Gondwana Research 21 (2012) 654–669

Contents lists available at ScienceDirect

Gondwana Research

j ourna l homepage: www.e lsev ie r.com/ locate /gr

A new U–Pb zircon age dating and palynological data from a Lower Permian sectionof the southernmost Paraná Basin, Brazil: Biochronostratigraphical andgeochronological implications for Gondwanan correlations

Ana Luisa Outa Mori a,⁎, Paulo Alves de Souza a, Juliana Charão Marques b, Ricardo da Cunha Lopes c

a Laboratório de Palinologia Marleni Marques-Toigo, Departamento de Paleontologia e Estratigrafia, Instituto de Geociências, Universidade Federal do Rio Grande do Sul. Avenida BentoGonçalves, 9500, CEP 91.540-000, Porto Alegre, RS, Brazilb Laboratório de Geologia Isotópica, Instituto de Geociências, UFRGS, Brazilc Universidade do Vale do Rio dos Sinos (UNISINOS). Av. Unisinos, 950—B. Cristo Rei/CEP 93.022-000—São Leopoldo, RS, Brazil

⁎ Corresponding author. Fax: +55 51 3308 6332.E-mail addresses: luisaouta@yahoo.com.br (A.L.O. M

(P.A. Souza), juliana.marques@ufrgs.br (J.C. Marques), rica

1342-937X/$ – see front matter © 2011 International Adoi:10.1016/j.gr.2011.05.019

a b s t r a c t

a r t i c l e i n f o

Article history:Received 28 October 2010Received in revised form 10 May 2011Accepted 29 May 2011Available online 17 June 2011

Handling Editor: E. Tohver

Keywords:PalynologyLower PermianParaná BasinBiostratigraphyRadiometrical dating

Index species useful for correlations with the International Stratigraphical Column are rare or absent in thePennsylvanian–Permian strata of the Paraná Basin in Brazil, preventing accuracy in geochronologicassignments. Besides, absolute datings are very scarce in comparison with other Gondwana basins. Thispaper presents palynological data from an outcrop on the surroundings of the Candiota coal mine, southmostBrazil, from several levels of the Rio Bonito and Palermo formations. The presence of certain index species ofspore–pollen allowed the recognition of two Permian palynozones: the Vittatina costabilis and theLueckisporites virkkiae zones. Furthermore, U–Pb in zircons from a volcaniclastic level interbedded in thecoal strata of the former unit was analyzed through LA-MC-ICP-MS method, providing a new absolute agedating of 281.4±3.4 Ma (Cisuralian, Early Permian). This dating is assumed as the oldest occurrence of the L.virkkiae Zone in Paraná Basin, which contains index species that are widespread in other Gondwana basins. Awell distributed surface boundary occurs in this section also, allowing local and regional correlations. Thesenew biostratigraphical and geochronological data are integrated, in order to offer a deep analysis on thestratigraphical significance for correlations across the Occidental Gondwana.

© 2011 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

1. Introduction

Late Paleozoic deposits of western Gondwana comprise severalfossil groups that played an important role in relative age assignmentsand regional correlations. However, these fossil assemblages aregenerally very distinct from those that characterize the Carboniferousand Permian stratotypes of the International Stratigraphic Column.Themost important fossils employed by these systems are graptolites,fusulinids, ammonites and conodonts, which are mainly derived fromLaurasia basins. Index species of these groups are generally scarce orabsent in most part of Gondwana sedimentary sequences. Thesignificant fossil records used for biostratigraphical purposes for theCarboniferous and Permian Gondwanan deposits of South Americawere summarized by Azcuy et al. (2007). The best geochronologicalcalibrations are only possible through the integration of datarecovered from invertebrates, such as fusulinids and conodonts.However, such occurrences are rare and restricted to certain

ori), paulo.alves.souza@ufrgs.brrdocl@unisinos.br (R.C. Lopes).

ssociation for Gondwana Research.

stratigraphic levels, preventing the establishment of biostratigraphicschemes. Furthermore, absolute age dating is relatively rare fromthese deposits, limiting the accuracy in stratigraphical assignmentsand correlations between the local, regional and global scales. Thistheme was analyzed by Césari (2007) who focused available data onabsolute age dating and their palynostratigraphical implications alongthe Gondwana, especially in Pennsylvanian and Permian sequences ofSouth America. A deeper discussion about absolute age dating for theUpper Paleozoic Sequences on the western Gondwana was recentlyintroduced by Césari et al. (2011), who established the time span forcertain Carboniferous Argentinian palynofloras.

Palynological associations recorded from the Pennsylvanian andPermian sequences of the Paraná Basin in Brazil are especiallyimportant for this Gondwana portion due the great abundance ofspore–pollen species, including common taxa used for correlations.General schemes initially proposed by Daemon and Quadros (1970)and Marques-Toigo (1988, 1991) have been deeply developed inrecent years (Souza and Marques-Toigo, 2003, 2005; Souza, 2006).In addition, absolute age dating carried out in conjunction withpalynology help to define the time span of certain species orbiozones, improving lateral correlations and paleogeographicalreconstructions. In the Paraná Basin, papers integrating absolute

Published by Elsevier B.V. All rights reserved.

655A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

age datings and palynology are available from Rio Bonito (Matos etal., 2001; Guerra-Sommer et al., 2005, 2008a,b; Rocha-Campos et al.,2006, 2007;) and Irati Formations (Rocha-Campos et al., 2006;Santos et al., 2006; Rocha-Campos et al., 2007). However, absoluteages for the former lithostratigraphical unit cover a time span toolong relative to the tonsteins recorded in Candiota coal mine. Thiscoal mine comprises several tonstein levels (see Matos, 1999), ofdifferent ages, probably related to Choiyoi explosive eruptions ofSan Rafael Block, western Argentina, which injected enormousvolumes of tephra in the troposphere–stratosphere (Rocha-Camposet al., 2011), acting as source of certain Permian ash fall layers inParaná Basin.

The coalfields of Candiota, Rio Grande do Sul State, southernmostBrazil, represent one of the most important economic deposits of coalin the region. These coalfields were the focus of several stratigraphicalstudies (e.g., Fontes and Cava, 1980; Alves and Ade, 1996; Holz, 1998;Holz et al., 2006), allowing geological refinements of these depositsand optimization of their exploitation.

This contribution presents an integrated palynological and U–Pbzircon age analysis, based on new data obtained from an outcropplaced in Candiota, comprising the Rio Bonito and the Palermoformations. Palynological associations were previously presented byMori and Souza (2010), furnishing biostratigraphical assignments tothe levels described. A deeper discussion on the stratigraphicalsignificance is given in this new contribution, once the samplesstudied herein are stratigraphycally above of those previouslychecked by many other authors (Matos et al., 2001; Guerra-Sommeret al., 2005, 2008a,b), which comprise age dating and/or palynologicaldata. These new data have important biostratigraphical and geochro-

Fig. 1. A. Distribution of the Paraná Basin alostratigraphic units. B. Stratigraphic

nological implications, for correlation with other Gondwanan paly-nological assemblages.

2. Geologic and palynological synopsis

2.1. Geological setting

The Paraná Basin is a large intracratonic basin located in the centraleastern portion of South America, covering an approximate area of1,700,000 km2, mainly in Brazil, but also in Uruguay, Argentina andParaguay (Fig. 1A). This basin bears a thick sedimentary–magmaticsequence, reaching up 6800 m in thickness. According to Milani(1997), six supersequences are recognized in this basin: Rio Ivaí(related to the Rio Ivaí Group, of Ordovician–Silurian age), Paraná(Paraná Group, Devonian), Gondwana I (Itararé, Guatá and Passa Doisgroups, Pennsylvanian to Permian), Gondwana II (Rosário do SulGroup, Middle to Upper Triassic), Gondwana III (São Bento Group,Jurassic–Cretaceous) and Bauru (Bauru Group, Cretaceous).

The Gondwana I Supersequence (Fig. 1B) corresponds to a largetransgressive–regressive cycle, comprising the Itararé and Guatá (RioBonito, Palermo and Tatuí formations) groups, which are dated asPennsylvanian (Bashkirian/Moskovian) to Cisuralian (Artinskian), aswell as the Passa Dois Group (Irati, Serra Alta, Teresina and Rio doRasto formations), dated as Cisuralian (Artinskian) to Lopingian (?Wuchiapingian), constituted by a thick sedimentary sequence of ca.2500 m that record paleoenvironmental changes in this portion of theGondwana, notably the shift from glacial to arid conditions (Milaniand Zalán, 1999).

al setting of the Gondwana I Supersequence (modified from Milani, 1997).

656 A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

Sedimentary post-glacial deposits of the Rio Bonito Formation areconstituted by fluvial sandstones, as well as coal beds and siltstones,originated in lagoonal and deltaic paleoenvironment. The PalermoFormation comprises siltstones and claystones, which are interpretedas marine in origin. A summary of stratigraphic and paleontologicaldata of these units are available in Holz et al. (2010).

2.2. Local stratigraphy

The Candiota coal mine is the southernmost coalfield in theBrazilian portion of the Paraná Basin, constituted by sedimentarydeposits of the Rio Bonito Formation. Its sedimentary basal rocks aremainly coarse to medium-grained, arkosic sandstones, interpretedas fluvial–estuarine in origin, that uncomfortably overlie thevarvites and tilites of the Itararé Group. The thickness of theformation varies according to the paleo relief, reaching up to a few totens of meters, with sparse intercalations of mudstones to organic-rich mudstones and thin lenticular coal seams. The middle part ofthe Rio Bonito Formation is locally composed of mudstones toorganic-rich mudstones, as well as thick, regionally extensive coal.The main coal bed is named as the “Candiota Coal Seam”, ranges inthickness from 1 to 9 m and is divided into two by a thick, clay-richbed in the middle. Overlying the main coal seams are quartz–sandstone beds, ranging in thickness from tens of centimetres to16 m, interpreted as a lagoonal–estuarine or barrier island environ-ment. This bed is succeeded by a new level of mudstones to organic-rich mudstones, widely distributed, constituted of thin coal seams(0.2 to 1 m each), named as “Candiota 4, 5, 6 and 7 Upper Seam”. Thetransgressive marine sediments of the overlying Palermo Formationare deposited o top of an erosive surface, named by Holz et al. (2006)as “SB3”, constituting a widespread surface boundary. The PalermoFormation comprises heterolithic deposits interbedded, fine-grained sandstones with hummocky cross-beds, representing shore-face to offshore deposits of a transgressive environment.

2.3. Palynostratigraphy

Several biostratigraphic zonations were proposed for the Penn-sylvanian–Permian interval of the Paraná Basin, using different scalesand concepts. The scheme of Daemon and Quadros (1970) isconsidered the most important in geographic and stratigraphicterms. Six interval zones were recognized, which are, in ascendingstratigraphic order: G, H (subdivided into subintervals H1, H2 and H3),I (subintervals I1, I2+I3+I4), J, K and L (L1, L2 and L3). Later, Marques-Toigo (1988, 1991) and Souza (2000) proposed an integratedpalynostratigraphic scheme for the southern and northeasternportions of this basin, improving the original proposal of Daemonand Quadros (1970), and including new data, as summarized in Souzaand Marques-Toigo (2003, 2005) and Souza (2006). Then, thepalynologic succession of the Pennsylvanian–Permian interval of theParaná Basin comprises four interval zones, named as Ahrensisporitescristatus (AcZ), Crucisaccites monoletus (CmZ), Vittatina costabilis(VcZ) and Lueckisporites virkkiae (LvZ) zones, in ascending strati-graphic order. The basal AcZ and CmZ units were presented in detailby Souza (2006), which are recognized only in the northeasternportion of the basin, in Paraná (PR) and São Paulo (SP) states. Thesezones were recognized within the lower and middle portions ofthe Itararé Group, interpreted as Pennsylvanian in age (Bashkirian/Moskovian to Gzhelian). The VcZ is observed from the uppermostlevels of the Itararé Group until the uppermost portion of the RioBonito Formation or the lowermost Palermo Formation. This biozonecomprises two units: Protohaploxypinus goraiensis and Hamiapolle-nites karrooensis subzones. The lower limit of the VcZ and thesubjacent LvZ is delimited by the first appearance of certain species ofbissacate and taeniate pollen grains, and represents a stratigraphicdatum recognized along the basin (Souza and Milani, 2007). The LvZ

occurs from the uppermost levels of Rio Bonito Formation or from thebasal levels of Palermo Formation, including the Irati Formation,reaching up the lowermost portion of the Rio do Rasto Formation(Neregato, 2007; Neregato et al., 2008). The age assigned to these lasttwo biozones is Asselian to Artinskian and Artinskian to Wuchiapin-gian, respectively, encompassing a significant interval of Permiantime. A synthesis of the main stratigraphic and paleontologic data forthe Pennsylvanian–Permian interval of the Paraná Basin is presentedin Holz et al. (2010).

2.4. Previous palynological data from Candiota

The first paleontological records from Candiota coal mine werebased on plant remains, megaspores and microspores, derived fromnot described and unmeasured beds (Carruthers, 1869; Plant, 1869;Liais, 1872; Pant and Srivastava, 1965; Tiwari and Navale, 1967).Palynological data of several localities of Candiota–Hulha Negraregion were presented by Nahuys et al. (1968), who was the first toascribe a Permian age to these beds. A significant palynotaxonomiccontribution was offered by Ybert (1975), while paleoenvironmentalinterpretations based on palynological associations and petrographicstudies are available in Marques-Toigo et al. (1975) and Corrêa daSilva andMarques-Toigo (1975). Marinemicrofossils, such asNavifusaand Cymatiosphaera, were recorded from certain coals levels andrelated deposits by Meyer and Marques-Toigo (2000), corroboratingstratigraphic interpretations of marine incursions within the coaldeposits (Alves and Ade, 1996; Holz, 1998).

Palynological studies of this coal mine presented by Cazzulo-Klepzig et al. (2002, 2005) and Guerra-Sommer et al. (2008a,b) arebased on previous papers, or restricted to taxonomic lists. Further-more, some of these studies included specimens previously illustratedfrom other localities of the Paraná Basin, as stated by Santos et al.(2006, p. 461), preventing a deeper analysis as well as detailedcomparisons.

2.5. Geochronological data from Candiota

Three tonstein levels occur between the coalbeds and siltstone atthe Candiota coal mine, named in ascending stratigraphic order:Tonstein A, B and C. The first tonstein level is included in “LowerCandiota Bed”, and the other two in the “Upper Candiota Bed”. Theseare the tonsteins levels fromwhich absolute age dating were obtainedfor the Rio Bonito Formation, presented by Matos et al. (2001),Guerra-Sommer et al. (2005, 2008a,b) and Rocha-Campos et al. (2006,2007). These ages range from 267.1±3.4 Ma to 299.1±2.6 Ma, assummarized in the Table 1, indicate some uncertainty as to the properage assignment, most likely due tomethodological improvements andto different stratigraphic position of the levels analyzed in each study.

Preliminary data from the tonstein analyzed herein was presentedbyMarques et al. (2007). A LAM-ICP-MS U/Pb zircon agewas obtained(278.2±1.8 Ma), but it is derived from few zircon grains. Additionalsampling was carried out in order to assess this age.

3. Materials and methods

This study is based on outcrop samples of the Rio Bonito andPalermo formations, collected on the km 152 on BR 293 Highway,located between Bagé and Candiota cities (UTM coordinates6523299 N/234914 E), Rio Grande do Sul State in the southernmostBrazil (Fig. 2A). Six samples were retrieved from the Rio BonitoFormation (named as C1, C2, C3, C4, C5 and C6) and two from thePalermo Formation (C7 and C8) (Fig. 2B). The contact between thesetwo lithostratigraphic units is erosive, located between samples C6and C7. The tonstein level used for absolute dating (Fig. 2C and D)occurs at the base of the C4 sample, within the Rio Bonito Formation.

Table 1Summary of radiometric datings available for the Candiota Coalmine, Rio Bonito Formation, Paraná Basin.

Matos et al. (2001) Guerra-Sommer et al.(2005)

Rocha-Campos et al.(2006, 2007)

Guerra-Sommer et al. (2008a) Guerra-Sommer et al. (2008b)

Stratigraphicallevel

Tonstein A (LowerCandiota Coal Seam)

Lower Candiota Coal Seam(1 tonstein level not clearlyspecified)

Lower Candiota Coal Seam(4 tonstein levels not clearlyspecified)

Tonstein A (Lower Candiota CoalSeam)/Tonstein C–(UpperCandiota Coal Seam)

Tonstein A (Lower Candiota CoalSeam)/Tonstein C (UpperCandiota Coal Seam)

AnalysisMethod

IDTIMS U/Pb IDTIMS U/Pb SHRIMP RG/SHRIMP 2 SHRIMP IDTIMS U–Pb

Main age 267.1±3.4 Ma 299.1±2.6 to296.9±1.4 Ma

– Tonstein A: 293,96 +3,83/−4,20 Ma Tonstein C: 289.36 +1.84/−1,58 Ma

Tonstein A: 296±4.2 MaTonstein C: 296.9±1.65 Ma

Average age – – 298.5±2.6 Ma 290.6±1.5 Ma –

657A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

Preparation for palynological analysis followed the basic proce-dure of maceration, chemical digestion of inorganic components usinghydrochloric and hydrofluoric acids, as described by Wood et al.(1996). The final residues were sieved to concentrate the size fractionbetween 20 and 250 μm. Excess amorphous organic matter wasremoved using a solution of KOH 10%. Twenty-eight palynologicalslides were mounted (codes MP-P: 5666 to 5694), which are stored atthe Laboratório de Palinologia “Marleni Marques Toigo”, Departa-mento de Paleontologia e Estratigrafia in the Instituto de Geociênciasat Universidade Federal do Rio Grande do Sul (IGeo/UFRGS).

The mineralogy of the tonstein was investigated using disorientedand oriented samples under X-ray Diffraction (XRD—Siemens D 5000—CuKα, 40 mA and 25 kV) and SEM images at a JEOL JSM 6060, bothat UFRGS. The XRD analysis of oriented samples were performed in air-dried (AD), ethylene glycol-saturated (EG) and heated to 550 °C for2 h.

For LAM-MC-ICP-MS U–Pb (Laser Ablation Microprobe, Multi-Collector, Inductively Coupled Plasma, Mass Spectrometer) zircon

Fig. 2. Location map of the studied outcrop (A), as shown in Fig. 1A; lithostratigraphicPaleontologia–Palinologia); general overview of the outcrop (C) and detail of the tonstein l

analyses, a total of 1.5 kg of tonstein sample was used (see Fig. 2D),which was crushed, powdered and sieved, concentrating the fractionbetween 75 and 78 μm. The concentration of heavy mineral wasobtained by panning, which was subsequently purified using bromo-form (tribromomethane, δ=2.89 g/cm3). The zircon grains wereselected andmounted in epoxy resin. Themount surface was polishedto expose the internal part of the grains. Images of zircons wereobtained using an optical microscope and a SEM JEOL JSM 5800 forback-scattering at the IGeo/UFRGS. U–Pb analyses by LAM-MC-ICP-MS were carried out using a Finnigan Neptune coupled to a Nd-YAGlaser (k=213 nm) ablation system (NewWave Research, USA) at theLaboratório de Geologia Isotópica (IGeo/UFRGS).

The analyses were performed in a single spot of 40 μm using thefollowing laser parameters: repetition rate of 10 Hz and energy of 0.5to 1.1 mJ/cm2. Faraday cup configuration of the MC-ICPMS was 206Pb,208Pb, 232Th, 238U and IC's on cup L4 with 202Hg, 204Hg, 204Pb, 207Pb,with acquisition of 50 cycles of 1.048 s of integration time. Main gasflow was 15 l/min Ar, auxiliary gas flow 0.8 l/m, while sample was

al section and sampled levels (B), including palynological slides (MP-P: Museu deevel (D). SB3 is according to Holz et al. (2006).

658 A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

carriedwith 0.75 l/min Ar plus 0.45 l/min He. Unknown analyseswerebracketed by measurements of the international standard GJ-1(Jackson et al., 2004), at every set of four to six zircon spots, andused to estimate the necessary corrections and internal instrumentalfractionation.

For every measured cycle of standard and sample, raw data werecorrected in Office Excel spreadsheet for background, instrumentalmass-bias drift and commonPb. The 204Pb valuewas corrected to 204Hg,considering the 202Hg/204Hg ratio as 4.355. The obtained ratios of207Pb/206Pb and 206Pb/238Uwere corrected for common Pb according toStacey and Kramers (1975) curve, assuming an initial estimated age.After the blank and common Pb corrections, ratios and respectiveabsolute errors (1σ) of 206Pb/238U, 232Th/238U and 206Pb/207Pb werecalculated. The interceptmethod proposed by Youden (1951) was usedassuming linear fractionation of 206Pb/238U. Hence, individual un-certainties are at 1σwhereas calculated ages on the plots are reported atthe 95% of confidence level. Plots on the Tera-Wasserburg diagram andages were calculated using ISOPLOT/Ex (Ludwig, 2003).

4. Results

4.1. Palynostratigraphic analysis

Rich and diversified spore–pollen associations were retrieved fromthe samples analyzed. A total of 76 species of palynomorphs wererecorded, corresponding to spores species (42), pollen grains (24) aswell asmicroplanktonic species (10). Samples from the Rio Bonito andPalermo formations showed compositional variations. In the RioBonito Formation, spores assigned to Punctatisporites, Granulatispor-ites, Horriditriletes, Lophotriletes are very abundant; pollen grains areless frequent. On the other hand, samples from the Palermo Formationrevealed high frequency of spores, mainly Punctatisporites andLundbladispora, and abundance of pollen grains, especially taeniate(Protohaploxypinus, Lunatisporites and Striatopodocarpites) and poly-plicate grains (Vittatina and Weylandites). Species of Zignemataceanalgae were also recorded (Brazilea, Tetraporina and Quadrisporites),which are more common from the Rio Bonito Formation. Palyno-morphs related to Fungi were also obtained in the two lithostrati-graphic units (e.g., Portalites gondwanensis). The complete taxonomiclist and photomicrographs were presented by Mori and Souza (2010).Stratigraphic distribution of each taxon and their respective relativefrequencies are shown in Tables 2 and 3. Furthermore, selectedspecies are illustrated in Fig. 3, mainly those ones of biostratigraphicrelevance.

The presence of Granulatisporites austroamericanus and Vittatinasubsaccata allowed the placement of C1 to C3 levels in the P. goraiensisSubzone, basis of the V. costabilis Zone of Souza and Marques-Toigo(2005). Immediately above, the records of Illinites unicus, Hamiapol-lenites fusiformis, P. goraiensis, V. costabilis, V. subsaccata, L. virkkiae andWeylandites lucifer within the C4 level suggest correlation with the L.virrkiae Zone of Souza and Marques-Toigo (2005). This biozone wasalso confirmed in C7 and C8 samples, based on the records of certainpollen grains (Lunatisporites variesectus, Striatopodocarpites cancella-tus, S. fusus, Striatoabieites multistriatus, L. virkkiae and W. Lucifer).Palynological data retrieved from C5 and C6 levels were not sufficientfor a biostratigraphic analysis, but they are assumed to be within inthe lower L. virkkiae Zone, taking into account species recognized fromthe underlying C4 level.

4.2. Absolute age dating

The tonstein outcrops as a thin (less than 5 cm), laterallycontinuous white layer (see Fig. 2D). Upper and lower abrupt contactswith the coal bed show no evidences of reworking. Compressed plantremains of undetermined origin occur at the top and the base of thetonstein layer. According to X-ray diffraction performed in both

disoriented and oriented samples, after semi-quantitative estima-tions, the tonstein consists of kaolinite (60%), smectite (30%) andquartz (10%) (Fig. 4A). In separate clay fraction, the proportions arekaolinite (90%) and smectite (10%), with no quartz (Fig. 4B). The clayminerals were also observed in SEM images (Fig. 4C and D). Thepoorly preserved primary minerals and transformation to kaolinitehave been described in the tonsteins intervals within the Permiancoals from Candiota region (Formoso et al., 1997, Matos et al., 2000)and Faxinal area (Simas et al., 2003; Guerra-Sommer et al., 2008c).These features are interpreted as resulting from transformation ofvolcanic ashes with provenance from felsic igneous distant source.

Morphological analysis on ca. 200 zircons collected from thetonstein, described using optical microscope and backscatteringimage, allowed the recognition of three different populations: twodominant and one subordinate. In addition, a couple of rounded grains,clearly detrital, were analyzed. These results are shown in the Table 4.

The zircons from the two dominant populations are transparent,colorless, prismatic, highly crystalline with sharp edges, typical ofmagmatic grains with no epiclastic transportation. However, onepopulation is comprised by 100 to 200 μm elongated zircons (4:1length/width) with some internal zoning, andmore fragile due to longvesicles aligned along C axis, while the second population consists of a100 to 150 μm short grains (2:1 to 3:1) with abundant inclusions(Fig. 5). Both populations were firstly believed to be coeval.Nevertheless, after LAM-MC-ICP-MS analysis, the results showed asignificant geochemical difference in Th/U ratios from each popula-tion, suggesting more than one magmatic episode. The Th/U ratio inthe first group is less than 1 (0.33 to 0.92), while in the secondpopulation it ranges from 2.03 to 4.74.

Plotting the results on the Tera-Wassenburg diagram, the low Th/Uratio group yields a 207Pb/206Pb age of 281.4±3.4 Ma (10 out of 11points, with 95% confidence, MSWD=0.46), which is considered thedepositional age of the volcanic ash (Fig. 6A). On the other hand, the207Pb/206Pb age of the shorter, inclusion-rich zircons with higher Th/Uratio is 295.4±4.5 Ma (12 out of 13 points, with 95% confidence,MSWD=2.5), which meaning of is addressed later (Fig. 6B).

The subordinate population constituted of fragmented, subhedral,short to long grains (2:1 to 4:1) with 100 to 200 μm, internal zoningand some ablation on the edges was interpreted as inherited grains(Fig. 7A and B). Two rounded grains presented Palaeoproterozoic toArchean age (see Fig. 7C and Table 3). Results on the 206Pb/238PbConcordia diagram (8 out of 9 points, with 95% confidence,MSWD=0.33) revealed a 1551±15 Ma upper intercept age (Fig. 8).

5. Discussion

5.1. Stratigraphy

Comparisons between other nearby sections were needed for abetter stratigraphical analysis, mainly to confirm the stratigraphicalpositioning of the outcrop studied herein. In this way, outcrops ofCandiota coal mine and subsurface data from two boreholes drilled inthe adjacent area (HN120RS and SV332RS) were selected for analysis.Location and geological descriptions of these other deposits arepresented in Figs. 2A and 9, respectively. As previously shown, severalcoal beds occur in the Candiota coal mine, among which “Lower” and“Upper Candiota Coal Seams” are the most expressive and useful onesfor correlations (Aboarrage and Lopes, 1986), due their wide lateraldistribution, observed even in the Uruguayan portion of the ParanáBasin. A thin claystone level (up to 1 m in thickness) occurs betweenthese two main coal beds, which is used as datum for correlation ofthese three sections (see Fig. 9). Radiometric age datings previouslypublished from Candiota coal mine (Matos et al., 2001; Guerra-Sommer et al., 2005; Rocha-Campos et al., 2006, 2007; Guerra-Sommer et al., 2008a,b) are derived from two tonsteins, named as“Tonstein A” and “Tonstein C”, which occur within the “Lower” and

Table 2Stratigraphic distribution and relative abundance of the recovered spore species. Frequences are interpreted as: x—rare (frequenceb2%); xx—commom (frequence between 2and 5%); xxx—abundant (frequence N5%).

Taxa Palynological samples

C1 C2 C3 C4 C5 C6 C7 C8

Apiculatisporis levis Balme and Hennelly 1956 x xApiculiretusispora sparsa Menendez and Azcuy 1979 x xBrevitriletes cornutus (Balme and Hennelly)Backhouse, 1991

x

Brevitriletes irregularis (Nahuys, Alpern and Ybert)Césari, Archangeslky and Seoane 1995

x x x

Calamospora exigua Staplin 1960 xxxCalamospora inderjensis (Malayavkina ex Luberand Valts) Hart 1965

xxx xxx

Calamospora liquida Kosanke 1950 x xCalamospora plicata(Luber and Valts) Hart 1965 x x xx xxCalamospora sinuosa Leschik 1955 x xxConverrucosisporites confluens(Archangeslky and Gamerro)Playford and Dino, 2002

x x x

Converrucosisporites micronodosus (Balmeand Hennelly) Playford and Dino, 2002

xxx x

Convolutispora candiotensis Ybert, 1975 x x xCristatisporites inconstans Archangeslky andGamerro 1979

x

Cyclogranisporites gondwanensis Bharadwaj andSalujha 1964

xx x xxx x xxx xx x

Cyclogranisporites parvigranulosus (Leschik) Ybert, 1975 x xxxDiatomonozotriletes subbaculiferous (Nahuys, Alpernand Ybert) Césari, Archangeslky and Seoane 1995

x xx

Foveolatisporites sp. xFoveosporites sp. xGranulatisporites austroamericanus Archangeslkyand Gamerro 1979

xxx x x xxx

Horriditriletes curvibaculosus Bharadwajand Salujha 1964

x x x x x

Horriditriletes gondwanensis (Tiwari and Moiz) Foster 1975 x xHorriditrieletes pathakeraensis Anand-Prakash 1970 x xHorriditriletes ramosus (Balme and Hennelly)Bharadwaj and Salujha 1964

x x

Horriditriletes superbus (Foster) Césari, Archangelskyand Seoane, 1995

xx x x

Horriditrieletes uruguaiensis (Marques-Toigo)Archangeslky and Gamerro 1979

x x xx x

Laevigatosporites desmoinensis (Wilson and Coe) Shopf,Wilson and Bentall 1944

x x x x

Laevigatosporites vulgaris Ibrahim 1933 xx x xx x xxLeiotriletes virkkii Tiwari 1965 xx x xx x xx x xxxLophotriletes pseudoaculeatus Potonié and Kremp 1955 xxx x xxx xxx xxx xxx x xxxLundbladispora braziliensis (Marques-Toigo and Pons)Marques-Toigo and Picarelli 1984

x x xxx xxx

Lundbladispora riobonitensis Marques-Toigoand Picarelli 1984

x xxx xxx

Murospora bicingulata Ybert, 1975 x xPsomospora sp. xPunctatisporites gretensis Balme and Hennelly 1956 xxx x xxx x xx xxx xxxReticulatisporites pseudopalliatus Staplin 1960 xRetusotriletes golatensis Staplin 1960 xxx xRetusotriletes nigritellus (Luber) Foster 1975 x xx xRetusotriletes simplex Naumova 1953 x xVallatisporites splendens Staplin and Jansonius 1962 x x xVallatisporites vallatus Hacquebard 1957 x x x xVerrucosisporites pseudoreticulatus forma minorYbert 1995

xxx x

Verrucosisporites sp. xV. costabilis Zone–P. goraiensisSubzone

L. virkkiae Zone

659A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

“Upper Candiota Coal Seams”, respectively. These deposits are locatedstratigraphically below the sections studied herein. The three coallevels of our outcrop were denominated as Cs4, Cs5 and Cs6(Aboarrage and Lopes, 1986), which correspond respectively to theC2, C4 and C6 palynological samples. These coals levels are notpreserved in the Candiota coal mine, but they were recognized in theHN120RS and SV332RS boreholes, just below of an expressive surface

boundary (SB3), that affected the uppermost deposits of the RioBonito Formation in this region (Holz et al., 2006).

5.2. Palynological correlation within Paraná Basin

Coal beds of southern Brazil were firstly included in theCaheniasaccites ovatus Subzone, middle portion of the Cannanoropollis

Table 3Stratigraphic distribution and relative abundance of the recovered pollen grains and microplanktonic species. Frequences are interpreted as: x—rare (frequenceb2%); xx—commom(frequence between 2 and 5%); xxx—abundant (frequence N5%).

Taxa Palynological samples

C1 C2 C3 C4 C5 C6 C7 C8

Pollen grainsCaheniasaccites elongatus Bose and Kar 1966 x xCycadopites sp. x xDistriomonaccites crucistriatus (Ybert) Césari et al., 1995 x x xHamiapollenites fusiformis Marques-Toigo 1974 x xIllinites unicus (Kosanke) Jansonius and Hills, 1976 xLimitisporites rectus Leschik 1955 xLueckisporites virkkiae (Potonié and Klaus) Clarke 1965 x xLunatisporites variesectus Archangelsky and Gamerro, 1979 x xMarsupipollenites sp. cf. Marsupipollenites striatus (Balmeand Hennelly) Foster 1975

x

Pakhapites fusus (Bose and Kar) Menendez 1971 x x xxPakhapites ovatus (Bose and Kar) Playford and Dino 2000 x x xx xPotonieisporites lelei Maheshwari 1967 xProtohaploxypinus goraiensis (Potonié and Lele) Hart, 1964 xProtohaploxypinus limpidus (Balme and Hennelly) Balmeand Playford 1967

x

Protohaploxypinus perfectus (Nauomova) Samoilovich 1953 xProtohaploxypinus samoilovichii (Jansonius) Hart, 1964 xStriatoabieites multistriatus (Balme and Hennelly) Hart, 1964 xStriatopodocarpites cancellatus (Balme and Hennelly) Hart 1963 xStriatopodocarpites fusus (Balme and Hennelly) Potonié 1958 xStriomonosaccites sp. cf. Striomonosaccites ovatusBharadwaj 1962

x

Vittatina costabilis Wilson 1962 x xVittatina saccata (Hart) Playford and Dino 2000 xxx xxxVittatina subsaccata Samoilovich 1953 x x xxx xxxWeylandites lucifer (Bharadwaj and Salujha) Foster 1975 x x x

Chlorophycean AlgaeBotryococcus braunii Kutzing 1849 xxBrazilea scisa (Balme and Hennelly) Foster 1975 x x xCongoites sp. xx xxLeiosphaeridae sp. xPilasporites calculus (Balme and Hennelly) Tiwariand Navale, 1967

xx xx

Quadrisporites horridus (Hennelly) Potonié and Lele 1961 x xxQuadrisporites lobatus (Tiwari and Navale) Ybert, 1975 xx xTasmanites sp. xTetraporina punctata (Tiwari and Navale) Kar and Bose 1976 x xx xx

FungiPortalites gondwanensis Nahuys, Alpern and Ybert 1968 xx xx x

V. costabilis Zone–P. goraiensisSubzone

L. virkkiae Zone

660 A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

korbaensis Zone of Marques-Toigo (1988, 1991). This zone wasrenamed as V. costabilis (Souza and Marques-Toigo, 2001, 2003).The C. ovatus Subzone was regarded an ecofacies within the P.goraiensis Subzone (Souza and Marques-Toigo, 2005). Thus, most ofthe coal beds from the Rio Bonito Formation in Rio Grande do Sul andSanta Catarina states were included in the underlying P. goraiensisSubzone. Several coal beds and related strata are being reexamined inorder to confirm this assumption (e.g., Smaniotto et al., 2006).Palynological associations previously described for the Rio BonitoFormation in the Candiota region (e.g., Nahuys et al., 1968; Ybert,1975; Corrêa da Silva and Marques-Toigo, 1975; Meyer and Marques-Toigo, 2000) are characterized by the abundance and diversity oftrilete spores, with scarce striate pollen grains species.

According to Cazzulo-Klepzig et al. (2002), the upper coal layer ofthe Candiota coal mine is correlated with the P. goraiensis Subzone.However, data provided by these authors do not allow a betteranalysis, since some of the taxa are listed only at the genus level.Cazzulo-Klepzig et al. (2005) reanalyzed the palynologic content ofseveral core samples from the Candiota region, resulting in paleo-floristic and paleoenvironmental reconstructions, based on quantita-tive data of the genera recorded. The absence of both a detailedtaxonomic list and diagnostic forms for biostratigraphic use preventsa more precise biostratigraphic placement of the levels studied by

those authors. Furthermore, in these papers, photomicrographs ofpalynomorphs are of uncertain origin, because some of these sampleswere originally described from other sites in the Paraná Basin (seeSantos et al., 2006, p. 461). Results provided by Cazzulo-Klepzig et al.(2002, 2005) were used by Guerra-Sommer et al. (2008a,b) for thebiostratigraphic correlation of the Candiota coal beds, which pre-sented new absolute datings from the tonsteins in the coalmine.Therefore, correlation provided by those authors is very poorrequiring the refinement of the taxonomy and a proper stratigraphiccontrol. The present contribution represents a step in the definition ofnew palynologic associations, with stratigraphical data from thesampling and original illustrations, as already detailed by Mori andSouza (2010).

The presence of G. austroamericanus and V. subsaccata in C1 andC3 levels suggests correlation with the P. goraiensis Subzone, basis ofthe V. costabilis Zone. The records of I. unicus, H. fusiformis, P.goraiensis, Vittatina costabilis, V. subsccata, L. virkkiae and W. luciferimmediately above allow to link the C4 level to the L. virrkiae Zone.This latter zone is confirmed in C7 and C8 samples, due to thepresence of L. variesectus, S, cancellatus, S. fusus, S., L. virkkiae and W.lucifer. Although the palynological content retrieved from the C5 andC6 levels are insufficient for biostratigraphic analysis, their biostrat-igraphic placement is based on diagnostic fossils identified on the

Fig. 3. Selected palynomorphs retrieved from the outcrop studied (MP-P: Museu de Paleontologia–Palinologia. Coordinates are given in England Finder Coordinates).A. Granulatisporites austroamericanus Archangelsky and Gamerro, 1979 (MP-P 5677, J54-3). B. Converrucosisporites confluens (Archangelsky and Gamerro) Playford and Dino, 2002(MP-P 5674, S50). C. Illinites unicus (Kosanke) Jansonius and Hills, 1976 (MP-P 5677, R59-3). D. Protohaploxypinus goraiensis (Potonié and Lele) Hart, 1964 (MP-P 5676, F43-2).E. Protohaploxypinus limpidus (Balme e Hennelly) Balme and Playford 1967 (MP-P 5685, N49-3). F. Striatopodocarpites cancelattus (Balme and Hennelly) Hart 1963 (MP-P 5692, U40-2).G. Striatopodocarpites fusus (Balme and Hennelly) Potonié 1958 (MP-P 5693, Q49). H. Lunatisporites variesectus Archangelsky and Gamerro, 1979 (MP-P 5692, F46-2). I. Hamiapollenitesfusiformis Marques-Toigo 1974 (MP-P 5693, W65). J. Lueckisporites virkkiae (Potonié and Klaus) Clarke 1965 (MP-P 5692, R43-2). K. Vittatina costabilis Wilson 1962 (MP-P 5692, J47).L.Weylandites lucifer (Bharadwaj and Salujha) Foster 1975 (MP-P 5694, H37-3). Scale bar: 10 μc.

661A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

subjacent and overlying samples; then, they are linked to the L.virrkiae Zone.

Taking into account the palynostratigraphic scheme of Souza andMarques-Toigo (2005), some species showed variations in their firstand last appearance, as discussed by Mori and Souza (2010). Therecord of species stratigraphically restricted to the V. costabilis Zoneassociated with others elements from the L. virrkiae Zone in the C4level, suggests that those species could have a larger stratigraphicrange, or that certain species of the L. virrkkiae Zone have firstappearances in a lower stratigraphic position. Both hypotheses arebeing studied by the present authors taking into considerationselected boreholes in the region, which exhibit a more completesection of this interval.

According to Holz et al. (2006), tectonic movements in this regioncould be the responsible for changes in the paleo-relief and thecoastline. The upper portion of the Rio Bonito and the lower portion ofthe Palermo formations correspond to the sequences 2 and 3 (S2 andS3) respectively, separated by the regional boundary sequence SB3.During the initial transgression that generated the SB3, uppermost

deposits of the Rio Bonito Formation were strongly eroded and, as aconsequence, the youngest part of the V. costabilis Zone is not fullypreserved. This stratigraphical model justifies the absence of the H.karrooensis Subzone and the record of the L. virrkiae immediatelyabove of the P. goraiensis Subzone.

The basal occurrence of the L. virkkiae Zone within the uppermostRio Bonito Formation confirms the Daemon and Quadros (1970) zonalscheme's for the Paraná Basin, once several index species of this zonewas also recorded within the interval K. However, the palynologicalassociations reported herein are clearly different from other onesalready published for Candiota Coalmine, which presented assemblieslinked to the underlying V. costabilis Zone.

5.3. Comparisons with other Gondwanan palynologic assemblages

The palynological associations studied herein are very similar incomposition to those ones described from Permian deposits of SouthAmerica, especially in the northern portion of Paraná Basin (Souza etal., 1999; Souza and Callegari, 2004), and from Argentina (e.g., Césari

Fig. 4. X-ray diffractometry (A and B) and scanning electron images of clay mineral components (C and D).

662 A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

and Gutiérrez, 2000) and Uruguay (Beri and Daners, 1996, 1998), aswell as other Gondwanan sectors, such as those found in currentlyAntarctica and Australia (e.g., Lindström, 1995a,b; Stephenson, 1998).

The most recent correlations between Permian palynozones ofSouth American were presented by Césari and Gutiérrez (2000),Azcuy et al. (2007) and Souza et al. (2007). According to Stephenson(2008), precise correlation for the different Gondwanan basins ishampered by the lack of standardization in the palynologic docu-mentation and by local phytogeographic variations. However, theincreasing in abundance and diversification of specific sporomorphgroups such as monosaccate pollen grains, Cheilocardioid type spores(“trilete cryptogam spores with proeminent labra and heart-shapedoutline in lateral view”, as cited by Balme, 1980, p. 49), as well ascertain bissacate and taeniate pollen grains, may reflect key eventsacross the Gondwana, in response to paleoclimatic control inequivalent paleolatitudes.

Based on Balme (1980), palynological associations of severalGondwanan sequences in the Lower Permian (Australia, SouthAmerica, Pakistan, India and southern Africa) show significantchanges close to the end of the glacial period. In this way, evolutionof different Gimnospermic groups was strongly favored, leading to thediversification and abundance of bissacate and taeniate pollen grains,in palynological terms. This characteristic is clearly observed inseveral Gondwana palynozonations for the Pennsylvanian–Permianinterval, for example, in South America. Even when there are localvariations in the stratigraphic ranges of the constituent elements ofthe different biozones, the V. costabilis and L. virkkiae zones can becorrelated, respectively, to: (1) the Cristatisporites and the Striatiteszones, of the Chaco-Paraná Basin (Russo et al., 1980; Vergel, 1993);(2) the F. fusus–V. subsaccata and the Lueckisporites–Weylanditeszones of the middle-west Argentina (Césari and Gutiérrez, 2000); and(3) the Cristatisporites inconstans–V. subsaccata and the Striatoabieitesanaverrucosus–Staurosaccites cordubensis zones of the PermianUruguay (Beri et al., 2004). The definition of these biozones ischaracterized by the presence and/or abundance of certain bissacateand taeniate genera (e.g., Protohaploxypinus, Lueckisporites, Vittatina),

as well as the abundance and diversification of these major pollengrain groups.

Comparisons of the V. costabilis and L. virkkiae zones can be madealso with other palynofloras described outside South America. InAfrica, the I/II zones of Falcon (1975) and the Hamiapollenitesbullaeformis/Cyclogranisporites gondwanensis zones of Modie and LeHerissé (2009) are related to the V. costabilis Zone; on the other hand,the L. virrkiae Zone finds correspondence with the III/IV zones (Falcon,1975) and to the Platysaccus papilionnis Zone of Modie and Le Herissé(2009). Based on data derived from of Antarctica, the palynoflorasfrom A and C sites of Lidkvarvet (Lindström, 1995a) are related to theV. costabilis Zone, while palynoassemblages from Fossilryggen andNunatak (Lindström, 1995b) are correlated to the L. virrkiae Zone. InAustralia, the Microbaculispora tentula zone (Jones and Truswell,1992) and the Stage 2, Pseudoreticulatispora confluens and P. reticulatazones of Backhouse (1991) are similar to the V. costabilis Zone, whilethe Striatopodocarpites fusus zone (Backhouse, 1991) are referred tothe L. virrkiae Zone. These palynozones show similar compositions,and the same general pattern verified in South America, with theprogressive increasing of bisaccate pollen grains, especially taeniateones, throughout the Permian.

5.4. Biostratigraphy×Absolute age dating

The V. costabilis Zone encompasses almost entirely the Rio BonitoFormation, including all coal beds. This palynozone is dated as EarlyCisularian (Souza and Marques-Toigo, 2003, 2005; Souza, 2006).Particularly, the P. goraiensis Subzone, base of the V. costabilis Zone,ranges approximately from the Asselian to the Sakmarian (299.0±0.8 Ma to 284.4±0.7 Ma), whereas the base of the H. karrooensisSubzone is around the Sakmarian/Artinskianboundary (284.4±0.7 Ma).

Based on the biostratigraphic discussion previously presented, theC4 level is positioned at the L. virkkiae Zone (Fig. 10). Because theuppermost V. costabilis Zone is not recognized here and the tonsteinlevel is located immediately below the C4 level, it is assumed that thenew absolute age dating is equivalent to the base of the L. virkkiae

Table 4In situ LAM-MC-ICP-MS zircon data of the Rio Bonito Formation analyzed herein. Corrections are 1-sigma (% for isotope ratios, absolute for ages). Rho=correlation factor between the errors of the 207Pb/235U and 206Pb/238U.Disc.% means discordance in%. Correction for common Pb: ƒ206=(206Pb/204Pb)common/(206Pb/204Pb)sample. * Samples plotted but removed from age calculation.

Spot number Analyses from low Th/U zircons

Age (Ma)

207Pb/235U Error 206Pb/238U Error Rho 238U/206Pb Error 207Pb/206Pb Error 206Pb/238U Error 207Pb/235U Error 207Pb/206Pb Error 232Th/238U Disc.% ƒ206

A-I-01 0.33979 4.94 0.04585 2.27 0.5 21.80819 2.27 0.05374 4.39 289 7 297 15 360 16 0.33 20 0.002A-I-10 0.3324 5.9 0.04474 2.83 0.5 22.35086 2.83 0.05388 5.18 282 8 291 17 366 19 0.69 23 0A-I-12* 0.34307 3.92 0.04792 2.29 0.6 20.86861 2.29 0.05192 3.18 302 7 299 12 282 9 0.35 −7 5E-04A-I-25 0.32465 7.18 0.04562 2.3 0.3 21.91856 2.3 0.05161 6.8 288 7 285 20 268 18 0.24 −7 0.002A-I-29 0.3222 3.31 0.04445 2.32 0.7 22.4949 2.32 0.05257 2.36 280 7 284 9 310 7 0.75 10 0.004A-I-35 0.30876 8.61 0.04421 2.69 0.3 22.61847 2.69 0.05065 8.18 279 7 273 24 225 18 0.86 −24 0.002C-III-19 0.31551 2.44 0.04432 1.84 0.8 22.56251 1.84 0.05163 1.59 280 5 278 7 269 4 0.79 −4 6E−04C-III-11 0.31634 2.53 0.04488 1.44 0.6 22.2832 1.44 0.05112 2.08 283 4 279 7 246 5 0.68 −15 6E−04C-III-12 0.32052 2.43 0.04449 1.41 0.6 22.47871 1.41 0.05226 1.98 281 4 282 7 297 6 0.42 5 6E−04C-III-16 0.31271 3.51 0.04407 2.3 0.7 22.68917 2.3 0.05146 2.65 278 6 276 10 261 7 0.87 −6 0.001C-III-09 0.31261 2.86 0.04365 1.98 0.7 22.90893 1.98 0.05194 2.06 275 5 276 8 283 6 0.92 3 2E−04

Spot number Analyses from high Th/U zircons

Age (Ma)

207Pb/235U Error 206Pb/238U Error Rho 238U/206Pb Error 207Pb/206Pb Error 206Pb/238U Error 207Pb/235U Error 207Pb/206Pb Error 232Th/238U Disc.% ƒ206

C-III-17 0.32617 2.87 0.04521 1.53 0.5 22.11801 1.53 0.05232 2.43 285 4 287 8 300 7 4.14 5 0.001A-I-18b 0.32078 7.57 0.0453 2.07 0.3 22.07555 2.07 0.05136 7.29 286 6 282 21 257 19 2.15 −11 0.003A-I-20 0.32763 7.15 0.04629 2.45 0.3 21.60331 2.45 0.05133 6.72 292 7 288 21 256 17 3.58 −14 0.002A-I-06 0.32923 5.49 0.04633 2.49 0.5 21.58291 2.49 0.05154 4.89 292 7 289 16 265 13 4.4 −10 0.001A-I-04 0.3294 6.47 0.046 2.41 0.4 21.73913 2.41 0.05194 6.01 290 7 289 19 283 17 2.88 −3 0.002A-I-05* 0.3143 6.01 0.04349 3.24 0.5 22.99251 3.24 0.05241 5.06 274 9 278 17 303 15 3.49 10 0.011A-I-11 0.33982 4.65 0.0458 2.58 0.6 21.83235 2.58 0.05381 3.87 289 7 297 14 363 14 4.08 20 0.013A-I-32 0.34164 8.25 0.04572 2.03 0.3 21.87149 2.03 0.05419 8 288 6 298 25 379 30 3.76 24 0.01A-I-13 0.35056 5.02 0.04993 2.28 0.5 20.02765 2.28 0.05092 4.47 314 7 305 15 237 11 3.09 −32 0.001A-I-15 0.35084 5.2 0.04861 2.35 0.5 20.57181 2.35 0.05234 4.64 306 7 305 16 301 14 4.74 −2 0.002A-I-16 0.355 3.56 0.04957 2.24 0.6 20.17357 2.24 0.05194 2.77 312 7 308 11 283 8 2.03 −10 7E−04A-I-23 0.34925 8.67 0.04782 2.13 0.3 20.91163 2.13 0.05297 8.41 301 6 304 26 327 28 2.17 8 0.005A-I-27 0.34118 9 0.04784 2.82 0.3 20.90124 2.82 0.05172 8.54 301 8 298 27 273 23 4.45 −10 0.001

Spot number Analyses from inherited and/or detrital zircons

Age (Ma)

207Pb/235U Error 206Pb/238U Error Rho 238U/206Pb Error 207Pb/206Pb Error 206Pb/238U Error 207Pb/235U Error 207Pb/206Pb Error 232Th/238U Disc.% ƒ206

C-III-05* 3.2206 1.57 0.24896 0.91 0.6 4.01666 0.91 0.09382 1.28 1433 13 1462 23 1505 19 0.36 5 0.003A-I-10 3.72733 1.6 0.28122 1.06 0.7 3.556 1.06 0.09613 1.2 1597 17 1577 25 1550 19 0.41 −3 0.001C-III-08 3.54165 1.78 0.26954 1.35 0.8 3.71005 1.35 0.0953 1.16 1538 21 1537 27 1534 18 0.34 0 6E−04A-I-04 3.73556 1.31 0.27981 1.05 0.8 3.57387 1.05 0.09683 0.77 1590 17 1579 21 1564 12 0.37 −2 0.002A-I-06 3.54834 1.6 0.26724 1.25 0.8 3.74202 1.25 0.0963 1 1527 19 1538 25 1554 16 0.48 2 0.002A-I-01 3.48277 0.92 0.2635 0.35 0.4 3.79501 0.35 0.09586 0.85 1508 5 1523 14 1545 13 0.29 2 4E−04A-I-02 3.47594 2.45 0.26109 1.69 0.7 3.83008 1.69 0.09656 1.77 1495 25 1522 37 1559 28 0.5 4 0.003C-III-14 3.31287 1.23 0.25018 0.91 0.7 3.99713 0.91 0.09604 0.83 1439 13 1484 18 1549 13 0.47 7 7E−04C-III-04 3.59787 1.21 0.2735 0.59 0.5 3.65636 0.59 0.09541 1.06 1559 9 1549 19 1536 16 0.43 −1 3E−04A-I-07 10.49591 2.66 0.43084 0.35 0.1 2.32105 0.35 0.17669 2.64 2310 8 2480 66 2622 69 0.48 12 5E−04A-I-12 13.16564 0.68 0.5116 0.31 0.5 1.95465 0.31 0.18664 0.61 2663 8 2692 18 2713 17 0.26 2 3E−04

663A.L.O

.Mori

etal./

Gondw

anaResearch

21(2012)

654–669

Fig. 5. Back-scattered electron images of low Th/U zircons (A, B, C) and high Th/U zircons (D, E, F) from tonstein sample. Dated spots are indicated with their respective age.

Fig. 6. Tera-Wasserburg diagrams for zircons from the tonstein analyzed by LAM-MC-ICP-MS. A. Low Th/U zircons meaning depositional age of the tonstein. B. High Th/U zirconsinherited during volcanic explosion.

Fig. 7. Back-scattered electron images of inherited zircons (A, B) and rounded detrital zircons (C).

664 A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

Zone. In the studied outcrop, the base of this last biozone wasrecognized by the first appearance of L. virkkiae and W. luciferdiagnostic species, represented by few specimens and in associationwith certain index species of the underlying V. costabilis Zone.

So, the absolute age of 281.4±3.4 Ma presented herein places thelocal upper portion of the Rio Bonito Formation in the middleArtinskian (Fig. 10). These new absolute age dating and palynologicaldata have local and regional significance, with implications on

lithostratigraphy, sequence stratigraphy and biostratigraphy forParaná Basin and correlated strata over the Gondwana. The absolutedating is derived from a level of tonstein which is located ca. 2 mbelow the lithostratigraphical contact between the Rio Bonito and thePalermo formations, which represent two distinct general post glacialenvironments developed within the Paraná Basin during the Permian.This limit corresponds to a regional unconformity named as SurfaceBoundary 3 (SB3) by Holz et al. (2006), with affected the top of the

Fig. 8. Concordia diagram for inherited Mesoproterozoic zircons.

665A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

former unit in various parts of the basin, with a great potential forcorrelation. Furthermore, the basal limit of the L. virkkiae Zone wasrecognized within the layer immediately above the tonstein bed. Thiszone is characterized by certain species of pollen grains which arewelldistributed inmost part of the Gondwana, and can be used as guide forage assignments and recognition of floral changes.

The absolute age dating of the volcanic zircons of 281.4±3.4 Ma isconsidered here as the depositional age of the ashes. The provenanceof the tonteins interbedded within coal layers in southern Rio Grandedo Sul seems to be derived from a distant felsic volcanic source,probably Choyoi Province, central-western Argentina, which charac-

Fig. 9. Geological sections of the HN120RS and SV332 boreholes and Candiota

terizes a long-lived eruption center, active for more than 30 millionyears (e.g., Coutinho and Hachiro, 2005).

Recently, Rocha-Campos et al. (2011) compared the volcanic rocksfrom Choyoi province to the ashes found in Paraná Basin providingnew geochronological data and a thoughtful discussion. Two agesrevealed by these authors are of interesting significance. The dating of281.4±2.5 Ma (zircon/SHRIMP) obtained at the base of the succes-sion of the Los Reyunos Formation (Cochicó Group) is very similar tothe one presented here (281.4±3.4 Ma), which was interpreted byRocha-Campos et al. (2011) as “the crystallization age of the zirconsand the record of the initial volcanic Permian episode. In addition, an297.2±5.3 Ma provenance agewas found in a sedimentary layer fromthe El Imperial Formation, just below the Choyoi succession, which issimilar to the 295.4±4.5 Ma obtained here also. Rocha-Campos et al.(2011) also report zircons of Mesoproterozoic (Grenvillian) age fromthis same layer, a basement age that is common in central-westernArgentina. Similarly, our older zircons found in the same tonstein areinterpreted as inherited from an outside volcanic province, as no earlyPermian or Mesoproterozoic magmatic rocks have been describedclose the southmost portion of the Paraná Basin. In addition, it isknown that debris from a long-lived volcanic pile or from countryrocks are normally incorporated in the volcanic column duringmagma migration or explosion, mixing material from different ages.

Comparisons with other Gondwanan age data for the Paraná Basinare needed, mainly those ones derived from Candiota Coalmine(Fig. 10). As previously shown, datings obtained to the Candiota coalmine ranges from 267.1±3.4 Ma (Matos et al., 2001) to 299.1±2.6 Ma (Guerra-Sommer et al., 2005). Regarding the discrepanciesof ages reported for tonsteins from Candiota and surroundingregion, is important to stress that older ages from previous papers(Guerra-Sommer et al., 2005, 2008a,b) were obtained from layers

Coalmine, which are used for correlation with the outcrop studied herein.

Fig. 10. Integrated scheme showing the palynozones and radiometric datings obtained for the Upper Paleozoic deposits of Paraná Basin. The references are related to radiometricdatings available for the Rio Bonito Formation: (1) Matos et al. (2001); (2) Guerra-Sommer et al. (2008a); (3) Guerra-Sommer et al. (2005, 2008b); (4) Rocha-Campos et al. (2006,2007) and Guerra-Sommer et al. (2005), including the new absolute dating presented herein (☼) and those ones obtained for the Irati Formation (*1: Santos et al., 2006; *2: Rocha-Campos et al., 2006, 2007).

666 A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

stratigraphically below the level sampled in this study. The maximumages of 299±2.6 Ma and 296±1.4 Ma (Guerra-Sommer et al., 2005)and 296.9±1.65 Ma to 296±4.2 Ma (Guerra-Sommer et al., 2008a),place the middle portion of the Rio Bonito Formation within theAsselian. However, these same ages were reassessed by Guerra-Sommer et al. (2008b), who obtained a main age of 290.6±1.5 Ma(average), which implies that the middle part of the Rio BonitoFormation belongs to the Sakmarian. Thus, it should be noted thatbesides the stratigraphical differences between tonstein samples,some of these contributions may be marred by methodologicalproblems with respect to the isotopic age analysis, if we consider thewide variation in the spectrum of the ages coming from the sametonstein level.

A more relevant inconsistency is the Wordian age of 267.1±3.4 Ma (Matos et al., 2001) to the Rio Bonito Formation whencompared to that Artinskian ages reported to the Irati Formation of278.4±2.2 Ma (Santos et al., 2006) and 279.9±4.8 Ma (Rocha-Campos et al., 2007), which is located above the Palermo Formationand, consequently, younger than Rio Bonito Formation. However, theage reported by Matos et al. (2001) was the first attempt to date thetonsteins and was obtained from an inferior intercept at Concordiadiagram, using discordant data and inherited zircon, which is not idealand may not represent the true age of the layer.

On the other hand, the ages proposed by Guerra-Sommer et al.(2008a,b) and that one obtained herein do not seem to conflict withthat proposed to the Irati Formation (Santos et al., 2006; Rocha-Campos et al., 2007). This unit comprises the most reliable agehitherto for the Permian deposits of the Paraná Basin, includingassociations recognized in Brazilian portion of this basin and withinother Gondwana adjacent areas, such as Uruguay, Argentina andsouthern Africa, where is known through different names (Mangrullo,White Hill, Huab and B2 formations). Fossil vertebrates (Mesosaurusand Stereosternum) and palynofossils concerning the L. virkkiae Zoneare well distributed in these units, suggesting accurate correlation.Geochemical evidence (stable isotopes, hydrogen index, microbial

blooms) corroborates these palaeontological data, demonstrating thatthis unit, distributed in different portions of the Occidental Gondwanahas stratigraphic equivalence (Faure and Cole, 1999).

Palynomorphs of the Striatites Biozone (see Césari et al., 1996;Césari, 2007) occur immediately above the magmatic deposits datedbyMelchor (2000) in Chacoparaná Basin, Argentina (266.3±0.82 Ma,obtained by the laser total fusion 40Ar/39Ar method on sanidine),concerning the Wordian (Middle Permian, Guadalupian). Thisbiozone is equivalent to the L. virkkiae Zone of the Paraná Basin. Ourdating extends the basal age of the L. virkkiae Zone, to the Artinskian(Early Permian, Cisuralian).

Absolute age data provided by Gulbranson et al. (2010) fromArgentinian sections bearing floras and faunas from the UpperPaleozoic, were the basis of established new chronostratigraphicranges of some palynozones recorded in the western sector ofGondwana (Césari et al., 2011). According to these authors, theabsolute age dating of Gulbranson et al. (2010) of 206U/238Pb of310.63±0.1 Ma is linked to the basal limit of the Pakhapites fusus–Vittatina saccata (FS Biozone), which is correlated to the V. costabilisZone. In addition, Glossopteris leaves were identified in certain stratarelated to this biozone. The Glossopteris Flora is a significantbiostratigraphic marker throughout Gondwana, with the first occur-rences found from the Pennsylvanian/Permian boundary (Césari,2007). An age of 296.09±0.085 was obtained by Gulbranson et al.(2010), just above the levels of occurrence of these Glossopterisremains, Accepting this age means that the lower limit of the V.costabilis Zone could be placed near the Carboniferous–Permianboundary, while its uppermost limit does not exceed the age of281.4±3.4 Ma.

Single taxa can also be useful for correlations across Gondwana.According to Stephenson (2008), C. confluens is one of the mainspecies in the Permian interval that allows this kind of comparison.This species is frequently associated to the end-glacial and post-glacialdeposits. Besides, it shows a wide geographic distribution, fromAntarctica (Lindström, 1995a), Oman and Saudi Arabia (Stephenson,

667A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

1998, 2004), Argentina (Archangelsky and Gamerro, 1979), Brazil(Souza and Callegari, 2004), Uruguay (Beri and Daners, 1996, 1998)and India (Tiwari and Singh, 1981). The first appearances ofConverrucosisporites confluens were registered by Stephenson (2009)in Gondwana deposits of Namibia, which was previously dated byBangert et al. (1999) within the uppermost Pennsylvanian (302±3 Ma, Gzhelian). This species is one of the biostratigraphical indexof the V. costabilis Zone of Paraná Basin. According to Stephenson(2009), the base of this zone could be close to the Pennsylvanian–Permian boundary. The complete range of this species in the ParanáBasin includes post-glacial sections of the Itararé Group up to marinebeds of the Palermo Formation. However, the more consistent recordsare in the Rio Bonito Formation, where most part of the index speciesof the P. goraiensis Subzone are commonly found. Then, the minimumapproximate range of this taxon comprises the Gzhelian to theArtinskian (ca. 20 Ma).

By other hand, the genus Lueckisporites seems to be a significativeindex taxon for correlation within the Gondwana basins, although ithas a long range in the Permian (see Stephenson, 2009, p. 326–327).Earlier appearances of this genus in the Paraná Basin are related to theuppermost portion of the Rio Bonito Formation (middle Artinskian),characterizing the basal limit of the eponymous Zone, to which thesamples C4–C8 are referred. The last occurrences of this species wasrecorded in the lowermost Rio do Rasto Formation (Neregato et al.,2008), which is regarded as Wordian/Capitanian in age (see Holz etal., 2010).

Therefore, considering the new dating of the uppermost Rio BonitoFormation presented herein (281.4±3.4 Ma) and the age of thelowermost Irati Formation (278.4±2.2 Ma), the deposition time ofthe Palermo Formation can be estimated roughly 3 Ma. Thiscorroborates the data provided by Holz et al. (2010), in which asimilar time span for this package was proposed, based on themaximum thickness of each lithostratigraphic unit and on thegeochronologic data available for this basin (Santos et al., 2006;Rocha-Campos et al., 2007).

6. Concluding remarks

The analysis of eight samples retrieved from the Rio Bonito andPalermo formations, collected in an outcrop stratigraphically above tothe Candiota coal mine, allowed the identification of rich and diversespore–pollen assemblages that, in turn, showed diagnostic elementsidentified in palynostratigraphic schemes established to the ParanáBasin. Two biozones were recorded through these levels, named as V.costabilis and L. virrkiae zones. These associations show palynologicalresemblances with other ones through Gondwana, from the ParanáBasin in Brazil, Uruguay and Argentina, as well as from Africa andAustralia. These correlations demonstrate a general phytogeograph-ical pattern recorded within Early Permian strata of the Gondwana.

The new absolute age data of 281.4±3.4 Ma obtained hereinplaces the upper portion of the Rio Bonito Formation in the middleArtinskian. These data do not conflict with the age of 278.4±2.2 Mafor the Irati Formation and others age data obtained from neighboringbasins, such as in Argentina and Africa, from deposits bearing L.virrkiae Zone correlated palynologic assemblages.

Samples analyzed contain the last appearances of C. confluens andthe first appearances of L. virkkiae, although they are represented byfew specimens. These taxa are important guide species for Permianpalynozones in several portions of the Gondwana, representing usefulmarker for correlations across Gondwana.

The absence of the H. karrooensis Subzone in the studied sectionsuggests that the uppermost deposits of the Rio Bonito Formationwere eroded at the beginning of the marine transgressive facies ofPalermo Formation, delimited by an erosive boundary surface.

The palynologic and radiometric data presented herein contributeto the age assignments for the Pennsylvanian/Permian palynostrati-

graphic scheme of the Paraná Basin. Furthermore, they can represent anew marker to understand the floral change in southernmost ParanáBasin during the Early Permian. The outcrop studied shows the firstevidence of a significant floral change, from a paleoflora dominated bypteridophytic elements (Filicophyta, Lycophyta, Sphenophyta), whichare represented by several spores species, to a gimnospermicvegetation, with dominance of bisaccate (non-taeniate and taeniate)and polyplicate pollen grains.

Our absolute dating corresponds to the oldest age for the ParanáBasin L. virkkiae Zone, which has biostratigraphic analogous unitsalong the Occidental Gondwana, mainly in the meridional portion ofSouth America. The integration of biostratigraphical data andradiometric age datings for Gondwana revels new advances to theknowledge of the palynological succession along different basins, andmust be addressed for the whole Permian interval, needing to betreated elsewhere.

Acknowledgements

This study was supported by research grants awarded by the“Conselho Nacional de Desenvolvimento Científico” (CNPq, Brazil,Projects 474153/2004-5, 480385/2010-6 and 401769/2010-0). Spe-cial thanks are given to Juan C. Cisneros for revision of the English text.This work constitutes part of the PhD Thesis of the first author at the“Programa de Pós-graduação em Geociências” of the UniversidadeFederal do Rio Grande do Sul.

References

Aboarrage, A.M., Lopes, R.C., 1986. Projeto Borda Leste da Bacia do Paraná. Porto Alegre.DNPM/CPRM. 18v. (unpublished).

Alves, R.G., Ade, M.V.B., 1996. Sequence stratigraphy and coal petrography applied tothe Candiota Coalfield, Rio Grande do Sul, Brazil: a depositional model.International Journal of Coal Geology 30, 231–248.

Archangelsky, S., Gamerro, J.C., 1979. Palinologia del Paleozoico Superior en el subsuelode la Cuenca Chacoparanaense, Republica Argentina. I. Estudio Sistematico de lospalinomorfos de tres perforaciones de la Provincia de Cordoba. Revista Española deMicropalentología 11, 417–478.

Azcuy, C.L., Beri, A., Bernardes-de-Oliveira, M.E.C., Carrizo, H.A., Di Pasquo, M., Saravia, P.Días, González, C., Iannuzzi, R., Lemos, V.B., Melo, J.H., Pagani, A., Rohn, R., Amenábar,C.R., Sabattini, N., Souza, P.A., Taboada, A., 2007. Bioestratigrafia del PaleozóicoSuperior de América del Sur: primera etapa de trabalho hacia una nueva propuesta.Revista de la Asociación Geológica Argentina, Publicação Especial 11, 9–65.

Backhouse, J., 1991. Permian palynostratigraphy of the Collie Basin, Western Australia.Review of Palaeobotany and Palynology 67, 237–314.

Balme, B.E., 1980. Palynology and the Carboniferous–Permian boundary in Australiaand other Gondwana continents. Palynology 4, 43–55.

Bangert, B., Stollhofen, H., Lorenz, V., Armstrong, R., 1999. The geochronology andsignificance of ash-fall tuffs in the glaciogenic Carboniferous–Permian DwykaGroup of Namibia and South Africa. Journal of South American Earth Sciences 29,33–49.

Beri, A., Daners, G., 1996. Palinomorfos de la perforación Cerro Largo Sur Nº4, PérmicoInferior Uruguay. Revista Chilena de Historia Natural 69, 163–170.

Beri, A., Daners, G., 1998. Estudio palinologico de la perforacion 201 (DI.NA.MI.GE),Pérmico, Uruguay. Boletín de la Real Sociedad. Española de Historia Natural(Sección Geologia) 94, 55–64.

Beri, A., Gutiérrez, P.R., Cernuschi, F., Balarino, L., 2004. Palinoestratigrafía del Pérmicode la Cuenca Paraná, Uruguay. In: Souza, P.A., Iannuzzi, R., Dutra, T., Bauermann,S.G., Meyer, K.B. (Eds.), XI Reunião de Paleobotânicos e Palinólogos, Gramado,Brasil: Boletim de Resumos, 29.

Carruthers, W., 1869. On the plant remains from the Brazilian coal beds with remarkson the genus Flemingites. Geological Magazine 6, 151–155.

Cazzulo-Klepzig, M., Guerra-Sommer, M., Formoso, N.L., Calarge, L.M., 2002. Geochem-ical and palynological evidence for the age determination of Permian coals,southern Brazil. Journal of South American Earth Sciences 25, 375–380.

Cazzulo-Klepzig, M., Menegat, R., Guerra-Sommer, M., 2005. Palynology and palaeo-botany in the reconstruction of landscape units from the Candiota Coalfield,Permian of Paraná Basin, Brazil. Revista Brasileira de Paleontologia 8, 83–98.

Césari, S.N., 2007. Palynological biozones and radiometric data at the Carboniferous–Permian boundary in western Gondwana. Gondwana Research 11, 529–536.

Césari, S.N., Gutiérrez, P.R., 2000. Palynostratigraphy of Upper Paleozoic sequences inCentral–Western Argentina. Palynology 24, 113–146.

Césari, S., Archangelsky, S., Seoane, L.V., 1995. Palinologia del Paleozoico Superior de laperforacion Las Mochas, Provincia de Santa Fe, Argentina. Ameghiniana 32, 73–106.

Césari, S.N., Meza, J.C., Melchor, R.N., 1996. Primer registro palinológico de la CuencaPérmica Oriental (Fm. Yacimiento Los Reyunos), Mendoza, Argentina. 12th Congr.Geol. Argent.and 3th Cong. Expl. Hidroc. Buenos Aires, Argentina. Actas, V, 49–63.

668 A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

Césari, S.N., Limarino, C.O., Gulbranson, E.L., 2011. An Upper Paleozoic biochronostrati-graphic scheme for the western margin of Gondwana. Earth-Science Reviews 1–2,149–160.

Corrêa da Silva, Z., Marques-Toigo, M., 1975. Carvão no Brasil: Mina de Candiota–RioGrande do Sul. In: Brodtkorb, M.K., Pexxutti, N., Pendola, H., Coira, B., Ametrano, S.,Homez, E. (Eds.), Congresso Ibero-Americano de Geologia Econômica, BuenosAires, Argentina. Tomo I, pp. 263–286.

Coutinho, J.M.V., Hachiro, J., 2005. Ditribution, mineralogy, petrography, provenanceand signicance of Permian ash-carrying deposits in the Paraná Basin. Revista doInstituto de Geociências da USP—Geologia, Série Científica 5, 29–39.

Daemon, R.F., Quadros, L.P., 1970. Bioestratigrafia no Neopaleozóico da Bacia do Paraná.In: Ladeira, E.A. (Ed.), XXIV Congresso Brasileiro de Geologia, Brasília, Brasil: Anais,pp. 359–412.

Falcon, R.M.S., 1975. Palyno-stratigraphy of the Lower Karoo sequence in the centralSebungwe District, Mid-Zambezi Basin, Rhodesia. Paleontologia Africana 18, 1–29.

Faure, F., Cole, D., 1999. Geochemical evidence for lacustrine microbial blooms in thevast PermianMain Karoo, Paraná, Falkland Islands andHuab basins of southwesternGondwana. Palaeogeography, Palaeoclimatology, Palaeoecology 152, 189–213.

Fontes, L.C., Cava, L.T., 1980. Análise ambiental preliminar do Gondwana na região deCandiota-RS. In: Corrêa da Silva, Z.C. (Ed.), XXXI Congresso Brasileiro de Geologia,Balneário Camboriú, Brasil: Anais, 2, pp. 755–765.

Formoso, M.L.L., Calarge, L.M., Garcia, A.J.V., Alves, D.B., Gomes, M.E.B., Misuzaki, A.M.,1997. Permian tonsteins from the Paraná Basin, Rio Grande do Sul, Brazil. In:Kodana, H., Mermut, A.R., Torrance, J.K. (Eds.), Proceedings of 11th InternationalClay Conference, 1997, Otawa, Canada, pp. 613–621.

Guerra-Sommer, M., Cazzulo-Klepzig, M., Formoso, M.L.L., Menegat, R., Basei, M.A.S.,2005. New radiometric data from ash fall rocks in Candiota coal-bearing strata andthe palynostratigraphic framework in southern Paraná Basin (Brazil). In: Pankhurst,R.J., Veiga, G.D. (Eds.), Abstracts. Gondwana 12, Mendoza, Argentina, p. 89.

Guerra-Sommer, M., Cazzulo-Klepzig, M., Santos, J.O.S., Hartmann, L.A., Ketzer, J.M.,Formoso, M.L.L., 2008a. Radiometric age determination of tonstein and strati-graphic constraints for the Lower Permian coal sucession in southern Paraná Basin,Brazil. International Journal of Coal Geology 74, 13–27.

Guerra-Sommer, M., Cazzulo-Klepzig, M., Formoso, M.L.L., Menegat, R., Mendonça-Filho, J.G., 2008b. U–Pb dating of tonstein layers from a coal sucession of thesouthern Paraná Basin (Brazil): a new geochonological aproach. GondwanaResearch 14, 474–482.

Guerra-Sommer, M., Cazzulo-Klepzig, M., Menegat, R., Formoso, M.L.L., Basei, M.A.S.,Barboza, E.G., Simas, M.W., 2008c. Geochronological data from the Faxinal coalsuccession, southern Paraná Basin, Brazil: a preliminary approach combiningradiometric U–Pb dating and palynostratigraphy. Journal of South American EarthSciences 25, 246–256.

Gulbranson, E.L., Montñez, I.P., Limarino, C.O., Isbell, J.L., Marenssi, S.A., Crowley, J.L.,2010. High-precision U–Pb calibration of Carboniferous glaciation and climatehistory, Paganzo Group NWArgentina. Geological Society of American Bulletin 122,1480–1498.

Hart, G.F., 1964. A review of the classication and distribution of the Permian miospore:Disaccate striatiti. 5th Congr. Strat. Geol. Carboniferous, Paris, Proceedings 3,1171–1199.

Holz, M., 1998. The Eo-Permian coal seams of the Paraná Basin in southernmost Brazil:an analysis of the depositional conditions using sequence stratigraphy concepts.International Journal of Coal Geology 36, 141–163.

Holz, M., Küchle, J., Philipp, R.P., Bischoff, A., Arima, N., 2006. Hierarchy of tectoniccontrol on stratigraphic signatures: base-level changes during the Early Permian inthe Paraná Basin, southernmost Brazil. Journal of South American Earth Sciences22, 185–204.

Holz, M., França, A.B., Souza, P.A., Iannuzzi, R., Rohn, R., 2010. A stratigraphic chart of theLate Carboniferous/Permian sucession of the eastern border of the Paraná Basin,Brazil, South America. Journal of South American Earth Sciences 29, 381–399.

Jackson, S.E., Pearson, N.J., Griffin, W.L., Belousova, E.A., 2004. The application of laserablation-inductively coupled plasma-mass spectrometry to in situ U–Pb zircongeochronology. Chemical Geology 211, 47–69.

Jansonius, J., Hills, L.V., 1976. Genera file of fossil spores and pollen. Special publication,Department of Geology, University of Calgary, AB.

Jones, M.J., Truswell, E.M., 1992. Late Carboniferous and Early Permian palynostrati-graphy of the Joe Joe Group, southern Galilee Basin, Queensland, and implicationsfor Gondwanan stratigraphy. Journal of Australian Geology and Geophysics 13,143–185.

Liais, E., 1872. Garnier Frères, Libraires Éditeurs: Climats, géologie, faune et geographiedu Brésil (Paris, 640 pp.).

Lindström, S., 1995a. Late Permian palynology of Fossilryggen, Vestfjella, DronningMaud Land, Antartica. Palynology 20, 15–48.

Lindström, 1995b. Early Permian palynostratigraphy of the northern Heimefrontfjellamountain-range, Dronning Maud Land, Antarctica. Review of Palaeobotany andPalynology 89, 359–415.

Ludwig, K.R., 2003. Isoplot 3.00: A Geochronological Toolkit for Microsoft Excel.Berkeley Geochronologixcal Center, Special Publication, 4 (70 pp.).

Marques, J.C., Chemale Jr., F.J., Lopes, R.C., Souza, P.A., Justino, D., Dussin, I.A., 2007. LA-ICP-MS U–Pb Zircon age form an ash fall rock of the uppermost Rio BonitoFormation, southern Paraná Basin. In: Milani, E.J., Pereira, E., Castro, J., Della Fávera,J.C., Almeida, L.F.B., Bergamaschi, S. (Eds.), 4º Simpósio sobre Cronoestratigrafia daBacia do Paraná. Rio de Janeiro, Brasil, Boletim de Resumos, p. 32.

Marques-Toigo, M. 1988. Palinologia, bioestratigrafia e paleoecologia do Neopaleozóicoda Bacia do Paraná nos estados do Rio Grande do Sul e Santa Catarina, Brasil.Doctoral Thesis, Programa de Pós-Graduação em Geociências, Universidade Federaldo Rio Grande do Sul.

Marques-Toigo, M., 1991. Palynobiostratigraphy of the southern Brazilian NeopaleozoicGondwana sequence. In: Ulbrich, H.E., Rocha Campos, A.C. (Eds.), 7st InternationalGondwana Symposium, São Paulo, Brasil, Proceedings, pp. 503–515.

Marques-Toigo, M., Corrêa da Silva, Z., Herter, G.G., 1975. Geology and palynology ofCandiota Coal Mine, RS, Brazil. In: Menéndez, C.A., Pascual, R., Aceñolaza, F.G.,Castellaro, H., Archangelsky, S., Cuerda, A., Bonaparte, J., Camacho, H., Amos, A.,Bertels, A., Gamerro, J. (Eds.), Primer Congresso Argentino de Paleontologia yBioestratigrafia, Tucuman, Argentina: Actas, I, pp. 401–427.

Matos, S.L.F. 1999. História deposicional e idade do intervalo portador de carvão daFormação Rio Bonito, Permiano da Bacia do Paraná no depósito de carvão de Candiota,RS. PhD. Thesis, Instituto de Geociências, Univerisdade de São Paulo, 158p.

Matos, S.L.F., Yamamoto, J.K., Hachiro, J., Coimbra, A.M., 2000. Tosntein da Formação RioBonito no depósito de carvão de Candiota, RS. Geociências 4, 679–684.

Matos, S.L.F., Yamamoto, J.K., Riccomini, C., Hachiro, J., Tassinari, C.C.G., 2001. Absolutedating of Permian ash-fall in the Rio Bonito Formation, Paraná Basin, Brazil.Gondwana Research 4, 421–426.

Melchor, R., 2000. New 40Ar/39Ar age for the base of Yacimiento Los Reyunos Formation(Permian), Mendoza, Argentina: stratigraphic and biostratigaphic implications.Ameghiniana 37, 271–282.

Meyer, K.E., Marques-Toigo, M., 2000. O Significado paleoambiental da microflora doscarvões da Malha IV, Mina de Candiota, RS, Permiano da Bacia do Paraná, Brasil.Revista Universidade de Guarulhos, Geociências V, número especial, pp. 17–20.

Milani, E.J. 1997. Evolução tectono-estratigráfica da Bacia do Paraná e seu relaciona-mento com a geodinâmica fanerozóica do Gondwana sul-oriental. Doctoral Thesis,Programa de Pós-Graduação em Geociências, Universidade Federal do Rio Grandedo Sul.

Milani, E.J., Zalán, P.V., 1999. An outline of the geology and petroleum systems of thePaleozoic interior basins of South America. Episodes 22, 199–205.

Modie, B.N., Le Herissé, A., 2009. Late Paleozoic palynomorph assemblages from theKaroo Supergroup and their potential for biostratigraphic correlation, KalahariKaroo Basin, Botswana. Bulletin of Geosciences 84, 337–358.

Mori, A.L.O., Souza, P.A., 2010. Palinologia das formações Rio Bonito e Palermo(Permiano Inferior, Bacia do Paraná) em Candiota, Rio Grande do Sul, Brasil: novosdados e implicações bioestratigráficas. Ameghiniana 47, 45–60.

Nahuys, J., Alpern, B., Ybert, J.P., 1968. Estudo palinológico e petrográfico de alguns carvõesdo sul do Brasil. Boletim Instituto Tecnológico do Rio Grande do Sul 46, 1–61.

Neregato, R. 2007. Estudo palinológico das formações Serra Alta, Teresina e Rio do Rastonos furos de sondagem SP-23-PR e SP-58-PR, centro-norte do Paraná (Permiano,Bacia do Paraná). Master Dissertation, Programa de Pós-Graduação em GeologiaRegional, Universidade Estadual Paulista. 107p.

Neregato, R., Souza, P.A., Rohn, R., 2008. Registros palinológicos inéditos nas formaçõesTeresina e Rio do Rasto (Permiano, Grupo Passa Dois, Bacia do Paraná): Implicaçõesbiocronoestratigráficas e paleoambientais. Pesquisas em Geociências 35, 9–21.

Pant, D.D., Srivastava, G.K., 1965. Some Lower Gondwana miospores from Brazil.Micropaleontology 11, 468–478.

Plant, N., 1869. The Brazilian coal fields. Geological Magazine 6, 147–150.Playford, G., Dino, R., 2002. Permian palynofloral assemblages of the Chaco-Paraná

Basin, Argentina: systematics and stratigraphic significance. Revista Española deMicropaleontología 34, 235–288.

Rocha-Campos, A.C., Basei, M.A.S., Nutman, A.P., Santos, P.R.dos., 2006. SHRIMP U–PbZircon geochronological calibration of the Late Paleozoic Supersequence, ParanáBasin, Brazil. In: Milani, E.J., Pereira, E., Castro, J., Della Fávera, J.C., Almeida, L.F.B.,Bergamaschi, S. (Eds.), 5° South American Symposium on Isotopic Geology (Puntadel Este), Uruguay, Short Papers, pp. 298–301.

Rocha-Campos, A.C., Basei, M.A.S., Nutman, A.P., Santos, P.R. dos, 2007. SHRIMP U–Pbzircons ages of the late Paleozoic sedimentary sequence, Paraná Basin, Brazil. 4°Simpósio sobre Cronoestratigrafia da Bacia do Paraná, Rio de Janeiro, Brasil:Boletim de Resumos, p. 33.

Rocha-Campos, A.C., Basei, M.A., Nutman, A.P., Kleiman, L.E., Varela, R., Llambias, E.,Canile, F.A., Rosa, O.C.R., 2011. 30 million years of Permian volcanism recorded inthe Choiyoi igneous province (W Argentina) and their source for younger ash falldeposits in the Paraná Basin: SHRIMP U–Pb zircon geochronology evidence.Gondwana Research 19, 509–523.

Russo, A., Archangelsky, S., Gamerro, J.C., 1980. Los depositos suprapaleozóicos en elsubsuelo de la llanura Chaco-Pampeana, Argentina. In: Castellaro, H.A. (Ed.), IICongresso Argentino de Paleontología y Biostratigrafía y I Congresso Latinoamer-icano de Paleontologia, Buenos Aires, Argentina: Actas, 4, pp. 157–173.

Santos, R.V., Souza, P.A., de Alvarenga, C.J.S., Dantas, E.L., Pimentel, M.M., de Oliveira,C.G., de Araújo, L.M., 2006. Shrimp U–Pb zircon dating and palynology of bentoniticlayers from the Permian Irati Formation, Paraná Basin, Brazil. Gondwana Research9, 456–463.

Simas, M.W., Formoso, M.L.L., Guerra-Sommer, M., Mendonça Filho, J.G., Klepzig, M.C.,2003. Mineralogy and palynofacies analyses of the tonstein of Faxinal Coalfield, analtered volcanic-ash layer from the Lower Permian of Paraná Basin, Brazil.International Journal of Coal Geology 46–76.

Smaniotto, L.P., Fischer, T.V., Souza, P.A., Iannuzzi, R., 2006. Palinologia do Morro doPapaléo, Mariana Pimentel (Permiano Inferior, Bacia do Paraná), Rio Grande do Sul,Brasil. Revista Brasileira de Paleontologia 9, 311–322.

Souza, P.A. 2000. Palinobioestratigrafia do Subgrupo Itararé, Carbonífero/Permiano, naporção nordeste da Bacia do Paraná (SP/PR, Brasil). PhD Thesis, Universidade de SãoPaulo, Instituto de Geociências, 199p.

Souza, P.A., 2006. Late Carboniferous palynostratigraphy of the Itararé Subgroup,northeastern Paraná Basin, Brazil. Review of Palaeobotany and Palynology 138, 9–29.

Souza, P.A., Callegari, L.M., 2004. An Early Permian palynoflora from the ItararéSubgroup, Paraná Basin, Brazil. Revista Española de Micropaleontologia 36,439–450.

669A.L.O. Mori et al. / Gondwana Research 21 (2012) 654–669

Souza, P.A., Marques-Toigo, M., 2001. Zona Vittatina: marco palinoestratigráfico doPermiano Inferior da Bacia do Paraná. Ciência-Técnica-Petróleo, Seção Exploraçãode Petróleo 20, 153–159.

Souza, P.A., Marques-Toigo, M., 2003. An overview in the palynostratigraphy of theUpper Paleozoic strata of the Brazilian Paraná Basin. Revista del Museo Argentinode Ciencias Naturales, Nueva Serie 5, 205–214.

Souza, P.A., Marques-Toigo, M., 2005. Progress on the palynostratigraphy of thePermian strata in Rio Grande do Sul State, Paraná Basin, Brazil. Anais da AcademiaBrasileira de Ciências 77, 353–365.

Souza, P.A., Milani, E.J., 2007. O pioneirismo do zoneamento palinoestratigráfico deDaemon & Quadros (1970): um marco no conhecimento estratigráfico da Bacia doParaná. Boletim de Geociências da Petrobrás 15, 118–130.

Souza, P.A., Vesely, F.F., Assine, M.L., 1999. Contribuição palinológica ao conhecimentodo Subgrupo Itararé na Serra dos Paes, sul do Estado de São Paulo. Revista doInstituto Geológico 20, 21–27.

Souza, P.A., Vergel, M.M., Beri, A., 2007. Pennsylvanian and Permian palynostratigraphyof the Paraná/Chacoparaná Basins in Brazil, Argentina and Uruguay: an integrativeanalysis. In: Dias Martínez, E., Rábano, I. (Eds.), 4th European Meeting on thePalaeontology and Stratigraphy of Latin America, 8. Cuadernos del MuseoGeominero, Madrid, pp. 361–366.

Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial lead isotope evolution by atwo-stage model. Earth and Planetary Science Letters 26, 207–221.

Stephenson, M., 1998. Preliminary correlation of palynological assemblages from Omanwith the Granulatisporites confluens Oppel Zone of the Grant Formation (LowerPermian), Caning Basin, Western Australia. Journal of South African Earth Sciences26, 521–526.

Stephenson, M., 2004. Early Permian spores from Oman and Saudi Arabia. In: Al-Husseini, M. (Ed.), Carboniferous, Permian and Early Triassic Arabian Stratigraphy:Geoarabia Special Publication, 3, pp. 185–215.

Stephenson, M., 2008. A review of the palynostratigraphy of Gondwana LateCarboniferous to Early Permiann glacigene sucessions. The Geologial Society ofAmerica. Special Paper 441, 317–330.

Stephenson, M., 2009. The age of the Carboniferous–Permian Converrucosisporitesconfluens Oppel Biozone: new data from the Ganigobis Shale Member (DwykaGroup) of Namibia. Palynology 33, 167–177.

Tiwari, R.S., Navale, G.K.B., 1967. Pollen and spore assemblage in some coals of Brazil.Pollen et Spores IX, 583–605.

Tiwari, R.S., Singh, V., 1981. Morphographic study of some dispersed trilete miospores(Sub-infraturma-Varitrileti) from the Lower Gondwana of India. The Palaeobota-nist 27, 253–296.

Vergel, 1993. Palinoestratigrafia de la secuencia neopaleozoica en la CuencaChacoparanaense, Argentina. In: Archangelsky, S. (Ed.), XII Congrès Internationalde la Stratigraphie et Géologie du Carbonifère et Permien. Buenos Aires, Argentina:Comptes Rendus, 1, pp. 201–212.

Wood, G.D., Gabriel, A.M., Lawson, J.C., 1996. Palynological Techniques—Processing andMicroscopy. In: Jansounius, J.R., McGregor, D.C. (Eds.), Palynology: principles andapplications: American Association of Stratigraphic Palynologists Foundation, 1, pp.29–50.

Ybert, J.P., 1975. Etude des miospores du bassin houiller de Candiota-Hulha Negra, RioGrande do Sul, Bresil. Pesquisas 5, 181–226.

Youden, W.J., 1951. In: John Wiley and Sons (Ed.), Statistical Methods for Chemists(New York, 126 pp.).