INTRODUCTION

1 The Campbellton Formation exposed at Campbellton, New Brunswick (Fig. 1), has been known for the fauna of ostracoderms, arthrodires, acanthodians and chondrichthyans (Whiteaves 1881, 1889; Woodward 1889, 1892; Traquair 1890, 1893) since the first vertebrate fossils were discovered in 1881 (Whiteaves 1899). Dineley and Williams (1968) called the vertebrate-bearing part of the formation the ‘Atholville beds’. In addition to fish, the assemblage includes ostracods (Jones 1889), molluscs (Whiteaves 1881), and eurypterids (Whiteaves 1881; Clarke and Ruedemann 1912). Plants have also been identified from the Campbellton locality (Gensel and Albright 2006). The Campbellton Formation east of Campbellton, from Dalhousie Junction to Point Pin Sec (Fig. 1A), has not yielded vertebrates, but since the mid-1800’s has produced a rich flora of zosterophyllophytes, trimerophytes and lycopsids (Gensel and Andrews 1984; Gensel et al. 1991). Recently an assemblage of terrestrial arthropods, including Eoarthropleura devonica and scorpion remains (Shear et al. 1996), and the millipede Gaspestria genselorum (Wilson 2006) has been identified near the Dalhousie Junction and Point la Nim areas (Fig. 1A). West of Campbellton at Atholville (Fig. 1B), the Atholville beds examined since 1994 have produced remains of chondrichthyans (Miller et al. 2003), ostracoderms, arthrodires, acanthodians and pterygotid eurypterids (Miller 1996, 2007).

Display large image of Figure 1

2 Pterygotids, which can reach more than two metres in length (Kjellesvig-Waering 1964; Chlupáč 1994), are relatively rare in the fossil record. The Pterygotoidea is believed to be the most cosmopolitan of the eurypterid clades (Tetlie 2007). It is also the most diverse with 56 species, although Tetlie (2007) believed the clade might be oversplit. Whiteaves (1881), at the Geological Survey of Canada, was the first to note Pterygotus sp. in the Lower Devonian Campbellton Formation from the Campbellton locality (Fig. 1B). His specimens (GSC 3239, 3239a-c; Bolton 1966) were later described as P. atlanticus by Clarke and Ruedemann (1912, p. 358) who considered their new species to be "of considerable dimensions" and compared it to Pterygotus anglicus Agassiz, 1844. They suggested that with larger collections the species might prove to be "a vicarious form" of P. anglicus, the type species of the genus from the Early Devonian of Scotland (Agassiz 1844; Huxley and Salter 1859; Woodward 1866-78; Kjellesvig-Waering 1964; Waterston 1964). Pageau (1969a) briefly mentioned finding pterygotid fragments associated with fish fossils from Campbellton.

3 This paper describes for the first time, pterygotid eurypterid fossils collected in the late 1800’s from the base of the Campbellton Formation exposed at the classic Campbellton locality (Fig. 1B), and reposited in the Natural History Museum, London (NHM I and NHM In). Although there are more than 40 specimens, a large collection for the Maritime Provinces of eastern Canada, little attention has been paid to them. In 1889, A.S. Woodward (1889, 1892) began to describe the fish fossils from Campbellton acquired by the British Museum (Natural History) (now the Natural History Museum, London) through fossil dealer R.F. Damon. Along with fish, Damon supplied pterygotid specimens in 1888, 1892 and 1893. One specimen, NHM In.18520, was later identified as Pterygotus atlanticus Clarke and Ruedemann (NHML specimen label notation by E.N. Kjellesvig-Waering), but otherwise the specimens have been identified only as Pterygotus sp. The National Museums of Scotland (NMS) also purchased a collection of Campbellton specimens from Damon in 1897 that included 12 small fragments with pterygotid remains; about half are undiagnostic cuticle fragments. Traquair (1890, 1893), at the museum in Edinburgh, also described vertebrates from the Campbellton locality.

LOCATION AND GEOLOGY

4 The Devonian Campbellton Formation (Williams et al. 1985; Wilson et al. 2004; New Brunswick Department of Natural Resources 2007) is exposed along the banks of the Restigouche River-Bay of Chaleur, New Brunswick, Canada from Atholville in the west to its eastern limit near Dalhousie (Fig. 1A). The Campbellton Formation unconformably overlies rhyolite of the Lower Devonian Val d’Amour Formation of the Dalhousie Group (Wilson et al. 2004). Dineley and Williams (1968) referred to the vertebrate-bearing beds, outcropping for almost 2 km on the south shore of the Restigouche River from Campbellton to Atholville, as the ‘Atholville beds’ (Fig. 1B). They are comprised of a steeply inclined basal coarse breccia overlain by a succession of interbedded sandstones and mud-stones (Fig. 2). Published descriptions (Whiteaves 1881) and specimen labels indicate collections of fish and eurypterids from the late 1800’s were all made toward the Campbellton end of the Atholville beds, east of Pratt Point (Fig. 1B) where the fossil-bearing unit today only consists of a thin veneer of mudstone overlying the irregular surface of a steep rhyolite cliff face. The basal fossil-bearing sediments display no clear bedding.

Display large image of Figure 2

5 Most specimens supplied to the Natural History Museum, London and the National Museums of Scotland occur in a grey mudstone rich in ostracods and gastropods, and sometimes containing volcanic clasts. Three of the pterygotid specimens described by Whiteaves (1881) at the Geological Survey of Canada are in a dark grey breccia of mudstone and rhyolite. Similar rocks occur in Campbellton just above the Val d’Amour Formation rhyolite, but are not seen above the rhyolite in out-crops at Atholville. Sampling of the eastern-most exposure of the formation at Campbellton (48°00.13N; 66°41.68W) has yielded fish and rare eurypterid remains in the breccia immediately overlying the Val d’Amour Formation, suggesting this may be the source of some specimens at the Natural History Museum, London. Most NHM specimens are not in the breccia and likely came from an overlying mudstone presumably depleted by early collectors. The mudstone breccia overlying the rhyolite appears to be locally deposited and may not form the basal unit of the entire Atholville beds of Dineley and Williams (1968). Slumping, industrial debris and vegetation obscure much of the outcrop.

6 Palynological data date the Campbellton Formation as Emsian (Blieck and Cloutier 2000). Richardson and McGregor (1986) identified the Emphanisporites annulatus-Camarozonotriletes sextantii miospore Assemblage Zone which approximately corresponds to the Polygnathus dehiscens to P. serotinus Conodont Zones of early Emsian to early late Emsian age (Blieck and Cloutier 2000). McGregor (1973, 1977) placed the outcrops along the Restigouche River within the Pragian "caperatus-emsiensis" and early to late Emsian spore zones, "sextantii-Grandispora" to "annulatus-lindlarensis". The Pterygotus-bearing beds are near the base of the Campbellton Formation, probably within the early Emsian, not far above the Pragian-Emsian boundary stratotype age at 407 ± 2.8 Ma (Ogg 2004).

7 Wilson et al. (2004) recognized an angular unconformity separating the Val d’Amour and Campbellton formations and suggested the Campbellton Formation straddled the Emsian-Eifelian boundary stratotype age at 397.5 ± 2.7 Ma (Ogg 2004). Their interpretation is based on a radiometric date of 407.4 ± 0.8 Ma in the Val d’Amour Formation (Fig. 1A) comprised of mafic to felsic volcanic rocks and minor sediments (Wilson et al. 2004). Reexamination of the contact, following the publication of Wilson et al. (2004), suggests perhaps a closer relationship than interpreted by these authors between the Val d’Amour Formation rhyolite and the Campbellton Formation sediments, consistent with palynological data. Some authors (Pageau 1968, 1969a, b) have considered the Campbellton out-crops equivalent to the Battery Point Formation or the LaGarde and Pirate Cove formations (Wilson et al. 2004; Bourque et al. 2005) found north of the Bay of Chaleurs in Gaspé, Quebec.

SYSTEMATIC PALEONTOLOGY

8 Dunlop et al. (2002) briefly discussed eurypterid systematics and recognized that a number of problems exist due in part to a large number of taxa based on incomplete or even nondescript fossils. They published a provisional cladogram with suggested synapomorphies for a selected set of taxa including Pterygotus. More recently Ciurca and Tetlie (2007) and Tetlie (2007) have reviewed eurypterid phylogeny and provided an updated cladogram. Miller (2007) provided an emended diagnosis for Pterygotus anglicus with additional details concerning the chelicera, considered one of the key diagnostic structures to distinguish species of pterygotids. Tollerton (1989) provided a summary of terminology to describe and differentiate eurypterids. Diagnoses below are a composite of descriptions applied to the Family and species description of Pterygotus anglicus.

CONCLUSION

9 The depositional environment of the Campbellton Formation has been considered fluvial (Williams et al. 1985) or described as a coarsening-upward alluvial-lacustrine sequence (Rust et al. 1989; Gamba 1990; Wilson et al. 2004). A lagoon or estuary environment is likely represented (Miller 2007) since rare prasinophyte algae (tasmanitids) from the upper part of the section at Atholville (Fig. 2) indicate a marine connection (Blieck and Cloutier 2000). In the basal sediments outcrop-ping at Campbellton, ostracods are abundant in the basal mudstone breccia and plant remains are common. The fish assemblage found in the breccia and the mudstone containing the pterygotids from the Campbellton locality includes two ostracoderms, Yvonaspis campbelltonensis and Y. jexi (Pageau 1969a; Belles-Isles 1989), three placoderms Phlyctaenius acadicus, P. atholi,and P. stenosus (Young 1983), and two acanthodians Mesacanthus semistriatus and Cheiracanthus? costellatus (Pageau 1969a). ‘Gyracanthus’(?) incurvus (Denison 1979) is considered as probably being a stem group gyracanthid (Turner et al. 2005). Two sharks, Doliodus problematicus (Woodward 1892; Miller et al. 2003; Turner 2004; Turner and Miller 2005a) and Protodus jexi (Woodward 1892; Turner and Miller 2005b) are also found at Campbellton. Spines of Climatius latispinosus (Whiteaves 1881) are common, although its position as an acanthodian is currently under review as it is most likely chondrichthyan (Miller et al. 2003). Spines identified as Ctenacanthus ornatus, are of uncertain affinity. Turner and Miller (2003) suggested that in this lowest part of the Campbellton Formation volcanic mudflows or lahars inundated coastal waters killing plants and animals, perhaps in a single event, leaving their remains in a chaotic mess. Fossil preservation supports an interpretation of rapid burial as some cephalaspids in collections of the Natural History Museum, London and the National Museums of Scotland include the headshield and the post-cranial body. Relatively complete acanthodians are known from the National Museums of Scotland collection and a three-dimensionally preserved acanthodian in the New Brunswick Museum collection (R. Cloutier, personal communication, 2004) was recovered from the basal breccia. Whether the deposits represent a lahar or not, the basal breccia exposed at Campbellton is a mudflow filled with a jumble of fossils and rhyolite fragments infilling fissures in the underlying rhyolite. The deposit seems to have been laid down over a limited geographical area. The Atholville beds outcrop sporadically along the Restigouche River shoreline and under the river from Campbellton to Atholville (Fig. 1B). Slumping, artificial fill, and gravel covered beaches have prevented measurement of a complete section and an adequate palaeoenvironmental interpretation.

10 Pterygotus anglicus from the Atholville locality (Miller 2007) occurs higher in the section in a relatively unfossiliferous mud-stone (Fig. 2), with only occasional to rare plant fragments and few ostracods. P. anglicus has also been found in a siltstone within a cross-bedded sandstone-siltstone unit (Fig. 2). The most complete specimens at Atholville are found in association with sparse cephalaspid and placoderm remains, relative to the Campbellton locality, along with rare acanthodian spines and the remains of only one shark species, Doliodus problematicus (Miller et al. 2003).

11 Pterygotids are believed to have had a wide environmental range (Plotnick and Baumiller 1988; Chlupáč 1994) with species of Pterygotus occurring from freshwater to marine benthic assemblage 5. At Early Devonian sites in the Scottish Midland Valley, fish beds belonging to the Arbuthnott Group are considered lacustrine. There, P. anglicus occurs with acanthodians (Mesacanthus, Ischnacanthus, and Euthacanthus) and ostracoderms (Cephalaspis pagei) and may have spent its whole life in freshwater, since the full size range of P. anglicus individuals are found in the lake beds (Trewin and Davidson 1996; Braddy 2000, 2001). In the Campbellton Formation P. anglicus accompanies sharks and placoderms, indicating part of its life cycle may have been spent in brackish estuaries or lagoons. The brackish shoreline environments may have provided a suitable site for moulting. Poschmann and Tetlie (2006) suggested a pterygotid found in New York State, Acutiramus macrophthalmus, may have inhabited different environments depending on the age of the animal. They surmised that solitary, medium-sized individuals represented a marine phase of the life cycle, while in lagoons numerous small individuals found with mature adults might represent a nursery. Braddy (2001) proposed a ‘mass-moult-mate’ hypothesis in which eurypterids migrated en masse into nearshore and marginal environments such as lagoons to moult and mate and that they probably required a quiet, current-free place to moult. The Atholville beds may represent part of a continuum of environments inhabited by Pterygotus anglicus, from freshwater to marine. Although a range of sizes of individuals occurs in the Campbellton Formation the total number represented is still small, perhaps less than a dozen animals.