada. Canadian Journal of similar to that described in Kokomopterus longicaudatus (Clarke Earth Sciences, 37:1167±1175. and Ruedemann, 1912) by Kjellesvig-Waering (1966) and BRADDY, S. J., P. A. SELDEN, AND T. D OAN NHAT. 2002. A new carci- Stùrmer (1974) than those from established hughmilleriids, while nosomatid eurypterid from the Upper Silurian of northern Vietnam. type B appendages are poorly known from carcinosomatids. The Palaeontology, 45:897±915. BRADDY, S. J., V. P. TOLLERTON JR. , P. P. R ACHEBOEUF, AND R. SCHALL- implications of this similarity are presently unknown; K. longi- REUTER. 2004. Eurypterids, phyllocarids and ostracodes, p. 255±265. caudatus is a stylonurid eurypterid, and thus would seem to be In B. D. Webby, M. L. Droser, and F. Paris (eds.), The Great Ordovician only distantly related to O. manitoulinensis on the basis of other Biodiversi®cation Event. Columbia University Press, New York. characters. The telson has a marginal rim, but does not have any BRANSON, E. B., AND M. G. MEHL. 1933. Conodont Studies. University marginal ornament of sclerotized scales, as is the case in the pter- of Missouri Studies, 8, 349 p., 28 pls. ygotids, as well as the genera Slimonia Page, 1856, Salteropterus BRANSON, E. B., M. G. MEHL, AND C. C. BRANSON. 1951. Richmond Kjellesvig-Waering, 1951, and the species Hughmilleria banksii conodonts of Kentucky and Indiana. Journal of Paleontology, 25:1±17. (Salter, 1856) (OET, personal observation). This suggests that BRETT, C. E., G. C. BAIRD, AND S. E. SPEYER. 1997. Fossil LagerstaÈtten: Stratigraphic record of paleontological and taphonomic events, p. 3± similarities between the ornament of O. manitoulinensis devel- 40. In C. E. Brett and G. C. Baird (eds.), Paleontological Events. Co- oped elsewhere, and that of the pterygotids are a product of con- lumbia University Press, New York. vergence and do not re¯ect a trait inherited from a common an- BURMEISTER, H. 1843. Die Organisation der Trilobiten, aus ihren leben- cestor. den Verwandten entwickelt; nebst systematische Uebersicht aller zeith- Some characters are consistent with those of the family Hugh- er beschriebenen Arten. G. Reimer, Berlin, 148 p. milleriidae Kjellesvig-Waering, 1951, such as the carapace shape, BYERLEY, M., AND M. CONIGLIO. 1991. Stratigraphy and sedimentology midsection second-order opisthosomal differentiation, coxal of the Upper Ordovician Georgian Bay Formation, Manitoulin Island shape, the telson keel, and some of the ornamentation. However, and Bruce Peninsula. Evaluation of hardgrounds, biostromes and storm other characters, such as the spinosity of appendages II±V, the beds for regional correlation. Ontario Geological Survey Miscellaneous form of podomeres 7±9 of appendage VI, the metastomal shape, Paper, 156:3±15. CASTER, K. E., AND E. N. KJELLESVIG-WAERING. 1964. Upper Ordovi- and the curvature of the preabdominal segments, are more con- cian eurypterids of Ohio. Palaeontographica Americana, 4:301±358. sistent with the Carcinosomatidae Stùrmer, 1934. A phylogenetic CHARTIER, M., M. COURNOYER, AND P. V EILLEUX. 2002. Upper Ordo- analysis of the entire Eurypterida, including this new form, is vician eurypterids from the RivieÁre des Hurons, southern QueÂbec. Ca- required to determine whether the new species belongs to the nadian Paleontology Conference 2002 Program and Abstracts, 12:7. carcinosomatid clade, the ``hughmilleriids,'' or another group, but CLARKE, J. M., AND R. RUEDEMANN. 1912. The Eurypterida of New this work requires various systematic revisions and is beyond the York. Memoirs of the New York State Museum of Natural History, 14: scope of this contribution. 1±439. CLAYPOLE, E. W. 1890. Carcinosoma newlini. American Geologist, 6: ACKNOWLEDGMENTS 400. COPELAND,M.J.,AND T. E. BOLTON. 1960. The Eurypterida of Canada. S. Donato, formerly of The University of Western Ontario Geological Survey of Canada Bulletin, 60:13±48. (UWO) (currently at McMaster University, Ontario), is thanked COPELAND,M.J.,AND T. E. BOLTON. 1985. Fossils of Ontario, Pt. 3, for the initial discovery and recognition of this material and his The Eurypterids and Phyllocarids. Ontario: Royal Ontario Museum, assistance in the ®eld. D. Tetreault (formerly at UWO, now at Life Sciences Miscellaneous Publications, 48 p. University of Windsor, Ontario), D. Rudkin (ROM), and R. A. COPPER, P. 1978. Paleoenvironments and paleocommunities in the Or- Moore (University of Bristol) offered constructive and helpful dovician±Silurian sequence of Manitoulin Island. Michigan Basin Geo- discussions. A. D. McCracken (Geological Survey of Canada) logical Society Special Paper, 3:47±61. reviewed the manuscript before ®nal submission. OET thanks J. COPPER, P. 1982. Early Silurian atrypoids from Manitoulin Island and Waddington (ROM) for making this material available for study. Bruce Peninsula, Ontario. Journal of Paleontology, 56:680±702. The paper was reviewed by B. Kues and one anonymous review- COPPER,P.,AND D. J. GRAWBARGER. 1978. Paleoecological succession leading to a late Ordovician biostrome on Manitoulin Island, Ontario. er. Field work and laboratory preparation of the fossils were fund- Canadian Journal of Earth Sciences, 15:1987±2005. ed through a NSERC research grant to J. Jin (UWO). OET ac- DENNISON, J. M. 1976. Appalachian Queenston delta related to eustatic knowledges funding from the NFR (Norwegian Research sea-level drop accompanying Late Ordovician glaciation centred in Af- Council) grant 145565/432 and the University of Bristol, United rica, p. 107±120. In M. G. Bassett (ed.), The Ordovician System: Pro- Kingdom. ceedings of a Palaeontological Association Symposium, Birmingham, U.K., September 1974. University of Wales Press and National Mu- REFERENCES seum of Wales, Cardiff. ARMSTRONG, H. A., AND A. W. OWEN. 2002. Euconodont diversity ELIAS, R. J. 1980. An Upper Ordovician eurypterid from Manitoba. Jour- changes in a cooling and closing Iapetus Ocean, p. 85±98. In J. A. nal of Paleontology, 54:262±263. Crame and A. W. Owen (eds.), Palaeobiology and Biodiversity Change: ETHINGTON, R. L. 1959. Conodonts of the Ordovician Galena Formation. The Ordovician and Mesozoic±Cenozoic Radiations. Geological Soci- Journal of Paleontology, 33:257±292. ety, London, Special Publications, 194. GLENISTER, A. T. 1957. The conodonts of the Ordovician Maquoketa BARNES, C. R., AND T. E. BOLTON. 1988. The Ordovician±Silurian Formation in Iowa. Journal of Paleontology, 31:715±736. boundary on Manitoulin Island, Ontario, Canada. Bulletin of the British GOLDMAN, D., AND S. M. BERGSTROÈ M. 1997. Late Ordovician graptolites Museum of Natural History (Geology), 43:247±253. from the North American Midcontinent. Palaeontology, 40:965±1010. BARNES, C. R., AND L. E. FAÊ HRAEUS. 1975. Provinces, communities, and HALL, J. 1852. Containing Descriptions of the Organic Remains of the 1174 JOURNAL OF PALEONTOLOGY, V. 79, NO. 6, 2005

Lower Middle Division of the New York System, (Equivalent in Part POLLOCK, C. A., C. B. REXROAD, AND R. S. NICOLL. 1970. Lower Si- to the Middle Silurian Rocks of Europe). Palaeontology of New York. lurian conodonts from northern Michigan and Ontario. Journal of Pa- Volume 2. New York Geological Survey, Albany, 362 p. leontology, 44:743±764. HALL, J. 1859. Eurypterida. Palaeontology of New York. Volume 3. New RHODES, F. H. T. 1953. Some British Lower Palaeozoic conodont faunas. York Geological Survey, Albany, p. 382±424. Royal Society of London Philosophical Transactions, series B, HEYMONS, R. 1901. Die Entwicklungsgeschichte der Scolopender. Zool- 237(647):261±334. ogica, 13(2±5):244. RITCHIE, A. 1968. Lanarkopterus dolichoschelus (Stùrmer) gen. nov., a JERAM, A. J. 1996. Chelicerata from the Escuminac Formation, p. 103± mixopterid eurypterid from the Upper Silurian of the Lesmahagow and 111. In H.-P. Schultze and R. Cloutier (eds.), Devonian Fishes and Hagshaw Hills inliers, Scotland. Scottish Journal of Geology, 4:317± Plants of Miguasha, QueÂbec, Canada. Verlag Dr. Friedrich Pfeil, Mu- 338. nich. SALTER, J. W. 1856. On some new Crustacea from the Uppermost Silu- KJELLESVIG-WAERING, E. N. 1951. Downtonian (Silurian) Eurypterida rian Rocks. Quarterly Journal of the Geological Society, 12:26±34. from Perton, near Stoke Edith, Herefordshire. Geological Magazine, SALTER, J. W. 1859. On some new species of Eurypterus; with notes on 88:1±24. the distribution of species. Quarterly Journal of the Geological Society, KJELLESVIG-WAERING, E. N. 1966. A revision of the families and genera 15:229±236. of the Stylonuracea (Eurypterida). Fieldiana Geology, 14:169±197. SARLE, C. J. 1903. A new eurypterid fauna from the base of the Salina KOBLUK, D. R. 1984. Coastal paleokarst near the Ordovician±Silurian of western New York. New York State Museum Bulletin, 69:1080± boundary, Manitoulin Island, Ontario. Bulletin of Canadian Petroleum 1108. Geology, 32:398±407. STAUFFER, C. R. 1940. Conodonts from the Devonian and associated KOHUT,J.J.,AND W. C. SWEET. 1968. The American Upper Ordovician clays of Minnesota. Journal of Paleontology, 14:417±435. Standard; X, Upper Maysville and Richmond conodonts from the Cin- STùRMER, L. 1934. Merostomata from the Downtonian Sandstones of cinnati Region of Ohio, Indiana, and Kentucky. Journal of Paleontol- Ringerike, Norway. Skrifter utgitt av Det Norske Videnskaps-Akade- ogy, 42:1456±1477. mie i Oslo, 1933, No. 10, 125 p., 12 pls. LEATHAM, W. B. 1984. Conodont biostratigraphy of the Ordovician/Si- STùRMER, L. 1974. Arthropods from the Lower Devonian (Lower Em- lurian systemic boundary in the Fish Haven and Laketown dolomites sian) of Alken an der Mosel, Germany, Pt. 4, Eurypterida, Drepanop- of northern Utah. Unpublished M.Sc. thesis, The Ohio State University, teridae, and other groups. Senckenbergiana Lethaea, 54:359±451. Columbus, 187 p. STOTT, C. A., P. M. GLASSER, AND M. G. DEVEREUX. 2001. A new LE FEÁ VRE, J., C. R. BARNES, AND M. TIXIER. 1976. Paleoecology of Late eurypterid LagerstaÈtte in Uppermost Ordovician strata, Manitoulin Is- Ordovician and Early Silurian conodontophorids, Hudson Bay Basin, land, Ontario. Canadian Paleontology Conference 2001 Program and p. 69±89. In C. R. Barnes (ed.), Conodont Paleoecology. Geological Abstracts, 11:45±49. Association of Canada Special Paper, 15. SWEET, W. C. 1979a. Conodonts and conodont biostratigraphy of the MILLER, S. A. 1874. Notes and descriptions of Cincinnatian Group fos- post±Tyrone Ordovician rocks of the Cincinnati Region. United States sils. Cincinnati Quarterly Journal of Science, 1:343±351. Geological Survey Professional Paper, 1066-G, 26 p. NORFORD, B. S., G. S. NOWLAN,F.M.HAIDL, AND R. K. BEZYS. 1998. SWEET, W. C. 1979b. Late Ordovician conodonts and biostratigraphy of The Ordovician±Silurian boundary interval in Saskatchewan and Man- the western Midcontinent Province. Brigham Young University, Ge- itoba. Eighth International Williston Basin Symposium, Saskatchewan ology Studies, 26(3):45±85. Geological Survey Special Publication, 13:27±45. SWEET,W.C.,AND S. M. BERGSTROÈ M. 1984. Conodont provinces and NOWLAN, G. S. 2001. Report on twenty-®ve samples from Ordovician biofacies of the Late Ordovician. Geological Society of America Spe- and Silurian strata, near Kagawong and West Bay, Manitoulin Island, cial Paper, 196:69±87. Ontario. Geological Survey of Canada Paleontological Report, 009- TOLLERTON JR., V. P., 1989. Morphology, taxonomy, and classi®cation GSN-2001, 18 p. of the order Eurypterida Burmeister, 1843. Journal of Paleontology, 63: NOWLAN, G. S., AND C. R. BARNES. 1981. Late Ordovician conodonts 642±657. from the VaureÂal Formation, Anticosti Island, QueÂbec. Geological Sur- TOLLERTON JR., V. P. 2004. Summary of a revision of New York State vey of Canada Bulletin, 329:1±49. Ordovician eurypterids: Implications for eurypterid palaeoecology, di- PAGE, D. 1856. Advanced Text-Book of Geology. William Blackwood versity and evolution. Transactions of the Royal Society of Edinburgh: and Sons, Edinburgh, 416 p. Earth Sciences, 94:235±242. PLOTNICK, R. E. 1999. Habitat of Llandoverian±Lochkovian eurypterids, TOLLERTON JR., V. P. AND E. LANDING. 1994. The myth of Ordovician p. 106±131. In A. J. Boucot and J. D. Lawson (eds.), Paleocommuni- eurypterids in New York State. Geological Society of America Ab- ties: A Case Study from the Silurian and Lower Devonian. Cambridge stracts with Programs, 26(3):76. University Press, Cambridge. WAY, H. G. 1936. The Silurian of Manitoulin Island, Ontario. Unpub- PLOTNICK,R.E.,AND D. K. ELLIOTT. 1995. A Lower Devonian stylon- lished Ph.D. thesis, University of Toronto, Ontario, 123 p. urid eurypterid from Arctic Canada. Journal of Paleontology, 69:399± 402. ACCEPTED 25 OCTOBER 2004