2.a. Foraminifers at the base of the Livian Substage in Belgium

The stratotype for the Livian base was selected in the transition between the carbonates of the Neffe and Lives formations (Conil et al. Reference Conil, Groessens and Pirlet1977), although an argillaceous bentonite of volcanosedimentary origin is located at the boundary, regionally named as the Banc d’Or de Bachant (Fig. 1), considered as the transition between the Cf4 and Cf5 chronozones. More recent studies (Hance et al. Reference Hance, Poty and Devuyst2006; Poty et al. Reference Poty, Devuyst and Hance2006, Reference Poty, Aretz and Hance2014; Poty, Reference Poty2016) have followed the lithostratigraphical subdivision proposed by Conil et al. (Reference Conil, Groessens and Pirlet1977), and new Mississippian foraminiferal zones were defined (MFZ) to standardize more widely distributed foraminiferal guides (Poty et al. Reference Poty, Devuyst and Hance2006). For these authors, some of the guide foraminifers defined for the subchronozones in Conil et al. (Reference Conil, Longerstaey and Ramsbottom1980, Reference Conil, Groessens, Laloux, Poty and Tourneur1991) had too much local character, and could not be identified outside of their influence areas in Belgium.

Fig. 1. Biostratigraphic sketch of the Moliniacian–Livian boundary interval stratotype at Lives section in Belgium. Data source: Poty & Hance (2006). Diagonal striped lines have no foraminiferal record because of unfavourable facies or a gap.

Since the studies of Conil et al. (Reference Conil, Pirlet and Lys1967, Reference Conil, Groessens and Pirlet1977), the guides for the base of the Livian, V2b or Cf5 zones were considered to be Pojarkovella nibelis (Durkina, Reference Durkina1959) and Koskinotextularia (named as Cribrostomum in 1967 or cribrate palaeotextulariids in 1977). Later, Archaediscus at concavus stage was also recognized as an auxiliary marker (Conil et al. Reference Conil, Longerstaey and Ramsbottom1980, Reference Conil, Groessens, Laloux, Poty and Tourneur1991), as well as more questionable or regional taxa such as primitive ‘Millerella’, Rhodesinella, Lituotubella, Nodosarchaediscus, Omphalotis minima (Rauzer-Chernousova & Reitlinger in Rauzer-Chernousova et al. Reference Rauzer-Chernousova, Belyaev and Reitlinger1936), Consobrinellopsis consobrina (Lipina, Reference Lipina1948) and Endothyranopsis (e.g. Conil, Reference Conil, Wagner and Holub1977; Conil et al. Reference Conil, Longerstaey and Ramsbottom1980, Reference Conil, Groessens, Laloux, Poty and Tourneur1991; Poty et al. Reference Poty, Devuyst and Hance2006). The evolutionary stages of the genus Archaediscus involving the shape of the sutures between the whorls, involutus, concavus, angulatus and tenuis were defined by Pirlet & Conil (Reference Pirlet and Conil1974) and later modified by Conil et al. (Reference Conil, Longerstaey and Ramsbottom1980) and Vachard (Reference Vachard1988a). From a taxonomical point of view they have no species value, but the presence of this character in a group of species of this genus has been used as a biostratigraphic tool in numerous studies in Europe (Conil et al. Reference Conil, Groessens and Pirlet1977, Reference Conil, Longerstaey and Ramsbottom1980; A.R.E. Strank, unpub. PhD thesis, University of Manchester,1981; Somerville, Reference Somerville2008) and in Asia (e.g. Hance et al. Reference Hance, Hou and Vachard2011; Zandkarimi et al. Reference Zandkarimi, Vachard, Najafian, Hamdi and Mosaddegh2017, Reference Zandkarimi, Vachard, Najafian, Mosaddegh and Ehteshami-Moinabadi2018). Another important character within the Archaediscus is the progressive reduction of the microgranular layer during its evolution, which led to Brenckle et al. (Reference Brenckle, Ramsbottom and Marchant1987) to consider the distinction between the genus Paraarchaediscus Orlova, Reference Orlova1955 (for those forms with well-developed microgranular layer) from Archaediscus Brady, Reference Brady1873 (for those forms with reduced microgranular layer) valid. This is a continuum process where the end-members are easily distinguished, whereas there are a multitude of intermediate forms in which it is difficult to decide if they belong to one or the other genus due to the ambiguity in the boundary between both genera. All the species recorded in the lower part of the middle Viséan substage would correspond to the genus Paraarchaediscus but, as a result of uncertainty and questionable taxonomic value of the genus Paraarchaediscus, ‘Archaediscus’ is used here. However, when data are based on previous studies by other authors, we have preserved the original notation as Archaediscus or Paraarchaediscus. The first occurrence of Pojarkovella nibelis is some distance above the base of the Lives Formation (at a level 14.3 m above the base, in the 30-m-thick basal Haut-le-Wastia Member in the Livian stratotype section; Fig. 1); in other regions of Belgium, this species first occurs much higher compared with the base of this substage (Poty & Hance, Reference Poty and Hance2006; Poty et al. Reference Poty, Aretz and Hance2014). Koskinotextularia occurs later than P. nibelis (Poty et al. Reference Poty, Devuyst and Hance2006). These authors also defined the rugose coral assemblage (RC5γ) for the Haut-Le-Wastian Member, characterized by the first occurrence of Siphonodendron irregular (Phillips, 1936), Axophyllum vaughani (Salée, Reference Salée1913), Caninophyllum archiaci (Milne-Edwards & Haime, Reference Milne-Edwards and Haime1852) and Clisiophyllum garwoodi (Salée, Reference Salée1913) (Fig. 1). Unfortunately, the first occurrence of this assemblage was situated ‘in the first open-marine bioclastic limestone’ and not from the base of the unit, which is mostly composed of massive limestones, intraclastic limestones or evaporitic limestones with stromatolites (Hance, Reference Hance1979).

Some of the previously described auxiliary foraminiferal guides for the Livian Substage are now confidently recognized as first occurring in the lower Viséan (Moliniacian) substage, such as Nodosarchaediscus, Omphalotis minima and Consobrinellopsis consobrina; in the most recent revision, they are considered as abundant species in the Livian assemblages (Poty et al. Reference Poty, Devuyst and Hance2006).

The Banc d’Or de Bachant is a volcanosedimentary bentonite, 10–30 cm thick with local pedogenic features, and is recognized as representing a gap in limestone deposition (erosion and/or hiatus) (Poty & Hance, Reference Poty and Hance2006). This Banc d’Or de Bachant is recognized in all the regions in Belgium and northern France, and is considered as the base of the sequence 7 (Hance et al. Reference Hance, Poty and Devuyst2001). Other authors have correlated this gap of the Banc d’Or de Bachant with similar rocks in Germany and England (A. Huyghe-Buissart, unpub. PhD thesis, University of Lille, 1993), although the stratigraphic position for this gap in England, at least, does not seem to correspond exactly to the position in Belgium. The presence of this hiatus has been criticized by some authors (e.g. Kalvoda, Reference Kalvoda2002; Hance et al. Reference Hance, Poty and Devuyst2006; Poty, Reference Poty2016), who have recommended the search for a new stratotype where there is a record of continuous sedimentation. Taking into consideration the studies in Britain (Conil et al. Reference Conil, Longerstaey and Ramsbottom1980), the Belgian subdivision can also be applied there. The hiatus between the Cf4 and Cf5 zones has been traditionally considered as negligible and, consequently, the possible influence and/or implication of the Banc d’Or de Bachant overlooked. Above the bentonite horizon, the Haut-le-Wastia Member is composed predominantly of lime mudstones, with stromatolites and local breccias due to evaporite dissolution. The lower part of the Lives Formation and the Banc d’Or are therefore not the most favourable facies for the recovery of foraminifers (Fig. 1). There is a 30-m-thick thick interval where the knowledge of typical lower and middle Viséan microfossils is poorly known, and even the positioning of the base of the middle Viséan substage at the base of the Banc d’Or de Bachant is not supported by any fossil guide.

2.b. The Arundian–Holkerian boundary interval in northern England

The stratotype for the base of the Holkerian Substage is situated at Barker Scar near Holker Hall in south Cumbria, northern England. The boundary was located at the contact between the Dalton Beds and the overlying Park Limestone, as originally defined by George et al. (Reference George, Johnson, Mitchell, Prentice, Ramsbottom, Sevastopulo and Wilson1976). The base of the Holkerian Substage was based on macrofossils, such as brachiopods (Davidsonina carbonaria (McCoy, Reference McCoy1852), Composita ficoidea (Vaughan, Reference Vaughan1903) and Linoprotonia corrugatohemispherica (Vaughan, Reference Vaughan1903)) and rugose corals (Axophyllum vaughani and Lithostrotion minus (McCoy, Reference McCoy1849)), as well as microfossils, the foraminifers Archaediscus at concavus stage and double-walled palaeotextulariids. However, in Barker Scar only Linoprotonia corrugatohemispherica occurs at the base of the Park Limestone (Rose & Dunham, Reference Rose and Dunham1977). These authors also considered Axophyllum cf. vaughani as typical of the lower beds in the Park Limestone, whereas they considered that Davidsonina carbonaria, absent from Barker Scar, occurs at equivalent levels elsewhere in England. In contrast, Composita ficoidea is commonly recorded in the upper part of the Dalton Beds (Fig. 2; beds e and i in Rose & Dunham, Reference Rose and Dunham1977). The double-walled palaeotextulariids is readily a misidentification, because they have never been recorded below the upper Viséan strata anywhere within the Palaeotethys; most probably, they were referring to the single-layered palaeotextulariids.

Fig. 2. Biostratigraphic sketch of the Arundian–Holkerian boundary interval stratotype at Barker Scar (Cumbria) and lower part of the Silverdale Borehole (Askrigg) in northern England. Bed thickness of the Barker Scar section according to Rose & Dunham (Reference Rose and Dunham1977) and Silverdale Borehole sensu Waters et al. (Reference Waters, Cózar, Somerville, Haslam, Millward and Woods2017). Base of the Park Formation according to Johnson et al. (Reference Johnson, Soper and Burgess2001). Diagonal striped lines have no foraminifers because of gaps.

Ramsbottom (Reference Ramsbottom1981) considered the same macrofossils as markers for the base of the Holkerian Substage and specified the foraminiferal list of potential markers. He considered the main guides as Koskinotextularia and Pojarkovella nibelis, but also mentioned Archaediscus at concavus stage and transitional forms to the angulatus stage, as well as the regional markers Holkeria and Dainella holkeriana Conil & Longerstaey in Conil et al. (Reference Conil, Longerstaey and Ramsbottom1980).

However, the conclusions in A.R.E. Strank (unpub. PhD thesis, University of Manchester, 1981), who also contributed in Ramsbottom (Reference Ramsbottom1981), were not exactly the same. She considered that the Archaediscus at concavus stage are not well-developed in England as in other basins, and that these forms appeared inconsistently from the Arundian to the Asbian substages. In her study, it is possible to find Omphalotis minima and Lituotubella from the base of the Barker Scar section (> 29 m below the boundary between the Dalton Beds and Park Limestone) and Archaediscus at concavus stage (A. varsanofievae Grozdilova & Lebedeva, Reference Grozdilova and Lebedeva1954 and A. stilus amplus Conil & Lys, Reference Conil and Lys1964) nearly from the base of the section (> 23 m below the boundary), both from bed a (Fig. 2). On the other hand, she recorded Consobrinellopsis (as Palaeotextularia sp.) in bed k (2.5 m above the boundary), and also Koskinotextularia and Pojarkovella nibelis in bed k (3.5 m above the boundary; Fig. 2). It can therefore be concluded that Omphalotis minima, Lituotubella and ‘Archaediscus’ at concavus stage first occur much earlier than Koskinotextularia and P. nibelis. Unfortunately, the Barker Scar section does not extend down far enough into the Arundian Substage as to allow more detailed first occurrences of the foraminifers to be recorded during the early Viséan Age; the Dalton Beds (Formation) reaches up to 255 m thick, but only the upper c. 30 m is exposed at Barker Scar.

Riley (Reference Riley1993) questioned the validity of the Barker Scar section as a stratotype because lateral sections with palaeosols (2 km apart) were correlated with bed i in Barker Scar where some sandy strata are recorded, and compared with the Davidsonina carbonaria beds of Garwood (Reference Garwood1913). He assumed a non-sequence zone in the lower part of the succession, also supported by the absence of the transition from fasciculate (Siphonodendron) to cerioid (Lithostrotion) rugose corals, as recorded in Ireland and Wales.

Johnson et al. (Reference Johnson, Soper and Burgess2001) revised the stratigraphy in the region and, at Plumpton quarries, they described the topmost 1.8 m of the Dalton Formation (lithological units were described for first time as formations in this work) as being composed of compact, pale brown, fine-grained dolomites and sandstones with traces of carbonaceous rootlets, interpreted as palaeosols. This confirms the non-sequence in the Dalton Formation, but not in the stratotype section at the base of the Park Limestone Formation. In Barker Scar, they considered that beds e–i contained similar sandy and partly dolomitized characteristics of the upper part at Plumpton, including the non-sequence interval (Fig. 2). Furthermore, they described bed j as a grey, dolomitized, bioclastic packstone with quartz grains, and compared it with the Park Limestone Formation, but less darker. They recognized this bed as a mappable horizon and, although no diagnostic fossils were recorded, they considered bed j as the base for the Park Limestone Formation (Fig. 2).

This lithological change at the base of the Park Limestone Formation was adopted by Waters et al. (Reference Waters, Dean, Jones, Somerville, Waters, Somerville and Jones2011, Reference Waters, Cózar, Somerville, Haslam, Millward and Woods2017), although the stratigraphic level for the base of the Holkerian Substage (with diagnostic fossils) was certainly not modified by Johnson et al. (Reference Johnson, Soper and Burgess2001) from the original position at the base of bed k (Fig. 2).

In the Askrigg Block, the lateral equivalent to the succession in South Cumbria is composed of the Chapel House Limestone and Kilnsey formations (Arthurton et al. Reference Arthurton, Johnson and Mundy1988). The Kilnsey Formation is subdivided into the Scaleber Force Limestone and Scaleber Quarry Limestone members, and the Arundian–Holkerian boundary is supposedly located at the boundary between both members. Foraminifers in the Scaleber Force Limestone Member are typically Conilidiscus bucculentus (Conil & Lys, Reference Conil and Lys1964), C. settlensis Conil in Conil et al. (Reference Conil, Longerstaey and Ramsbottom1980), Uralodiscus sp., Planoarchaediscus and Glomodiscus miloni (Pelhâte, Reference Pelhâte1967), an assemblage assigned to the Arundian (Strank in Arthurton et al. Reference Arthurton, Johnson and Mundy1988). Important macrofauna in the Scaleber Quarry Limestone Member are Lithostrotion vorticale (Parkinson, Reference Parkinson1808), Composita ficoidea, Davidsonina carbonaria, Linoprotonia corrugatohemispherica associated with small ‘concavus stage and partial angulatus stage Archaediscus and Nodosarchaediscus’, as well as Koskinotextularia cribriformis and Pojarkovella nibelis (Arthurton et al. Reference Arthurton, Johnson and Mundy1988). Fossil assemblages can be readily compared with that of the Barker Scar section as well as some of the macrofauna, and the base of the Scaleber Quarry Limestone Member can therefore be correlated with the base of the Park Limestone Formation.

The occurrence of the non-sequence interval in the upper part of the Chapel House Limestone Formation and Archaediscus at concavus stage from the base of the Kilnsey Formation led Waters et al. (Reference Waters, Cózar, Somerville, Haslam, Millward and Woods2017) to correlate the base of the latter with the base of the Park Limestone Formation. At the preserved base of the Chapel House Limestone Formation in the Silverdale Borehole (this formation has a maximum thickness of 12 m; Murray, Reference Murray1983), the assemblage is composed of numerous typical upper Tournaisian and lower Viséan forms (Caligella, Dainella, Eosinopsis, Septabrunsiina, Eoparastaffella, Septatournayella, Eblanaia, Spinobrunsiina, Tournayellina, and Spinochernella, as well as Glomodiscus sp., G. miloni, Uralodiscus adindanii Brenckle & Marchant (Reference Brenckle and Marchant1987), U. elongatus (Conil & Lys, Reference Conil and Lys1964), U. rotundus (Chernysheva, Reference Chernysheva1948), and Conilidiscus settlensis) which, in addition to Latiendothyranopsis menneri solida (Conil & Lys, Reference Conil and Lys1964), Nodosarchaediscus, Plectogyranopsis moraviae (Conil & Longerstaey in Conil et al. Reference Conil, Longerstaey and Ramsbottom1980) and P. settlensis (Conil & Longerstaey in Conil et al. Reference Conil, Longerstaey and Ramsbottom1980)), allow the assemblage to be assigned to the Cf4δ subzone (Figs 2, 3). The upper part of the formation is dolomitized and does not contain foraminifers, and there is a non-deposition and development of a palaeosol (Fig. 2). At the base of the overlying Kilnsey Formation, Archaediscus at concavus stage, Omphalotis minima and Pojarkovella sp. were recorded. Specimens of Pojarkovella sp., Holkeria? and Ninella are recorded 7 m above the base, and typical Pojarkovella nibelis, Koskinotextularia, Archaediscus concavus trans. angulatus stage and Endothyranopsis compressa (Rauzer-Chernousova & Reitlinger in Rauzer-Chernousova et al. Reference Rauzer-Chernousova, Belyaev and Reitlinger1936) occur 2 m above (Fig. 2). In higher levels of the formation, and occurring in several steps, are Archaediscus at angulatus stage, Pseudoendothyra and Endothyranopsis crassa (Brady, Reference Brady1876) as well as Koskinobigenerina, which suggest a relatively condensed succession in Arundian – lower Asbian strata compared with other regions in northern England, but including all the foraminiferal subzones defined by Conil et al. (Reference Conil, Longerstaey and Ramsbottom1980).

Fig. 3. Selected foraminiferal key species from the Askrigg Block in northern England and in the Montagne Noire in southern France. Scale bar: (a–g) 200 μm and (h–t) 400 μm. (a) ‘Archaediscus’ at concavus stage, level Pc5339, Valuzière section. (b) ‘Archaediscus’ at concavus stage, level EWJ1565, Silverdale Borehole. (c) ‘Archaediscus’ at concavus stage transitional to the angulatus stage, level EWJ1565, Silverdale Borehole. (d) Nodosarchaediscus cornua (Conil & Lys, Reference Conil and Lys1964), level Pc5339, Valuzière section. (e) Nodosarchaediscus cf. saleei (Conil & Lys, Reference Conil and Lys1964), level Pc4660, Puech de la Suque section. (f) Nodosarchaediscus sp., level Pc5148, Puech de la Suque Hill 248 section. (g) ‘Archaediscus’ at concavus stage transitional to the angulatus stage, level Pc5337, Valuzière section. (h) Pojarkovella sp., level EWJ1563, Silverdale Borehole. (i) Pojarkovella sp., level EWJ1566, Silverdale Borehole. (j) Pojarkovella sp., level EWJ1562, Silverdale Borehole. (k) Pojarkovella nibelis, level EWJ1561, Silverdale Borehole. (l) Endothyranopsis aff. compressa, level EWJ1573, Silverdale Borehole. (m) Latiendothyranopsis menneri solida, level EWJ1573, Silverdale Borehole. (n) Plectogyranopsis moraviae, level EWJ1573, Silverdale Borehole. (o) Plectogyranopsis settlensis, level Pc4900, Pic de Vissou section. (p) Ugurus intermedius Vachard et al. (Reference Vachard, Coronado, Izart and Cózar2018), level Pc5537, Valuzière section. (q) Endothyranopsis compressa, level EWJ1562, Silverdale Borehole. (r) Omphalotis minima, level Pc4983, Puech de la Suque section. (s) ‘Millerella’ aff. infulaeformis Ganelina (Reference Ganelina1951), level Pc4896, Pic de Vissou section. (t) Koskinotextularia sp., level Pc4660, Puech de la Suque section.

It is necessary to highlight the interval from the first occurrence of Archaediscus at concavus stage and Pojarkovella sp. up to the first occurrence of Koskinotextularia and P. nibelis. The latter taxa allow the correlation with bed k of the Park Limestone of the Barker Scar stratotype; this interval is therefore correlated with the upper part of the Dalton Beds and assigned to the Arundian Substage. The level with Koskinotextularia/P. nibelis in other outcrops where the Scaleber Force Limestone and Scaleber Quarry Limestone members are well differentiated is coincident with the boundary between both members, as well as the abovementioned macrofauna (Dunham & Wilson, Reference Dunham and Wilson1985; Arthurton et al. Reference Arthurton, Johnson and Mundy1988). The boundary between Scaleber Force Limestone and Scaleber Quarry Limestone members mimics the fossil record of that in the boundary between Dalton Formation and Park Limestone Formation.

In the Kilnsey Crag, the type locality for the Kilnsey Limestone Formation, the younger member is well-exposed and only the top 1–2 m of the Scaleber Force Limestone Member is seen (see Waters et al. Reference Waters, Cózar, Somerville, Haslam, Millward and Woods2017, fig. 6). The occurrence of Endothyranopsis compressa is recorded at the top of the lower member.

2.c. Lower–middle Viséan boundary interval in southern France

In the Montagne Noire, the transition between the lower and middle Viséan substages is rarely recognized in sections because of the poor exposure and commonly hostile facies for the foraminifers (Cózar et al. Reference Cózar, Izart, Vachard and Coronado2017; Vachard et al. Reference Vachard, Coronado, Izart and Cózar2018). Archaediscus at concavus stage has only been recorded at Valuzières section, at the same level as Pojarkovella sp. and Endothyranopsis compressa (Figs 3, 4). However, Archaediscus at concavus stage transitional to the angulatus stage and Pojarkovella nibelis, as well as the regional markers such as Ugurus, primitive ‘Millerella’ and Cribrospira, occur only 0.8 m above this level (Fig. 3p, s). This small difference in the Valuzières section between the first appearances of the taxa does not allow a clear conclusion if this intermediate zone at the top of the lower Viséan strata is present. It could be poorly developed, but additional sections should confirm this short interval. Nevertheless, the succession in shallow-water platforms of the Montagne Noire can also be considered as relatively condensed, with a limited thickness. Moreover, below the Valuzières Formation a typical griotte limestone occurs, the Combe Roland Formation, which is rather poor in foraminifers; in Tourière (Fig. 4) and in the type sections, it also contains Archaediscus at concavus stage as well as transitional forms to the angulatus stage (Cózar et al. Reference Cózar, Izart, Vachard and Coronado2017; Vachard et al. Reference Vachard, Coronado, Izart and Cózar2018). Calciturbidites of the slope facies do not contain the transition to the lower Viséan substage; however, the lower part of the middle Viséan Colonnes Formation also shows the same first occurrences of the taxa (Fig. 5). The stratigraphic records of the taxa are rather similar to those in northern England.

Fig. 4. Biostratigraphic sketch of the Valuzières and Tourière sections, Montagne Noire (southern France). CB – Combe Rolland Formation; Tou – Tourière Formation.

Fig. 5. Biostratigraphic sketch of the Puech de la Suque-Hill 248 and Pic de Vissou sections, Montagne Noire (southern France).

2.d. Lower–middle Viséan boundary interval in the western Meseta of Morocco

This study is limited to the western Meseta because the lower and middle Viséan substages are not documented in the eastern Meseta (M. Berkhli, unpub. MSc thesis, University of Lille, 1993). The occurrence of the lower Viséan substage in Morocco has been a much debated point since early studies in the region, because assemblages older than the V2a are rarely documented (Huvelin & Mamet, Reference Huvelin and Mamet1997). Subsequent to the chronostratigraphical changes introduced by Conil et al. (Reference Conil, Groessens and Pirlet1977), many sections were considered from their bases to be middle Viséan (V2a) in nature; problematic lower Viséan strata were later documented (e.g. Verset, Reference Verset1983, Reference Verset1988), although it is not clear whether the successions based on those assemblages compared with the V2a were considered as lower or middle Viséan (e.g. Vachard et al. Reference Vachard, Termier and Termier1977). To avoid problems with the comparison with the European zonal scales, these authors proposed a regional zonal scheme for Morocco using foraminifers and calcareous algae (Vachard, Reference Vachard1988b; Vachard & Tahiri, Reference Vachard and Tahiri1991); this has been updated (in stratigraphic ranges and guides) in subsequent studies (Vachard & Berkhli, Reference Vachard and Berkhli1992; Berkhli et al. Reference Berkhli, Vachard and Paicheler2000; Izart et al. Reference Izart, Tahiri, El Boursoumi, Vachard, Saidi, Chèvremont and Berkhli2001, Reference Izart, Vachard and Berkhli2017).

The basal zone Cfm1 is characterized by the primitive forms of Uralodiscus and Glomodiscus, with U. rotundus as a marker and Archaediscus convexus Grozdilova & Lebedeva in Dain & Grozdilova (Reference Dain and Grozdilova1953) in its upper part. This zone is typically represented in the Sidi Sebaa Formation (Fig. 6), and was correlated with the upper part of the Moliniacian Substage and lower part of the Livian, upper Arundian and lower Holkerian, upper MFZ11 and lower MFZ12, upper V2a and lower V2b, upper Cf4δ and lower Cf5 substages (Izart et al. Reference Izart, Vachard and Berkhli2017). The base of the succession in the western Meseta seems to be only present at the Sidi Sebaa section where, in the lithological unit E2 (Fig. 6), Uralodiscus rotundus, Glomodiscus oblongus, Lapparentidiscus nanus (Reitlinger, Reference Reitlinger1969) and Archaediscus at involutus stage (Vachard & Fadli, Reference Vachard and Fadli1991) occur. This assemblage can be compared with the Cf4γ-Cf4δ zones of Conil et al. (Reference Conil, Longerstaey and Ramsbottom1980, Reference Conil, Groessens, Laloux, Poty and Tourneur1991), most of the MFZ11 zone of Poty et al. (Reference Poty, Devuyst and Hance2006), and the upper Moliniacian Substage (Devuyst et al. Reference Devuyst, Hance and Poty2006). Above this level the first change in the assemblages is characterized by the appearance of Archaediscus convexus (at concavus stage), which, in addition to Sidi Sebaa, it is also recognized at north Oulmès (Vachard & Tahiri, Reference Vachard and Tahiri1991; Izart et al. Reference Izart, Tahiri, El Boursoumi, Vachard, Saidi, Chèvremont and Berkhli2001), Agouraï (M. Ben Abbou, unpub. MSc thesis, Univiversity of Sidi Mohamed ben Abdellah, Fès, 1990) and Skoura boutonnière (Izart et al. Reference Izart, Poty and Vieslet1989; unpublished data). In this interval, Endothyranopsis compressa has been documented, although generally as ex group species or in open nomenclature (E. cf. compressa and E. aff. compressa) (Vachard & Tahiri, Reference Vachard and Tahiri1991; Izart et al. Reference Izart, Tahiri, El Boursoumi, Vachard, Saidi, Chèvremont and Berkhli2001).

Fig. 6. Biostratigraphic sketch of the Sidi Sebaa section, Morocco. Bed thickness based on D. Fadli (unpub. PhD thesis, University of Rabat, 1990).

In an early attempt, Izart et al. (Reference Izart, Tahiri, El Boursoumi, Vachard, Saidi, Chèvremont and Berkhli2001) subdivided the Cfm1 zone into the Cfm1a subzone characterized by Uralodiscus rotundus and a younger Cfm1b subzone characterized by Archaediscus convexus (Fig. 7), although these subzones were not used in later studies.

Fig. 7. Correlation of the different proposals for the MFZ11B subzone in eastern Palaeotethys basins and the Cfm1b subzone from Morocco, according to their authors.

Assemblages composed of Uralodiscus/Glomodiscus/‘Archaediscus’ at concavus stage are predominant in this upper interval, which can be correlated with the unnamed zone recorded in England at the top of the Arundian Substage.

The Cfm2 zone is currently characterized by Uralodiscus and Conilidiscus as markers (e.g. Berkhli et al. Reference Berkhli, Vachard and Paicheler2000), but also by Endothyranopsis compressa and Forschia parvula Rauzer-Chernousova (Reference Rauzer-Chernousova1948b) in older works (Vachard, Reference Vachard1988b). The zone was correlated with parts of the Cf5, MFZ12, Livian and Holkerian substages and upper part of the V2b zone (Izart et al. Reference Izart, Vachard and Berkhli2017). Its type locality is Tizi ben Zizouit (Termier et al. Reference Termier, Termier and Vachard1975). In this case, the foraminiferal stratigraphic record in different sections is not as consistent as for the previous stratigraphic records. In the Mdakra Massif and El Hamman Ridge, Conilidiscus appears much earlier than Endothyranopsis compressa (Vachard & Fadli, Reference Vachard and Fadli1991; Izart et al. Reference Izart, Tahiri, El Boursoumi, Vachard, Saidi, Chèvremont and Berkhli2001), whereas in Oulmès, E. compressa occurs much earlier (Vachard & Tahiri, Reference Vachard and Tahiri1991). In Agouraï, both taxa occur from the base of the succession, with E. compressa in the basal conglomeratic/sandy unit (Berkhli et al. Reference Berkhli, Vachard and Paicheler2000). In Skoura, only Conilidiscus occurs (Izart et al. Reference Izart, Poty and Vieslet1989, as Tubispirodiscus). The subdivision of the Cfm1b and Cfm2 zones therefore does not seem useful, because the late occurrence of Conilidiscus compared with taxa such as ‘Archaediscus’ at concavus stage and E. compressa is not clear enough and the succession is in need of revision. The suite of foraminifers recorded in this interval might also be compared with those of the unnamed interval at the top of the Arundian Substage.

The Cfm3 zone is characterized by Koskinotextularia bradyi (von Möller, Reference von1879) as a guide. It is correlated with the upper part of the Cf5, Holkerian, Livian and MFZ12 and with the V3a zones. The second marker, Pojarkovella nibelis, is only known in one section close to Tiflet town (Vieslet, Reference Vieslet1983), whereas the taxon is mostly recorded in the upper Viséan limestones of Morocco (Vachard & Tahiri, Reference Vachard and Tahiri1991). The type locality of this zone is at Irhzer Aoujgal (Termier et al. Reference Termier, Termier and Vachard1975). Unquestionable middle Viséan markers are only recognized from this Cfm3 zone. The virtual lack of Pojarkovella nibelis is a handicap for comparison with Europe because Koskinotextularia is a comparatively rare taxon; on contrast, P. nibelis from its basal range is commonly recorded and therefore more useful for the recognition of middle Viséan carbonates.

These chronozones are temporally slightly out of phase compared with European zones, and some of the European markers arrived later to the Meseta (Vachard & Tahiri, Reference Vachard and Tahiri1991). The occurrences of typical taxa of the Cfm1 and Cfm2 zones in sections assigned to the Cfm3 zone (e.g. Vieslet, Reference Vieslet1983) significantly complicate the recognition of the lower and middle Viséan substages in Morocco.

3.a. Lower–middle Viséan boundary interval in China

Devuyst et al. (Reference Devuyst, Hance, Hou, Wu, Tian, Coen and Sevastopulo2003) and Devuyst (unpub. PhD thesis, Université catholique de Louvain, 2006) suggested the presence of an interval zone at the top of the Pengchong section (Tournaisian–Viséan boundary global boundary stratotype section and point), unrecognized in Belgium. This interval zone was based on the presence of primitive archaediscids (Uralodiscus, Glomodiscus and Conilidiscus) after the first occurrence of Pojarkovella nibelis. This zone was named as zone G in Devuyst et al. (Reference Devuyst, Sevastopulo, Hance, Hou, Kalvoda and Wu2004) following the zones defined previously in the section by Hance et al. (Reference Hance, Muchez, Hou and Wu1997). However, the Pengchong section is characterized, overall, by poor archaediscid assemblages, attributed by the authors to a lack of deep-water foraminifers in turbidites, due to little downslope mixing of the fauna (Devuyst et al. Reference Devuyst, Sevastopulo, Hance, Hou, Kalvoda and Wu2004). The occurrence of primitive assemblages of archaediscids at the base of inferred middle Viséan rocks is not sufficient to justify the occurrence of this new zone, because this fact is also recognized in Morocco, Moravia, Russia and Kazakhstan (e.g. Reitlinger et al. Reference Reitlinger, Vdovenko, Gubareva, Shcherbakov, Wagner, Winkler Prins and Granados1996; Kalvoda, Reference Kalvoda2002; Brenckle & Milkina, Reference Brenckle and Milkina2003; Kulagina et al. Reference Kulagina, Gibshman and Pazukhin2003); this can only be a matter of longer persistence of the taxa for regional ecological factors.

Hance et al. (Reference Hance, Hou and Vachard2011) subdivided the MFZ11 – Uralodiscus rotundus Zone into a lower subzone (MFZ11A), containing mostly the classical markers defined in Poty et al. (Reference Poty, Devuyst and Hance2006), and an upper subzone (MFZ11B), characterized by the first occurrence of Pojarkovella (Fig. 7); the specimens first occurring at this level are more primitive than P. nibelis, a marker for the MFZ12 Zone. This MFZ11B subzone was only recognized in two sections, Mopanshan and Jiusi, but neither contains the transition into the MFZ12, meaning that the transition into the middle Viséan substage cannot be studied in detail. Other sections in the region contain poorer assemblages in the upper part of the lower Viséan substage, and the MFZ11B subzone could not be distinguished and, contrary to the previously studied Pengchong and Yajiao sections by Devuyst et al. (Reference Devuyst, Hance, Hou, Wu, Tian, Coen and Sevastopulo2003, Reference Devuyst, Sevastopulo, Hance, Hou, Kalvoda and Wu2004) and Devuyst (unpub. PhD thesis, Université catholique de Louvain, 2006), Hance et al. (Reference Hance, Hou and Vachard2011) only recognized the MFZ11 and MFZ12 zones in the same sections. It has to be highlighted that the latter authors considered Endothyranopsis ex gr. compressa as typical of the lower Viséan MFZ11 zone, but was only illustrated from the MFZ11B subzone. It must also be noted that the first occurrence of Conilidiscus is located by those authors in the MFZ11A subzone.

The unusual record of the primitive archaediscids in China represents a significant complication when trying to calibrate the first occurrences of other taxa, as well as establish a precise definition of the foraminiferal zones. But it also has to be added that the first occurrence of Archaediscus at concavus stage and the transitional forms to the angulatus stage are both recorded from the MFZ12 zone (e.g. Hance et al. Reference Hance, Hou and Vachard2011, pl. 51, fig. 14) and only primitive Pojarkovella first occurs below this substage.

3. Lower–middle Viséan boundary interval in Turkey

Okuyucu et al. (Reference Okuyucu, Vachard and Göncüoğlu2013) studied material assigned to the MFZ11B subzone in the Istanbul terrane. These authors considered the first appearances of Uralodiscus rotundus (and other species of the genus), Paraarchaediscus, Cribranopsis and Conilidiscus and the disappearance of Eoendothyranopsis, Eoparastaffella, Pseudolituotubella, Condrustella and Eotextularia characteristic of the MFZ11 zone. They interpreted that Paraarchaediscus (‘Archaediscus’ herein) and Conilidiscus might be guides for the MFZ11B subzone. However, these taxa are recorded from near the base of the MFZ11 zone in Belgium (Poty et al. Reference Poty, Devuyst and Hance2006), as well as from the MFZ11A subzone in China (Hance et al. Reference Hance, Hou and Vachard2011); they might therefore be used as regional markers, but not to characterize a global MFZ11B subzone.

Compared with the European zones, this modification into the MFZ11B should be correlated with the Cf4γ-δ; clearly, it cannot be compared with the MFZ11B in China (Fig. 7). There is no mention of other important taxa in Turkey such as primitive Pojarkovella, the first Endothyranopsis, and therefore no evidence of any unusual foraminiferal assemblage below the typical middle Viséan strata.

3.c. Lower–middle Viséan boundary interval in Iran

Zandkarimi et al. (Reference Zandkarimi, Najafian, Vachard, Bahrammanesh and Vaziri2016) identified the MFZ11B subzone on the basis of advanced Glomodiscus and Archaediscus in NW Alborz. This definition is rather similar to that in Turkey (Fig. 7). The rest of the assemblage contains Archaediscus/Paraarchaediscus at involutus stage, and Ammarchaediscus, Planoarchaediscus and Eotextularia; the assemblage should therefore also be correlated with the Cf4γ-δ subzones of Conil et al. (Reference Conil, Longerstaey and Ramsbottom1980, Reference Conil, Groessens, Laloux, Poty and Tourneur1991), and with the classical concept of the MFZ11 in Poty et al. (Reference Poty, Devuyst and Hance2006). The first occurrence of primitive Pojarkovella is at the same horizon as P. nibelis, and no difference exists.

Zandkarimi et al (Reference Zandkarimi, Vachard, Najafian, Hamdi and Mosaddegh2017, Reference Zandkarimi, Vachard, Najafian, Mosaddegh and Ehteshami-Moinabadi2018) only recognized the MFZ11 and MFZ12 zones in northern Alborz, and they did not subdivide the former zone. The occurrence of Pojarkovella spp. is synchronous in the main stratigraphic sections at Naserabad and Dozdehban, and also coincident with the first appearance of Koskinotextularia. Zandkarimi et al. (Reference Zandkarimi, Vachard, Najafian, Mosaddegh and Ehteshami-Moinabadi2018) described the occurrence of Endothyranopsis compressa from the MFZ11 zone, although this species (generally very rare in the region) was used for the recognition of the base of the middle Viséan substage at the Naserabad section (at sample Ns102) whereas P. nibelis occurs 1 m above (at Ns103). Other specimens recorded in older rocks are identified with a question mark, and its presence in the lower Viséan deposits of Iran is therefore questioned.

However, although not acknowledged by the authors, ‘Archaediscus’ at concavus stage are recorded from the early Viséan Age (Zandkarimi et al. Reference Zandkarimi, Najafian, Vachard, Bahrammanesh and Vaziri2016, figs 7.35, 7.38; Zandkarimi et al. Reference Zandkarimi, Vachard, Najafian, Hamdi and Mosaddegh2017, figs 7.29, 7.30). Furthermore, although Zandkarimi et al. (Reference Zandkarimi, Vachard, Najafian, Hamdi and Mosaddegh2017) recognized the occurrence of Paraarchaediscus at concavus transitional to the angulatus stage from the middle part of the middle Viséan strata, specimens are recorded from closer positions to the base of the middle Viséan Stage (Zandkarimi et al. Reference Zandkarimi, Najafian, Vachard, Bahrammanesh and Vaziri2016, fig. 9.12; Zandkarimi et al. Reference Zandkarimi, Vachard, Najafian, Hamdi and Mosaddegh2017, figs 7.37, 7.38, 7.40).

Fassihi et al. (Reference Fassihi, Kulagina, Esfahani and Nurgalilev2018) subdivided the succession at the Sanandaj-Sirjan region in Central Iran into two local zones. The older zone was correlated with the MFZ11B subzone due to the occurrence of Glomodiscus, Uralodiscus and Paraarchaediscus and assigned a latest early Viséan age (Fig. 7). The youngest local zone was characterized by the appearance of Lapparentidiscus bokanensis Vachard (Reference Vachard1980) and the disappearance of dainellids. These authors correlated this zone with the lowest ‘upper’ Viséan or MFZ12 strata of Belgium. (It must be noted that, in the Russian scale, the Viséan Stage is only divided into lower and upper, and the Tulian substage, more or less equivalent to the middle Viséan substage in western Europe, is the lower part of the upper Viséan substage.) However, there is no foraminifer which suggests a middle Viséan age within this assemblage, and the entire succession should be assigned to the Cf4γ-δ subzones of early Viséan age.