We present the first description of organic-walled microfossils from Cambrian strata of the Bourinot belt, central Cape Breton Island. Age-diagnostic acritarchs have been recovered from the Dugald and MacMullin formations and from probable levels within the upper part of the Eskasoni Formation, which permit detailed correlations with acritarch-based zones in Newfoundland and Spain. The assemblage of acritarchs from the Dugald Formation confirms earlier assignments to the early middle Cambrian eteminicus Zone, but it also indicates that the upper part of the formation belongs to the hicksi Zone of the Drumian Stage. Acritarchs from the MacMullin Formation provide the first biostratigraphic evidence that this unit extends into the forchhammeri Zone of the Guzhangian Stage. These acritarchs are present in the lower part of the MacMullin Formation, putting into question earlier identification of hicksi Zone trilobites in this unit and raising the possibility of an unconformity. The data from the Bourinot belt provide additional evidence for the biostratigraphic utility of acritarchs in the Cambrian Acado-Baltic province.
The Bourinot belt is a narrow fault-bounded band of lower Palaeozoic sedimentary and volcanic rocks that extends for about 30 km in a southwest–northeast direction in central Cape Breton Island, Nova Scotia (White et al. 1994; Figs. 1, 2). Bimodal volcanic rocks, shale, and sandstone of the Bourinot Group form the lower part of the succession and are divided, from base to top, into the Eskasoni, Dugald, and Gregwa formations (Hutchinson 1952; White et al. 1994). The Dugald Formation contains middle Cambrian brachiopods and trilobites (Walcott 1912, p. 133; Hutchinson 1952), and the Bourinot Group has been considered to be early middle Cambrian in its entirety (Hutchinson 1952; Landing 1996). The Bourinot Group is overlain by the more sandstone-rich MacMullin Formation, from which Middle Cambrian trilobites have been recorded, and locally by the Furongian MacNeil Formation, consisting largely of dark shale (Hutchinson 1952; White et al. 1994).
The relationship of the Bourinot Group to other Cambrian successions in the vicinity, such as that in the Mira River area to the southeast, has long been controversial. The controversy arises because the Bourinot Group lacks obvious correlative units in the Mira River area, particularly in that equivalents of the volcanic Eskasoni and Gregwa formations are missing. There are also differences in the basement on which these successions rest (e.g., Barr et al. 1998). These differences have been explained either by the successions in the Bourinot belt and Mira River area having developed on different terranes (e.g., White et al. 1994; Barr et al. 1998) or as regional differences within a common terrane (e.g., Landing 1996). Successions west of Mira River have been proposed to provide lithostratigraphic similarities to the Bourinot Group with the presence of basalt and rhyolite (e.g., Landing 1996), but other studies have shown that Cambrian volcanic rocks are not present in that area (Barr et al. 1996).
In this paper, we present the first data on organic-walled microfossils from Cambrian strata in the Bourinot belt. The material is well-preserved and includes numerous age-diagnostic taxa, which enable comparison with acritarch-based zones established in other areas. Particular emphasis is made here on the comparison to the Mira terrane (Palacios et al. 2009) and to acritarch-based zones developed in southeastern Newfoundland (Martin and Dean 1981, 1983, 1984, 1988) and Spain (Palacios 2008, 2010; Palacios et al. 2006). No evidence is known for provincialism in the distribution of Cambrian acritarchs (e.g., Moczydłowska 1998, p. 38). For example, Cambrian strata of Baltica, and many of the peri-Gondwanan terranes, now spread across eastern North America and southern Europe (i.e., approximating the Acado-Baltic faunal province; see for example Cowie 1971; Sdzuy 1972) and have been the focus of numerous studies of organic-walled microfossils with a generally uniform distribution of taxa (e.g., Palacios et al. 2009). Thus, while the assemblages of organic-walled microfossils described here do not yield new insights into the provenance and tectonic evolution of the Bourinot belt, they enable more secure temporal correlation of the Cambrian sequence of the Bourinot belt to other regions. The material described also includes occurrences of little reported taxa and expands the known geographical distribution of many taxa. Much of the results refer to strata belonging to Cambrian series 3, the base of which has not been formally defined. In the text, we therefore largely maintain the traditional usage of Lower/Early Cambrian and Middle Cambrian.
The Bourinot belt is located in the southern part of the Bras d’Or terrane of somewhat controversial relation to other peri-Gondwanan terranes but generally considered to be part of Ganderia (e.g., Barr et al. 1998; Hibbard et al. 2006). The Cambrian volcanic–sedimentary succession, consisting of the Bourinot Group and overlying MacMullin and MacNeil formations, is in faulted contact with older plutonic and metamorphic units. The basal unit of the Bourinot Group is the Eskasoni Formation, which consists largely of bimodal volcanic rocks, with compositions indicative of formation in an intra-continental rift-system (White et al. 1994). Its base is not known. The Northern Boisdale Hill volcanic unit of White et al. (1994) is also included here in the Eskasoni Formation. White et al. (1994) placed it in a separate unit because its U–Pb (zircon) age of 505 ± 3 Ma (Late Cambrian according to the time scale of Palmer 1983) appeared to be in conflict with the Middle Cambrian age of the Bourinot Group based on fossil evidence. However, that apparent conflict no longer exists because in the current time scale, 509 Ma is in the middle part of the Cambrian (e.g., Walker and Geissman 2009).
The Dugald Formation consists of siltstone with minor sandstone and wacke, and also contains tuffaceous layers. White et al. (1994) observed that its contact with the Eskasoni Formation is everywhere faulted. The Dugald formation locally contains abundant and diverse inarticulate brachiopods, including Acrothele prima Matthew, 1886 and Acrotreta gemmula (Matthew, 1894) (e.g., Matthew 1903; Walcott 1912). Other fossils include bradorids described in a series of papers by Matthew (e.g., Matthew 1903). In a recent revision, Siveter and Williams (1997) recognized Bradoria scrutator Matthew, 1899 and Indiana lippa Matthew, 1902, as the only legitimate species among those described from the Dugald Formation. Scarce trilobite material is known from the upper part of the Dugald Formation, consisting of Solenopleura bretonensis Matthew, 1903 and Andrarina linnarssoni bretonensis Hutchinson, 1952. The assemblage of brachiopods in the Dugald Formation has been considered Middle Cambrian (Walcott 1912), and the trilobites indicative of an early Middle Cambrian age (Hutchinson 1952).
The uppermost unit of the Bourinot Group, the Gregwa Formation, consists of lithic tuff and minor siltstone, shale, and volcanogenic conglomerate. No fossils have been reported from this formation. The overlying MacMullin Formation consists of interbedded micaceous siltstone, sandstone, and shale, locally with carbonate nodules. Trace fossils are common, and early Middle Cambrian trilobites have been reported from several localities (Hutchinson 1952). The MacNeil Formation is dominated by black shale. It yields Furongian (late Cambrian) trilobites and brachiopods (Hutchinson 1952).
Sample material and methods
The material studied here was collected by the authors in 2008 and 2009. The approximate sample locations are shown on Fig. 2, together with their stratigraphic context as established on the basis of earlier mapping by White et al. (1994). Specific sample locations and characteristics are summarized in Table 1.
From the collected material, rock samples of ca. 50 g were treated with standard palynological methods, mounted on glass slides with PetroPoxy resin, and studied under transmitted light with a ZeissAxio Imaginer M1 microscope with a computerized Axiocam Hrc microcamera. Figured and representative material are stored with the collections of the Nova Scotia Museum of Natural History, Halifax, Nova Scotia (numbers NSM010GF041.001–NSM010GF041.025).
Organic-walled microfossils in the Bourinot Group and MacMullin Formation
The majority of the collected samples yielded organic-walled microfossils. Three samples were barren (Bo09:4, Bo09:14, Bo09:20), and one sample yielded only filaments (Bo09:16) (Figs. 4C, 4D). The assemblages of acritarchs in the positive samples are discussed in approximate stratigraphic order, starting with the lowest sample. A comprehensive list of the taxa identified is given in Fig. 3.
Sample Bo09:21 (Figs. 4A, 5A, 5C, 5E, 5F, 6G) is from an isolated outcrop of micaceous siltstone and volcaniclastic material, in an area with heavy cover by soil and vegetation. White et al. (1994) identified a tectonic contact between the Eskasoni and Gregwa formations in this area, with the Dugald Formation faulted out. The assemblage of acritarchs suggests that this sample is the oldest in this study, and it likely represents sedimentary intervals in the uppermost part of the Eskasoni Formation. Of particular note for the age of the sample is the occurrence of Eliasum llaniscum Fombella, 1977, Heliosphaeridium notatum (Volkova) Moczydłowska, 1998 (Fig. 5A), and Retisphaeridium dichamerum Staplin, Jansonius and Pockock, 1965. This sample also contains poorly preserved material of Skiagia sp. (close to Skiagia cf. insignis of Downie 1982; Figs. 5E, 5F). The stratigraphically lowest occurrence of Eliasum llaniscum on a global scale is close to the traditional Lower–Middle Cambrian transition (Moczydłowska 1998). In Newfoundland, the lowest occurrence of Eliasum llaniscum is ca. 5 m above the base of the Chamberlains Brook Formation in strata assigned to the Paradoxides bennetti Zone. Heliosphaeridium notatum is common in strata assigned to the Protolenus Zone and the Paradoxides oelandicus Superzone on Baltica, and equivalent levels in Spain (Palacios and Moczydłowska 1998). The combined evidence suggests that this sample has a position close to the base of Cambrian series 3.
Samples Bo09:17, Bo09:18, and Bo09:19 (Figs. 4B, 5B, 5D, 5G, 8A, 8B) were collected from the lower part of the Dugald Formation on Dugald Brook. In common with sample 21, they contain Eliasum llaniscum and Retisphaeridum dichamerum (Figs. 8A, 8B). In addition, there are Comasphaeridium silesiense Moczydłowska, 1998 (Fig. 5B; in sample 17) and Heliosphaeridium serridentatum Moczydłowska, 1998 (Fig. 5D; in samples 17 and 18). Comasphaeridium silesiense has been considered to have a first appearance close to the base of the Middle Cambrian (Moczydłowska 1999). It was described from the middle Cambrian (oelandicus Superzone) of Silesia (Moczydłowska 1998), and it is also known from coeval strata in Sweden. Heliosphaeridium serridentatum was described from the middle Cambrian (oelandicus Superzone) of Silesia (Moczydłowska 1998). Sample 18 contains a form that Martin and Dean (1984) reported as “Acritarch gen. and sp. nov” from the upper part of Chamberlains Brook Formation and lower part of the Manuels River Formation in Newfoundland. Acritarch gen. and sp. nov. is one of the name-bearing forms for Microflora A0-1 (Eliasum jennessi and Acritarch gen. et sp. nov. assemblage) in Newfoundland, though it ranges into the Rugasphaera terranovana and Adara alea zones (Martin and Dean 1988). Additional common taxa of this zone are Retisphaeridum dichamerum and Eliasum llaniscum, both with relatively long stratigraphic ranges. In Spain, the IMC1 Zone is characterized by the presence of Comasphaeridium silesiense, Eliasum llaniscum, and Acritarch gen. and sp. nov. (Palacios 2008).
The acritarchs from samples Bo09:17, Bo09:18, and Bo09:19, suggest correlation with assemblage A0-1 in Newfoundland and the upper part of the IMC1 Zone in Spain. The absence of Cristallinium cambriense argues against correlation with the Rugasphera terranovana Zone in Newfoundland and the IMC2 Zone in Spain. The A0-1 assemblage corresponds to the eteminicus Zone, though it may extend to the bennetti Zone, as this zone is poorly characterized by acritarchs in Newfoundland.
Samples Bo09:7 and Bo09:8 (Figs. 6B, 6D–6F), from the Dugald Formation in Gregwa Brook, contain several specimens in common with samples Bo09:17 and Bo09:18, notably Acritarch gen. et sp. nov. (Figs. 6D, 6E), and Eliasum llaniscum (Fig. 6B). Sample 8 additionally contains Abacum normale Fombella, 1978 (Fig. 6F), a taxon known from the Oville Formation of northern Spain, where it has a long stratigraphical range (IMC2–IMC5 zones, Palacios 2010), and assemblage BB1 in the Booley Bay Formation of Ireland (Vanguestaine and Brück 2008).
The acritarchs from samples Bo09:7 and Bo09:8, suggest correlation with assemblage A0-1 in Newfoundland and the upper part of the IMC1 Zone in Spain. The absence of Cristallinium cambriense argues against correlation with the Rugasphaera terranovana Zone. However, the presence of Abacum normale could indicate correlation with the IMC2 Zone, or higher in Spain, although the absence in the assemblage of other diagnostic taxa of this zone may instead suggest an extended lower range of Abacum normale.
Sample Bo09:15, from the upper part of the Dugald Formation on Dugald Brook, contains a low diversity of acritarchs, which includes Multiplicisphaeridium parvum (Hagenfeldt) Moczydłowska, 1998, a species with heteromorphic processes (simple to branching) with a length about equal to the vesicle diameter. This species was first described from beds attributed to the oelandicus Superzone in Sweden (Hagenfeldt 1989). Multiplicisphaeridium parvum also occurs in the Nant-y-big Formation on the St. Tudwal’s Peninsula, where it was reported as Heliosphaeridium? llynense (Young et al., 1994) in beds attributed to the Tomagnostus fissus Zone (equivalent to the hicksi Zone, Fletcher 2007). Timofeevia tacheddirtensis Vanguestaine and van Looy, 1983, from Morocco is here considered to be flattened specimens of M. parvum. In Morocco, M. parvum occurs with Cristallinium cambriensis and Adara alea Martin in Martin and Dean, 1981 (Celtiberium cf. geminum in Vanguestaine and van Looy 1983, pl. 1, figs. 5, 6), suggesting correlation with the Adara alea Zone (and IMC3). The acritarchs in sample Bo09:15 do not provide fine biostratigraphical control.
Sample Bo09:13 (Figs. 6A, 6C, 7A–7F, 8C) from the Dugald Formation on Dugald Brook contains Multiplicisphaeridium parvum (Fig. 7), and several species diagnostic of the Rugasphera terranovana Zone (= Zone A0) in Newfoundland, corresponding to the hicksi Zone, with the appearance of Cristallinium cambriense (Fig. 8C) and Vulcanisphera lanugo (Fig. 6C) (Martin and Dean 1984, 1988). In northern Spain, the corresponding zone IMC2 has the first appearance of Cristallinium cambriense and is defined on the stratigraphic range of V. lanugo (Palacios 2008). Vulcanisphaera lanugo was originally described from the lower part of the Manuels River Formation, where it is restricted to the middle part of the hicksi Zone (Martin and Dean 1988). V. lanugo is also known from the lower part of the Playon Formation in southern Spain (Palacios et al. 2006) and the lower part of the Oville Formation of the Cantabrian Mountains, northern Spain, where it is associated with Eliasum asturicum Fombella, 1977, Cristallinium cambriense and Celtiberium dedalinum Fombella, 1978, in the Badulesia Zone (Palacios 2008, 2010). Vanguestaine and Brück (2008) reported V. lanugo from their BB1 assemblage of the Booley Bay Formation of Ireland, which they concluded to span the Ptychagnostus atavus, Hypagnostus parvifrons, and Ptychagnostus punctuatus zones, opening the possibility that the stratigraphic position of the Irish V. lanugo is somewhat higher than in Newfoundland. However, this occurrence is apparently based on a single specimen (Vanguestaine and Brück 2008, pl. 3, fig. 7), which has a morphology similar to Vulcanisphaera spinulifera. Eliasum asturicum (Fig. 6A) spans the IMC2 and IMC3 zones in northern and southern Spain (Palacios 2008). Sample Bo09:13 can be confidently correlated with the lower part of the Manuels River Formation and a relatively confined interval within the hicksi Zone, or more generally with zone IMC2 in Spain.
Samples Bo09:1, Bo09:2, Bo09:3, Bo09:5, and Bo09:6 (Figs. 8D–8F, 9A–9F, 10A–10C, 11A, 11B, 11E, 11H, 11K, 11L), from the lower part of the MacMullin Formation on Gregwa Brook, contain a diverse assemblage of acritarchs, with the occurrence of Aranidium granulatum Welsch, 1986 (Fig. 11E), Cristallinium dubium Volkova, 1990 (Figs. 8E, 8F), Symplassosphaeridium cambriense Slaviková, 1968 (Figs. 10A, 10B), Timofeevia lancarae (Cramer and Diez) Vanguestaine, 1978 (Figs. 9A, 9B), T. microretis Martin in Martin and Dean, 1983 (Figs. 11A, 11B), T. phosphoritica Vanguestaine, 1978, T. sp. A (Figs. 9C, 9D), T. sp. B (Figs. 9E, 9F), and Tubulosphaera sp. (Fig. 10C). The presence of A. granulatum in the MacMullin Formation is of particular interest. Aranidium granulatum has a polygonal to spherical vesicle with a circular pylome, and processes with conical bases. In northern Norway, this form is restricted to assemblage AI, occurring with, among others, Cristallinium ovillense, and a diverse assemblage of small acanthomorphic acritarchs (Welsch 1986). The AI assemblage lies within the Paradoxides paradoxissiumus Superzone and underlies a trilobite assemblage attributed to the Ptychagnostus puntuosus Zone (Nikolaisen and Henningsmoen 1990). A similar assemblage from the upper part of the paradoxissimus Superzone defines the SK2a Zone in the East European Platform (Volkova and Kir’yanov 1995), which contains several species of Aranidium and which precedes levels with the first Cristallinium dubium and various species of Timofeevia. The first occurrence of A. granulatum and Cristallinium dubium in northern Spain is in the upper part of the IMC5 Zone (Palacios 2008, 2010). Symplassosphaeridium cambriense, found in samples Bo09:5 and Bo09:6 occurs with Tubulosphaera sp. in the Oville Formation, northern Spain, a short stratigraphical distance below an occurrence of Paradoxides cf. davidis (Sdzuy 1961) in the lower part of IMC5 Zone (Palacios 2008, 2010). In the Mira River area, S. cambriense occurs with C. dubium in an interval that contains Paradoxides davidis (Palacios et al. 2009). Samples Bo09:1, Bo09:2, Bo09:3, Bo09:5, and Bo09:6, contain diverse species of Timofeevia. However, the identification and stratigraphic range of several species of Timofeevia need restudy (cf. Palacios et al. 2009). For example, specimens of Timofeevia lancarae, figured by Martin and Dean (1981, 1988) from Newfoundland and Welsch (1986) from northern Norway, do not correspond to the type material of this species illustrated by Cramer and Diez (1972). The specimens here identified as T. lancarae (Figs. 9A, 9B) are only those materials that share diagnostic characteristics with the type specimens described by Cramer and Diez (1972) from the Oville Formation in the type area in the Cantabrian Mountains, northern Spain. Timofeevia sp. A (Figs. 9C, 9D) is a species with short conical processes. Identical forms have been referred to in the literature as T. phosphoritica, T. manata, and T. janishewskyi, for example from the Sosinsk Formation by Dean et al. (1997; Figs. 9A, 9B). However, the type material of T. phosphoritica and T. manata have tubular processes, and T. phosphoritica exhibits rupture into paraplates. Timofeevia sp. B (Figs. 9E, 9F) is distinguished from T. lancarae by a greater number of processes. In northern Spain, T. sp. B, has a long stratigraphic range (IMC4–IMC6; Palacios, unpublished observations). These two forms of Timofeevia represent new species that will be formally described in a future publication. The assemblage of acritarchs in these samples suggest correlation with the IMC5 Zone in Spain.
Samples Bo09:9, Bo09:10, and Bo09:11 (Fig. 11C) from the MacMullin Formation on Indian Brook contain relatively poorly preserved acritarchs, but the combined presence of Timofeevia lancarae, T. phosphoritica (Fig. 11C), Cristallinium cambriense, C. dubium, and Symplassosphaeridium cambriense indicates correlation with the IMC5 Zone in Spain and is closely comparable to the assemblage of acritarchs found close to the transition between the Trout Brook and MacLean Brook formations in the Mira River area (Palacios et al. 2009), with a position close to the transition between the Paradoxides davidis and Paradoxides forchhammeri zones. The samples from the MacMullin Formation are indicative of the upper part of the davidis Zone, to the lower part of the forchhammeri Zone.
Samples Geo08:1, Geo08:2, and Geo08:3 (Figs. 11D, 11F, 11G, 11I, 11J) from the MacMullin Formation in the northern part of the Bourinot Belt include taxa also reported from the middle and upper part of the MacLean Brook Formation in the Mira River area, such as Stelliferidium magnum Palacios in Palacios et al., 2009 (Fig. 11F) and Petaloferidum lacrimiferum Palacios in Palacios et al., 2009 (Figs. 11I, 11J) (Palacios et al. 2009 and unpublished observations). This assemblage indicates correlation with the IMC6 Zone in Iberia, which is characterized by the first appearance of Stelliferidium magnum, Timofeevia microretis, and T. phosphoritica (Palacios 2010).
Age of the Bourinot Group and MacMullin Formation
The assemblages of acritarchs described in the preceding section provide new constraints for the age of the Bourinot Group and MacMullin Formation (Fig. 12). The most specific pronouncements on these Cambrian strata are those of Hutchinson (1952), Landing (1996, p.43, fig. 5), and Landing and Westrop (1998, fig. 20). Landing (1996) assigned the Dugald Formation to the Hartella bucculenta Zone (= Paradoxides eteminicus Zone), with the Eskasoni Formation occupying the upper part of the bennetti Zone, the Gregwa Formation assigned to the hicksi Zone, and the MacMullin Formation spanning the top of the hicksi Zone to the Homagnostus obesus Zone. This largely coincides with the age assignments given by Hutchinson (1952), except that he restricted the MacMullin Formation to the hicksi Zone. Landing’s (1996) extended range for the upper part of the MacMullin Formation was on the assumption that it is equivalent to the MacLean Brook Formation in the Mira River area. Landing (1996, p. 43) suggested that the volcanic rocks of the Eskasoni Formation correlate with volcanic rocks in the Chamberlains Brook Formation at Trinity Bay, southeastern Newfoundland. The data from acritarchs presented here largely support these assignments, but also leaves open the possibility that the Eskasoni Formation includes older volcanic rocks, and it brings into question the age of trilobite-bearing portions of the MacMullin Formation.
Highly relevant to the age question of the Eskasoni Formation are the published U–Pb (zircon) dates from the northern part of the Bourinot belt. Rhyolite on Long Island likely belonging to the Eskasoni Formation yielded an age of 505 +/– 3 Ma (White et al. 1994). A syenogranite sample from the nearby Mount Cameron pluton yielded an age of 509 +/– 2 Ma. Petrological similarities suggest that the rhyolite and syenogranite are likely co-magmatic and their U–Pb ages overlap within error. These ages of 509–505 Ma are close to recent estimates for the base of Cambrian series 3 at ca. 510 Ma (Fig. 12). Hutchinson (1952) inferred the transition between the Eskasoni and Dugald formations to be gradual because of shale intervals in the Eskasoni Formation with early Middle Cambrian brachiopods comparable to those in the Dugald Formation. Although its position is somewhat uncertain because it is from an isolated outcrop, it is probable that sample Bo09: 21 is from such a shaly interval near the top of the Eskasoni Formation. A sample (Bo09:20) of cleaved siltstone from the upper part of the Eskasoni Formation along Dugald Brook was processed for organic-walled microfossils but found barren. The combined data from geochronology and acritarchs suggest that the Eskasoni Formation extends to at least the lower part of Cambrian stage 5 (Fig. 12), but the base of the formation is not known. Landing et al. (2008, p. 889) cited Barr et al. (1996) for a U–Pb date of ca. 522 Ma for the lower part of the Eskasoni Formation, but no such data were presented in that publication, and hence the validity of the 522 Ma date is uncertain.
The acritarch assemblage from the lower part of the Dugald Formation agrees with Landing’s (1996) assignment to the eteminicus Zone (although not excluding the possibility that it enters the bennetti Zone) and temporal correlation to the upper part of the Chamberlains Brook Formation. However, the highest sample in the Dugald Formation on Dugald Brook contains an assemblage clearly correlated with the hicksi Zone. This sample has a position close to the Dugald–Gregwa transition. Of particular note is the presence in this sample of Vulcanisphaera lanugo, also known from the lower part of the Manuels River Formation in Newfoundland. A scarce trilobite faunule was reported 3.6 m below the Dugald–Gregwa contact on Dugald Brook, comprising Solenopleura bretonensis and Andrarina linnarssoni bretonensis (Matthew 1903; Hutchinson 1952; Geological Survey of Canada (GSC) 18583 on Fig. 2). Neither of these taxa has been reported outside of the area. Hutchinson (1952) remarked that Solenopleura rushtonensis Cobbold, 1934, from the Upper Comley Sandstone of Shropshire, England, may be a junior synonym of Solenopleura bretonensis, and Landing (2006, p. 43) built on this suggestion, when he used the co-occurrence of Solenopleura rushtonensis with Ptychagnostus gibbus as evidence to assign the Dugald Formation to the Hartella bucculenta Zone. Rushton et al. (2007, p. 137), on the other hand, rejected the synonymy of Solenopleura bretonensis and S. rushtonensis. The precise stratigraphic position of the acritarch sample relative to the trilobite-bearing horizon is not known but is assumed to approximate it. Based on the assemblage of acritarchs, the upper part of the Dugald Formation extends to the hicksi Zone and correlates temporally with the lower part of the Manuels River Formation.
Hutchinson (1952) interpreted the lower contact of the MacMullin Formation to represent a disconformity in its northern area of exposure in the Bourinot belt because of greatly reduced thickness, although this may also be the result of faulting (White et al. 1994). In the Indian Brook area the contact has been interpreted as being conformable, although the contact is generally poorly exposed (Hutchinson 1952, pp. 24–25). Acritarchs from the lower part of the MacMullin Formation on Gregwa Brook suggest correlation with strata close to the transition of the davidis and forchhammeri zones, or in the forchhammeri Zone. Comparison with the acritarch record from Newfoundland in much of the davidis and forchhammeri zones is complicated because of poor preservation, intervals with missing data, and a likely hiatus that may span a substantial part of these zones in Newfoundland.
Hutchinson (1952, pp. 25–26) reported an assemblage of trilobites from sandy shales of the MacMullin Formation on Indian Brook (locality GSC 18572, see Fig. 2A), with Acrocephalops matthewi Hutchinson, 1952, Holasaphus centropyge Matthew, 1895, Cotalagnostus barrandei (= Lejopyge barrandei), and Bailaspis sp., that he attributed to the hicksi Zone, largely on account of the identification of Lejopyge barrandei (Hicks, 1872). Lejopyge barrandei is known from the lower part of the Manuels River Formation at Manuels River (Howell 1925; Landing and Westrop 1998; Fig. 7), where it occurs in the hicksi Zone, including the bed that yields the main record of Vulcanisphaera lanugo (bed 51 of Howell 1925). Howell’s (1925) highest citation of Lejopyge barrandei in the Manuels River section coincides with the top of the Rugasphera terranovana Zone. The ages indicated by acritarchs from the lower part of the MacMullin Formation on Gregwa Brook seem inconsistent with the reports of hicksi Zone trilobites. Hutchinson (1952) noted that the contact between the Gregwa and MacMullin formations on Grewa Brook is separated by a 30 cm thick zone of deeply weather material of uncertain composition. This zone may indicate that the contact here is faulted and that the lowest part of the MacMullin Formation is missing in this section. However, the succession of Cambrian strata between Gregwa Brook and Dugald Brook is represented by a truncated inverted limb that is younging in a southeasterly direction (White et al. 1994; Fig. 7A). The location of trilobite locality GSC 18572 suggests that it belongs to a higher stratigraphic interval than the acritarch samples on Gregwa Brook. It is possible, therefore, that trilobites from locality GSC 18572 are younger than previously thought and that their identification needs to be revised. The record of Lejopyge barrandei in this locality is based on a single poorly preserved cephalon and is doubtful (P. Ahlberg, personal communication, 2010). Acrocephalops matthewi and Holasaphus centropyge are both unique to the MacMullin Formation. Dean (1972) described a second species of Holasaphus, H. mesopotamicus from the upper part of the Sosinsk Formation of Turkey. Holasaphus mesopotamicus occurs in an interval that has yielded Timofeevia lancarae and forms transitional between T. lancarae and T. phosphoritica (Dean et al. 1997). Both the acritarch assemblage and relationship to other trilobites in the Sosink Formation show that H. mesopotamicus is late middle Cambrian (Dean et al. 1997). Dean (1972, p. 277) remarked that “The similarities between H. centropyge and H. mesopotamicus are sufficiently close to attempt one into postulating a similar age for the 2 species, but the stratigraphical evidence is not conclusive.” The data presented here on acritarchs support the possibility that Holasphus centropyge and H. mesopotamicus are coeval. It should also be noted that Dean (1982) remarked on the strong resemblance to H. mesopotamicus of a cranidium from the basal portion of the Elliot Cove Formation in beds of the Lejopyge laevigata Zone, in the Manuels River section, that had been previously recorded as Andrarina costata (Angelin, 1854).
A richer and better preserved trilobite assemblage has been reported from the MacMullin Formation in the northern part of the Bourinot belt in limestone concretions on the shore of St. Andrews Channels (GSC 18570, Fig. 2B), where Holasaphus centropyge is recorded with Paradoxides abenacus Matthew, 1886, Acadagnostus acadicus Hartt, 1866, Ptychoparia bretonensis Hutchinson, 1952, and Acrocephalops matthewi (Hutchinson 1952). Hutchinson (1952) assigned this assemblage to the hicksi Zone. The identification of Acadagnostus acadicus was considered questionable by Robison (1995), thus making the identification of Paradoxides abenacus critical. Hutchinson (1952, p. 76) considered differences between the Cape Breton Island material and the type material from New Brunswick to be the result of differences in preservation. Nevertheless, the lack of ridges on the front of the glabella in the MacMullin Formation material suggests that the assignment to P. abenacus is uncertain. In view of the data presented here, a new look at the identification of the MacMullin Formation Paradoxides is warranted. The acritarch assemblages recovered in the MacMullin Formation in the northern part of the Bourinot belt come from outcrops (Fig. 2B) that yield acritarchs of the forchhammeri Zone. Landing (1996, p. 43) also noted hicksi Zone trilobites in what he described as a Manuels River Formation-like tongue of dark shale and limestone nodules in the lower MacMullin Formation, but without giving any details on the taxa identified.
No acritarchs of the Adara alea Zone were recovered during this study. Adara alea has a narrow stratigraphic range in all of its occurrences and may represent a short time interval. In Newfoundland, the Adara alea range Zone (= A1 Zone (Adara alea – Eliasum llaniscum assemblage)) spans the upper part of the hicksi Zone in the Manuels River Formation and questionably enters the davidis Zone. In Spain, IMC3 is a corresponding zone defined on the range of Adara alea and approximates the Pardailhania Zone of the middle Caesaraugustian. Also missing from the MacMullin Formation (and from Newfoundland) are taxa characteristic of the IMC4 Zone of Spain, such as Adara matutina Fombella, 1977, Celtiberium geminum Fombella, 1977, and Timofeevia raquelinae (Cramer and Diez) Cramer and Diez, 1979 (Palacios 2010).
The data from acritarchs indicate that the possibility of a disconformity at the base of the MacMullin Formation (Hutchinson 1952) needs further consideration. Future sampling along trilobite-bearing strata at St. Andrews Channel and Indian Brook, and on MacMullin Brook, which has the best exposed transition between the Gregwa and MacMullin formations in the Indian Brook area, may shed additional light on the conflicting data from trilobites and acritarchs. A sample from the MacNeil Formation on Indian Brook (Bo09:12) was processed for organic-walled microfossils but found to be barren.
Implications for the age of the British Acrothele prima Shale
The Dugald Formation has been for long time a reference for the age of the Acrothele prima Shale of the Wrekin area, Shropshire, England. The Acrothele prima Shale is based on material from a temporary excavation (Cobbold and Pocock 1934), and its placement within the regional stratigraphy has been uncertain. Both the Dugald Formation and Acrothele prima Shale contain the brachiopods Acrothele prima and Acrotreta gemmula, and the bradorid arthropod Bradoria scrutator. The Acrothele prima shale was long thought to be Lower Cambrian following Ulrich and Bassler’s (1931) assignment of the Dugald Formation fossils to the Lower Cambrian, and attributed to the Lower Comley Sandstone. This error was only recently corrected, and the Acrothele prima Shale is now placed in the Middle Cambrian Upper Comley Sandstone (Williams and Siveter 1998; Rushton et al. 2007). Matthew (1903) described a great number of species of Bradoria from the Dugald Formation on Dugald Brook, but Siveter and Williams (1997) considered all to be synonyms of Bradoria scrutator. In the section on Dugald Brook, Bradoria scrutator is known from Matthew’s (1903) divisions E1b to E3f, thus spanning most of the Dugald Formation. Bradoria scrutator spans several of the acritarch samples examined here (with certainty Bo09:17, Bo09:18, and Bo09:19), confirming earlier assignements of this bradorid to the eteminicus Zone, whereas the stratigraphically highest occurrences may be in the hicksi Zone. Under the assumption of a similar stratigraphical range for this species in its different occurrences, the data presented here on organic-walled microfossils support the assignment of the Acrothele prima Shale to the gibbus, or possibly fissus, Zone (Williams and Siveter 1998; Rushton et al. 2007).
Regional context of the Bourinot Group
The data presented here on organic-walled microfossils from the Bourinot belt provide tighter biostratigraphical age control on the Cambrian sedimentary succession than was previously available (Hutchinson 1952) and broadly support the age relations suggested by Landing (1996, p. 43, fig. 5) and Landing and Westrop (1998, fig. 20). Thus, volcanic rocks of the Eskasoni Formation may be coeval with volcanism in the Chamberlains Brook Formation at Trinity Bay (but could be older) and tuffaceous material in the Gregwa Formation with ash beds in the Manuels Rivers Formation. Landing et al. (2008) reported Bradoria sp. cf. B. scrutator from a Cambrian succession with abundant volcanic material at Beaver Harbour, southern New Brunswick (Fig. 1, inset), and considered this observation their best available biostratigraphical constraint on shales that follow upon volcanic rocks of their Wades Lane Formation. We sampled the Cambrian successions in the Beaver Harbour area for organic-walled microfossils, but all samples were barren. However, if the correlation potential of Bradoria scrutator is accepted, the stratigraphic position of volcanic rocks in the Wades Lane Formation is consistent with their being coeval with the Eskasoni Formation. Inferences on the proximity of these areas on the basis of volcanic activity is problematic, as volcanic activity was widespread along the northern margin of Gondwana during this period of time as part of rifting related to the incipient stages in the opening of the Rheic Ocean (e.g., Pollock et al. 2009; Nance et al. 2010). For example, extensive deposits of pyroclastic rocks, and other volcanic material, including ignimbrite, formed in the same time period in the transition from the Vallehondo to Playon formations in the Ossa Morena Zone of southern Spain (e.g., Etxebarria et al. 2006). As presently understood, acritarchs do not show palaeogeographical differentiation in the Cambrian, and so cannot help to resolve the relative positions of the various peri-Gondwanan terranes. This study, however, is an additional example of the widespread distribution and biostratigraphical utility of acritarchs in the Cambrian series 3 across a region broadly coinciding with the Acado-Baltic faunal zone.
TP and SJ acknowledge funding from the Spanish Ministry of Science and Innovation, through grant CGL-2008-04373 (co-financed by FEDER). SMB’s work in Cape Breton Island is supported by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada. We thank the journal referees Paul Strother and Stewart Molyneux and Associate Editor Brendan Murphy for suggestions that improved the manuscript. Museum numbers were provided by Deborah Skilliter, Nova Scotia Museum of Natural History, Halifax, N.S.
↵1 This article is one of a series of papers published in CJES Special Issue: In honour of Ward Neale on the theme of Appalachian and Grenvillian geology.
- Received October 3, 2010.
- Accepted January 28, 2011.
- Published on the NRC Research Press Web site at http://cjes.nrc.ca on December 22, 2011.
- Published by NRC Research Press