Comment on

‘‘Earliest Zanclean age for the Colombacci and uppermost Di Tettoformations of the « latest Messinian » northern Apennines:New palaeoenvironmental data from the Maccarone section

(Marche Province, Italy)’’ by Popescu et al. (2007)Geobios 40 (359–373)§

« La formation de Colombacci et le sommet de la formation Di Tetto

(‘‘Messinien terminal’’ des Apennins septentrionaux) sont d’age

zancleen : nouvelles donnees paleoenvironnementales

sur la coupe de Maccarone (Marche, Italie) »

par Popescu et al. (2007) Geobios 40 (359–373)

Marco Roveri a,*, Adele Bertini b, Paola Cipollari c, Domenico Cosentino c,Agata Di Stefano d, Fabio Florindo c, Rocco Gennari a, Elsa Gliozzi c,Francesco Grossi c, Silvia Iaccarino a, Stefano Lugli e, Vinicio Manzi a

a Dipartimento di Scienze della Terra, Università di Parma, Via G.P. Usberti, 157/A, 43100 Parma, Italyb Dipartimento di Scienze della Terra, Università di Firenze, Via La Pira 4, 50121 Firenze, Italy

c Dipartimento di Scienze Geologiche, Università Roma-Tre, L.go S. Leonardo Murialdo 1, 00146 Roma, Italyd Dipartimento di Scienze Geologiche, Università di Catania, Corso Italia 55, 95129 Catania, Italy

e Dipartimento di Scienze della Terra, Università di Modena e Reggio Emilia, Piazza S. Eufemia 19, 41100 Modena, Italy

Received 3 December 2007; accepted 18 January 2008

Available online 15 August 2008

Abstract

Two possible alternative interpretations of the claimed Zanclean age (Popescu et al., 2007) of two historical lithostratigraphic units of theNorthern Apennines, usually referred to as Late Messinian in age and recording the so called Lagomare final event of the Messinian salinitycrisis (MSC), are here discussed. The wrong age attribution of the Colombacci and ‘‘tetto’’ Fms. is ruled out based on data from theMaccarone and other sections showing that the Colombacci-Argille Azzurre Fm. boundary is basin wide synchronous and coincident with theMiocene-Pliocene boundary as far as it has been formally defined in the Eraclea Minoa GSSP. Alternatively, the opportunity of emending theZanclean GSSP to a stratigraphically lower horizon recording the first evidence of marine influences in the Mediterranean following the MSCpeak, seems not suitable, as (1) the marine signature of uppermost Messinian deposits is weak and still controversial and (2) no significantbio- and magnetostratigraphic events, well chronologically defined and recognizable at a global scale appear to be available to such apurpose.# 2008 Elsevier Masson SAS. All rights reserved.

http://france.elsevier.com/direct/GEOBIO/

Disponible en ligne sur www.sciencedirect.com

Geobios 41 (2008) 669–675

§ DOIs of original articles: 10.1016/j.geobios.2006.11.005, 10.1016/j.geobios.2008.05.001.

* Corresponding author.E-mail address: marco.roveri@unipr.it (M. Roveri).

0016-6995/$ – see front matter # 2008 Elsevier Masson SAS. All rights reserved.doi:10.1016/j.geobios.2008.01.002

M. Roveri et al. / Geobios 41 (2008) 669–675670

Résumé

Deux interprétations possibles sont discutées à la suite de la toute récente attribution de deux unités lithostratigraphiques historiques desApennins septentrionaux du Messinien supérieur au Zancléen, unités qui enregistrent l’événement du Lagomare à la fin de la crise de salinitémessinienne (CSM). La supposée mauvaise attribution chronologique initiale des formations Colombacci et « tetto » résulte des données deMaccarone ainsi que d’autres sections qui suggèrent que la limite entre les formations Colombacci-Argille Azzurre est synchrone à l’échelle dubassin et coïncident avec la limite Miocène-Pliocène tout comme formellement établie dans le GSSP à Eraclea Minoa. Alternativement,l’opportunité d’étendre le Zancléen défini par le GSSP à un niveau inférieur correspondant à l’enregistrement des premiers influx marins enMéditerranée après le pic de la CSM ne semble pas convenable car (1) la signature marine dans les dépôts du Messinien supérieur est faible etencore controversée, et (2) nous manquons d’événements bio- et magnétostratigraphiques chronologiquement bien établis et reconnaissables àl’échelle globale pour corroborer cette proposition.# 2008 Elsevier Masson SAS. All rights reserved.

Keywords: Messinian; Zanclean transgression; Lago-Mare; Biostratigraphy; Northern Apennines

Mots clés : Messinien ; Transgression zancléenne ; Lago-Mare ; Biostratigraphie ; Apennins septentrionaux

1. Introduction

In a recent paper Popescu et al. (2007), based on areappraisal of the Maccarone section (northern Apennines),claimed an earliest Zanclean age for two historical lithostrati-graphic unit of the Messinian post-evaporitic succession of theApennine foredeep basin, i.e. the ‘‘tetto’’ and Colombacci Fms.Similar conclusions, already anticipated in Clauzon et al.(2005), have been also reached by Sprovieri et al. (2007), basedon the revisitation of some key sections of the NorthernApennines. From this new age attribution Popescu et al. (2007)derive palaeoenvironmental implications supporting the Mes-sinian salinity crisis scenario proposed by the same authorsgroup in previous papers (Clauzon et al., 2005).

Setting apart some problems of formal lithostratigraphicnomenclature related to these units (see Vai, 1988 and Roveriet al., 1998) and with the obvious premise that even the mostconsolidated interpretations may hopefully change when newand more convincing data are provided, the Popescu et al.(2007) paper contains several statements that prompted us towrite this comment to contributing in a constructive way to thiscomplex matter.

The common reader, not necessarily interested in Messinianproblems or in the details of the Apennines Neogenestratigraphy, when looking at the paper’s title, is forced tobelieve that new data of the two formations are being presentedwhich demonstrates their Early Zanclean age, as far as thisstage has been formally defined.

This is of great interest indeed and encourages to gobeyond the title; however, at the end of the paper it is not clearwhether the authors wish to demonstrate that the current ageattribution of the Colombacci and ‘‘tetto’’ Fms. to theMessinian Stage was wrong or, on the contrary, that the formaldefinition of the Zanclean Stage itself does not work andshould be emended and lowered at a level currently placed inthe late Messinian.

These are two very different scenarios and the authorsshould clarify which one actually corresponds to their intention.

This uncertainty is reinforced by the paper’s conclusions (p.371) where the authors state that:

� th

e results of their work allow to close a long-lived debateabout the Messinian versus Pliocene age of the ColombacciFm., which is now ‘‘unequivocally’’ placed into the Pliocenetogether with the upper part of the ‘‘tetto’’ fm.; � th e Zanclean transgression has to be lowered in the regional

stratigraphy.

In our opinion, it seems that, despite the title, the authors aremore inclined towards the second hypothesis, even if muchuncertainty remains. However, in our comment we will discussboth the hypothesis starting from the first one.

2. Hypothesis 1–A wrong age for the Colombacci andtetto Fms.?

Proposing an Early Zanclean age (MPl1) for the Colombacciand upper ‘‘tetto’’ Fms. means that the base of the overlyingArgille Azzurre Fm. should be, at least in the Maccaronesection, younger than commonly thought. The Zanclean Stageis well constrained by bio- and magnetostratigraphic events thatallow to obtain, through astronomical tuning to the insolationcurve, a very high-resolution stratigraphy of the basal Pliocene.The key element is the recognition of the Thvera normal chron,which occurs five precessional cycles above the Miocene/Pliocene boundary, according to its formal definition in theEraclea Minoa GSSP (Van Couvering et al., 2000). This basalinterval, even when evidences of lithological cyclicity arelacking, can be well constrained due to the occurrence of anordered sequence of bioevents: two successive sinistral coilingshifts of Neogloboquadrina acostaensis in cycles 2 and 3, theacme of Sphaeroidinellopsis spp. between cycles 2 and 6 andthe FCO of Globorotalia margaritae at cycle 10 (Di Stefanoet al., 1996), i.e. within the Thvera magnetic event.

As a consequence, any study aiming at the recognition of theMiocene/Pliocene boundary should necessarily and carefullylook for these bio- and magnetic events; this implies thecollection and multidisciplinary analysis of a good number ofsamples.

However, in the Popescu et al. (2007) study no magnetos-tratigraphic analysis have been carried out and only three

M. Roveri et al. / Geobios 41 (2008) 669–675 671

samples have been collected at the base of the Argille AzzurreFm. For this reason we have some doubts about the realintention of the authors to recognize the Zanclean base as far asit has been formally defined and accepted by the internationalscientific community.

Instead of defining in as more as possible detail the age of thefirst open marine sediments of the Argille Azzurre Fm., theauthors suggest a Zanclean age and the marine nature of theColombacci and upper ‘‘tetto’’ Fm., based on the discovery ofsome specimen of the calcareous nannofossil Ceratolithusacutus, a marker of the basal Pliocene, associated to theoccurrence of marine dinoflagellate cysts and planktonicforaminifers.

Besides specific problems concerning the significance ofthe recovered faunal and floral assemblages, that will bebriefly discussed hereafter, it seems that the inferredPliocene age of these units is strictly related to their marinenature. Even considering the peculiar history of the formaldefinition of the Pliocene Series base, mixing palaeoenvir-onmental and chronostratigraphic arguments seems to us notcorrect and possibly misleading, in any case more pertinentto the arguments concurring to formulate the secondhypothesis.

2.1. Foraminifers

Since many researches put forward the hypothesis thatmarine fauna occasionally occurred within the Messinian post-evaporitic successions, a number of section from northern Italy,Calabria, Sicily and southern Spain were studied by our groupin the last years in order to better define the micropaleonto-logical characteristics of the post-evaporitic phase of the MSC(Gennari, 2007). Even if intervals lacking hypohaline faunawere found, no significant marine assemblage was collected;this is common to both marginal and basinal successions. Thescarce and small sized globigerinids commonly found in theNorthern Apennines sections as well as in the southern Spainbasins cannot testify the establishment of an open marinesetting. Moreover, the most abundant foraminifera assemblagesare often mixed with the classical Lago-Mare fauna (molluscs,ostracods and dinocysts), and, even if sometimes extinct taxaare lacking, paleoenvironmental considerations suggest theirreworked origin.

Concerning the M-P boundary in the Northern Apenninesand Tuscany our data (Gennari et al., 2006; Riforgiato et al.,2006) document that marine conditions are settled in thenorthern Mediterranean synchronously to the southern Med-iterranean (Sicily, Crete and ODP sites) and, as defined in theZanclean GSSP, in coincidence with the insolation peak 510from the present. The fact that some section, as it is the case ofthe Riccò section (Sprovieri et al., 2007), lacks the sinistralN. acostaensis influxes could be indicative of a particularmorphotectonic condition (intra-basinal structural high) asreconstructed by Artoni et al. (2007); it is worthnoting that,within the same basin, a few hundreds of meters to the south, allthe bioevents marking the base of the Zanclean are present(Gennari et al., 2006).

2.2. Calcareous nannoplankton

The recognition of the Miocene/Pliocene boundary in theMediterranean region by means of calcareous nannofossil hasbeen widely debated. Three important biological events takeplace very close to the Zanclean base and are very wellconstrained outside the Mediterranean: the Last Occurrenceof Triquetrorhabdulus rugosus, the First Occurrence ofCeratolithus acutus, and the LO of Discoaster quinqueramus,respectively astronomically calibrated at 5.231, 5.372,5.537 Ma by Backman and Raffi (1997). The LO ofT. rugosus and the FO of C. acutus have been recentlyrecalibrated by Lourens et al. (2004) and placed respectively at5.28 and 5.35 Ma. All these biohorizons anyway show lowapplicability in the Mediterranean region. The presence oftypical Discoaster quinqueramus is a controversial point. Manyauthors (e.g. Raffi et al., 2003 and reference therein) argue thatthis species and the co-distributed and similar D. bergrenii, arenot present in the Mediterranean area. On the other hand, thepresence of typical specimens is documented in the Medi-terranean area (Mazzei, 1985; Cipollari and Cosentino, 1995)but the scattered distribution and the low frequencies do notallow to define neither a FO nor a LO synchronous level in theMediterranean area. The absence of the species in thenannofossil assemblage cannot itself demonstrate that thesediments which do not contain this species is Pliocene in age.Both T. rugosus and C. acutus are rare in the Mediterraneanarea. Rio et al. (1984) and Di Stefano et al. (1996) did notmention the presence of C. acutus in the Zanclean of CapoRossello area and southern Italy. Castradori (1998) points outthe presence of very rare C. acutus only in one sample in thebasal Zanclean sediment of Site 969B succession (EasternMediterranean). The presence of such abundant and con-tinuously distributed specimens in the Maccarone sectionshould deserve a photographic documentation and quantitativeanalysis.

In any case, typical Pliocene species are widely documentedin late Miocene (e.g. Discoaster tamalis, Helicosphaera sellii):this fact does not mean that Messinian sediments are Pliocenein age in absence of other evidences.

2.3. Dinoflagellate cysts

Corradini and Biffi (1988) recorded and described for thefirst time the dinoflagellate cysts Impagidinium (?) sp. 1, 2, 3and Galeacysta etrusca in late Messinian deposits of theTyrrhenian side (Cava Serredi, Tuscany).

A later study at Maccarone provided additional insights onthe ecobiostratigraphic significance of such dinocysts and theiraffinities with Paratethyan taxa as well as their insertion withinthe typical taxa of the Lago-Mare biofacies (Bertini, 1992,2006; Bertini and Corradini, 1998).

Lago-Mare assemblages are strongly dominated by taxashowing Paratethyan affinities (e.g. Sütone-Szentai, 1988); thelatter show diverse abundance and distribution in dependenceof ecological parameters in a dominant brackish environmentsubmitted to repeated changes as pointed out at both Maccarone

M. Roveri et al. / Geobios 41 (2008) 669–675672

and Cava Serredi but also in other late Messinian Mediterra-nean sites (Bertini, 1992, 2006; Londeix et al., 2007; Popescuet al., 2007). Their arrival often postdated that of ostracods andapproximates the spread of Pediastrum. Lago-Mare assem-blages are really peculiar and clearly distinct from either theunderlying (post-evaporitic) and overlying ones. The former aredominated by very high percentages of Cretaceous andPaleogene reworked taxa; dinocysts show low diversity andare devoid of key marine taxa consisting principally ofSpiniferites ramosus group and Operculodinium spp. The EarlyPliocene assemblages are marked by the absence of the Lago-Mare dinocysts and the abrupt arrival of taxa indicative of theonset of open marine conditions. In the Northern Apennines,the palynological record from late Messinian deposits reveals,at present, only one Lago-Mare event, stratigraphicallysandwiched between an ash layer (130 m below) dated at5.5 Ma and the beginning of the Pliocene, as observed atMaccarone.

2.4. Ostracods

Mediterranean ostracod assemblages from the late Messi-nian Lago-Mare are clearly distinguishable either from the pre-evaporitic Messinian or from the Early Pliocene ones. Whilethe Early Messinian and the Early Pliocene assemblages aredefinitely marine (circalittoral and bathyal, respectively), theLago-Mare ostracod faunas are characteristic of brackishenvironments (variable from oligohaline to mesohaline waters)and made up by a very important contingent of Paratethyanspecies, which migrated in the Mediterranean area only duringthe Messinian postevaporitic interval (Gliozzi et al., 2007).

3. The Maccarone section

In the last two years, the Maccarone section has been sampledin detail, every 50 cm, for a total number of 460 collectedsamples and multidisciplinary analyses have been performed onall of them by our group (Cosentino et al., unpublished data).Previous studies had involved pollen and dinocysts (Bertini,1992, 2006). In the samples pertaining to the ‘‘tetto’’ andColombacci Fms. foraminifers and calcareous nannoplanktonare represented only by reworked species and no C. acutusspecimens have been recovered, while the ostracod assemblagesare typically brackish and dominated by Paratethyan species. Inparticular, the ostracod associations of the Colombacci Fm.,made up mainly by candonids, show the existence of anoligohaline environment up to the Colombacci/Argille AzzurreFms. boundary. In the ‘‘tetto’’ fm. reworked pre-Neogenedinocysts are present throughout and generally in large amount.Dinocyst assemblages contain also marine taxa having a largerstratigraphic range; within them the possibility of somereworking is not completely excluded. Only in the uppermostpart of the unit, in the transitional interval to the p-ev2, the arrivalof the first ‘‘Paratethyan’’ contingent (Impagidinium (?) sp. 1 ofCorradini and Biffi, 1988) of the Lago-Mare biofacies marks asignificant palaeoenvironmental change ca 8 m below the first‘‘colombaccio’’. The concomitant widespread presence of

Pediastrum, an algae typical of freshwater environments,together with a close dolomite-rich layer (Coradossi et al.,1991) support a significant fresh water input linked to the Lago-Mare event. Impagidinium (?) sp. 1 is followed by the progressivearrival of other ‘‘Paratethyan’’ taxa such as Impagidinium (?) sp.2, I. (?) sp. 3 (Corradini and Biffi, 1988) and the best knownGaleacysta etrusca, which dominate the p-ev2 assemblages.They show significant and distinctive ecophenotypical variationslinked to repeated salinity changes. Though never dominant,some marine cysts continued to be present together with the‘‘Paratethyan’’ taxa. Among the former, however, the mostabundant taxa (e.g. Lingulodinium machaerophorum andS. ramosus group) are not indicative of the instauration of openmarine conditions. In correspondence of the Colombacci/ArgilleAzzurre Fms. boundary the arrival and rise up to dominance ofopen marine dinocysts, notably of Impagidinium patulumtogether with the sudden disappearance of ‘‘Paratethyan’’ taxa,indicate a drastic and abrupt change in the aquatic environment.Across this boundary brackish ostracods completely disappear,while bathyal species occur just from the base of the ArgilleAzzurre Fm. The calcareous nannoplankton assemblages changeas well at the Colombacci/Argille Azzurre Fms. boundary,showing a dramatic drop of reworking and a significant presenceof Amaurolithus delicatus; moreover, the genus Discoaster ispresent with species never detected in the underlying samples(D. pentaradiatus, D. brouwerii, D. surculus). Finally, Trique-trorhabdulus rugosus has been observed only in the ColombacciFm. whereas it is always absent in the Argille Azzurre Fm. as wellas C. acutus. The foraminifers assemblages are very abundantand bear the Pliocene markers. In particular, the followingplanktonic foraminifer events have been recognized: left-coiledN. acostaensis, Sphaeroidinellopsis acme, FCO of G. margaritaeand FAD of G. puncticulata. Magnetostratigraphic analysis hasevidenced the Thvera normal chron above the base of the ArgilleAzzurre Fm. The FCO of G. margaritae has been found withinthe Thvera and all the bioevents that characterize the lowermostfive precessional cycles of the Zanclean stage as defined at theEraclea Minoa GSSP have been recognized in the intervalcomprised between the Colombacci/Argille Azzurre Fms.boundary and the base of the Thvera.

According to Popescu et al. (2007, Fig. 3) the ‘‘new’’Zanclean boundary in the Maccarone section should be moveddownward at around 110 m. Probably the proposed boundarycoincides with the sharp increase of Pinus abundance pointed outby Bertini (1992, 2006) and interpreted by Popescu et al. (2007)as an indication of more distal conditions and thus atransgression. Actually the over-representation of Pinus is awell-known taphonomic basis in coincidence of transgressivephases; the latter however are usually testified by significant andconcomitant changes in the aquatic assemblages too. AtMaccarone no substantial changes towards ‘‘more’’ marineconditions have been recorded by dinocysts in correspondence ofthe break in the pollen diagram (at 116.5 m). Tectonics and adominant ‘‘freshwater input’’ would provide evidence for analternative explanation. In fact the Pinus increase event as well asthe successive first significant change in the dinocyst assem-blages (arrival of the Lago-Mare dinocysts and Pediastrum acme

M. Roveri et al. / Geobios 41 (2008) 669–675 673

phase) fall in the transitional interval between p-ev1 and p-ev2

correlatable to a regional unconformity. The arrival of non-marine waters could have produced a scenario similar to thatfollowing a transgression with the instauration of more distalenvironments (and consequent Pinus over-representation).

However, this limit would be coincident with the regionaldiscontinuity marking the p-ev1/p-ev2 boundary, which doesnot sound like a good choice (see below). No absolute age isreported in schemes and sections for this boundary and, lookingat Figs. 3, 7 and 9, it is not clear whether the MPl1 zone at thebase of the Argille Azzurre Fm. is complete or not.

Actually, this section, as well as many other studied in thelast years, supports the high-resolution chronostratigraphicframework for the Late Messinian proposed by Roveri et al.(1998, 2001, 2004) and Roveri and Manzi (2006) based onregional-scale correlation of precessionally-driven sedimentarycyclicity particularly well developed in the Colombacci Fm.According to this stratigraphic model the p-ev1/p-ev2 boundaryhas been calibrated at 5.42 Ma and the age of the lowermost‘‘colombaccio’’ limestone bed would be around 5.39 Ma. As aconsequence, the oldest specimens of C. acutus, found byPopescu et al. (2007) in interval underlying the lowermost‘‘colombaccio’’, would have an age older than its FO in theAtlantic, dated at 5.35 Ma (Lourens et al., 2004). However, wefeel that the correct recognition of C. acutus is a real problemthat deserves to be more specifically addressed in the future bythe specialists, considering its importance also for a bettercomprehension of the relationships between the Mediterraneanbasin and the Paratethys.

Based on these considerations, we believe that thehypothesis 1, i.e. the wrong age attribution of the Colombacciand ‘tetto’ Fms., can be ruled out. However, besides being notadequately supported by the data presented by Popescu et al.(2007), we suppose that such a hypothesis was not actually themain intention of the authors.

4. Hypothesis 2–Lowering the Zanclean and PlioceneSeries GSSP?

According to Popescu et al. (2007) and Clauzon et al.(2005), two successive marine transgressions occurred duringthe MSC; the first one at around 5.52 Ma and the second onecorresponding to the Zanclean transgression, immediatelyfollowing the deep desiccation event at 5.40 Ma and leading tothe end of the MSC with the Mediterranean refill. This lattertransgression would be the one recorded in the Maccaronesection within the upper tetto-Colombacci Fms. interval.

Furthermore, the marine character of these deposits, basedon the occurrence of foraminifers, calcareous nannoplankton,dinoflagellate cysts and also supported by the recent discoveriesof marine fish otoliths (Carnevale et al., 2006, 2008), wouldhave been overlooked till now, as more importance aspaleoenvironmental proxies has been given to ostracods andmolluscs, both indicating a low salinity water column.

Setting apart the long debate about the palaeoenvironmentalmeaning of the uppermost Messinian deposits (the Lago-Mareevent) and assuming a marine nature for the upper tetto-

Colombacci Fms., it follows that, automatically, they have to beconsidered Pliocene in age and hence the Zanclean base shouldbe lowered.

This point, which is not clearly addressed in Popescu et al.(2007), is much more explicitally explained in Clauzon et al.(2005) where they state (p. 451) that ‘‘. . .the uppermost 5 m of theLago-Mare Formation are missing at the Eraclea Minoareference section. Such data lead us to reject the classicalstratigraphical position of the Arenazzolo Formation and to shiftit into the earliest Zanclean. . .’’. The same arguments andconclusions, even with more details, are put forward by Londeixet al. (2007); at pp. 238–239, while describing the stratigraphy ofthe Eraclea Minoa section they state that ‘‘. . .in some outcropsfrom Eraclea Minoa, an erosional discontinuity obviouslyseparates Arenazzolo from Lago-Mare (the Lago-Mare presentsa significantly reduced thickness with disappearance of its upperreference beds, and the Arenazzolo sediments are onlapping theresidual Lago-Mare ones. Arenazzolo silts. . ..might actuallybelong to the Pliocene series. . .’’. In the conclusions (p. 246)their point of view is unequivocally clarified: ‘‘If one accepts anearliest Zanclean age for the Arenazzolo Fm. (cf. Clauzon et al.,2005), the mean sea-surface temperature would not havechanged significantly from the base of the Messinian to theearliest Zanclean. In addition such result document the idea thatthe Zanclean marine waters invaded coastal Mediterraneanbasins prior to the entrance of foraminifer markers. . . As aconsequence, the Miocene-Pliocene GSSP, as located at EracleaMinoa between Arenazzolo and Trubi Fms., should be discussedagain’’. As in the Eraclea Minoa section the Zanclean GSSP hasbeen defined and the Arenazzolo Fm. clearly lies below it, itfollows that the authors actually wish to emend the Miocene/Pliocene boundary, lowering it to include the interval duringwhich, according to their interpretation, a gradual refill of theMediterranean by Atlantic waters took place before the 5.33 Ma,giving rise to a fully re-establishment of ocean connections.

This probably descends from a rigorous observance of thehistorical definition of the base of the Pliocene Series thatshould correspond to the first evidence of the return to openmarine conditions in the Mediterranean basin.

We observe that: (1) the paleontological assemblages ofthe Colombacci and upper ‘‘tetto’’ Fms., even taking intoconsideration the possible in situ character of the observedmarine taxa, do not support the idea of the establishment offully marine conditions before the base of the Argille AzzurreFm., as the marine signal is very weak and could be eventuallyindicative of marine influences or of the persistence of aresidual marine water body in the Mediterranean at the end ofthe MSC; (2) lowering the Zanclean GSSP means to define anew isochron boundary well constrained by events that can beunequivocally recognized at a global scale; to this respect, thedata and the arguments presented in Clauzon et al. (2005),Popescu et al. (2007) and Londeix et al. (2007) are quiteambiguous and do not allow to individuate a reliable alternativeto the current Zanclean GSSP, even admitting that a openmarine environment was established within the Mediterraneanbasin well before 5.33 Ma. In fact, the base of the ArenazzoloFm., in their opinion, would be a (subaerial) unconformity

M. Roveri et al. / Geobios 41 (2008) 669–675674

formed on marginal basins during the desiccation anddeposition of evaporites in the deep Mediterranean basins;the associated hiatus is calculated by Clauzon et al. (2005) tospan a �200 ka interval. According to international rules, theGSSP of a chronostratigraphic unit cannot be coincident withan unconformity; as a consequence, even accepting thisinterpretation, the base of the Arenazzolo Fm. is not suitable forplacing the M/P boundary. Moreover, some contradictionsappear when going back to what proposed by some of theauthors (J.P.S. and G.C.) during the phase that led to define theZanclean GSSP. In Suc et al. (1997), in a general introduction ofthe problem and referring to the location of chronostratigraphicboundaries they stated that ‘‘. . . one must not neglect its globalimportance as a reference point. . . Nevertheless, such a choiceis not devoid of a geopolitical aspect, all the Neogenestratotypes. . . being located. . . mostly in Italy’’. Later on, in thesame paper, sustaining their proposal to place the Zanclean andPliocene GSSP outside the Mediterranean and more preciselyin the Bou Regreg section (Morocco) but at a levelcorresponding to the classical boundary marking the Medi-terranean flooding at the end of the Messinian, they state that‘‘the Miocene/Pliocene boundary should be drawn at astratigraphic level representative of a rise in sea-level, i.e.,in correspondence with an interglacial peak. Peaks TG15,TG11 and TG9 cannot be selected because they are coeval withthe Salinity Crisis and such a boundary would not have a globalsignificance’’. In other words, no problems with the 5.32 Ma(subsequently recalibrated at 5.33 Ma) boundary, only someobjections to another GSSP placed in a Mediterranean (Italian)section. Maybe that the problems raised with the formalizedZanclean boundary date back to this discussion but thealternative solution seems at the moment not suitable.

In any case, if this second hypothesis corresponds to theAuthor’s intention, we think that the use of a chronostrati-graphic unit in a sense that differs from its formalized definitionis not acceptable as it causes ambiguities and confusion.

5. Concluding remarks

In Roveri and Manzi (2006), a paper submitted in 2002following an oral presentation at the Eeden Meeting of Sabadell(Spain) in 2001, several MSC open problems were taken intoconsideration and discussed, among which those concerningthe Mio/Pliocene boundary. It was suggested, starting from therecognition of the isochronous character of the M/P boundarybased on detailed studies carried out in many cores and landsections (Iaccarino et al., 1999), that as ‘‘the same stratigraphicdetail is actually lacking in the other basins. . .. it cannot beruled out that. . . detailed studies could point out a differentscenario. . .’’. In particular, at the end of the chapter, it was alsosuggested that ‘the Lago-Mare could have persisted longer insome distal areas. This implies that some Lago-Mare depositscould have an Early Pliocene age’’.

This was not a prophetic view but only a working hypothesisto be verified through a detailed multidisciplinary analysis ofsections straddling the M/P boundary in different basins and notstudied in detail yet. This was the objective of a specific task of a

research project on the Lago-Mare event, carried out since 2003by the group writing these notes. A large number of old and newsections from the Apennine basin, Sicily, Calabria, Tuscany,Crete, Spain have been studied within this project and the resultshave been presented in September 2006 at the RCMNS Meetingheld in Parma (Italy). The studied sections also comprise theMaccarone and other northern Italy sections for which Sprovieriet al. (2007), similarly to what stated by Popescu et al. (2007),reported a Pliocene age for the uppermost Colombacci Fm., aswell as the Cava Serredi section in Tuscany. Our data documentthe perfect isochron character of the M/P boundary in the studiedbasins and its correspondence to the Eraclea Minoa GSSP, as allthe bio- and magnetic events characterizing the lowermost part ofthe Zanclean have been found in the basal part of lithostrati-graphic unit historically assigned to the Pliocene (Gennari et al.,2006; Riforgiato et al., 2006).

Summarizing, we agree with Popescu et al. (2007) that thepalaeoenvironmental picture of the final stage of the MSC is farmore complicated than previously thought and that the marinesignals should not be neglected but incorporated in new,comprehensive scenarios, and this is what we obtained from ourresearches from which it results that the Lago-Mare stillmaintains a robust chronostratigraphic value (Roveri andManzi, 2006). However, we strongly believe that any effortshould be made to keep separated the chronostratigraphic andpalaeoenvironmental aspects of the problem and for this reasonwe are confident that the authors will contribute to thediscussion by clarifying their point of view.

References

Artoni, A., Rizzini, F., Roveri, M., Gennari, R., Manzi, V., Papani, G., Bernini,M., 2007. Tectonic and climatic controls on sedimentation in Late MioceneCortemaggiore Wedge-Top Basin (Northwestern Apennines, Italy). In:Lacombe, O., Lavé, J., Roure, F., Vergés, J. (Eds.), Thrust Belts andForeland Basins – From Fold Kinematics to Hydrocarbon Systems. Sprin-ger-Verlag, pp. 431–456.

Backman, J., Raffi, I., 1997. Calibration of Miocene nannofossil events toorbitally tuned cyclostratigraphies from Ceara Rise, in: Shackleton, N.J.,Curry, W.B., Richter, C., Bralower, T.J., (Eds.), Proceedings of OceanDrilling Program, Scientific Results 154, pp. 83–99.

Bertini, A., 1992. Palinologia ed aspetti ambientali del versante adriaticodell’Appennino centro-settentrionale durante il Messiniano e lo Zancleano.PhD Thesis, University of Modena, Italy.

Bertini, A., 2006. The Northern Apennines palynological record as a contributefor the reconstruction of the Messinian palaeoenvironments. SedimentaryGeology 188/189, 235–258.

Bertini, A., Corradini, D., 1998. Biostratigraphic and paleoecological signifi-cance of Galeacysta etrusca in the ‘‘Lago-mare’’ facies from the Medi-terranean area (Neogene). Dino 6. NTNU Vitenskapsmuseet RapportBotanisk Serie 1998, Abstract, 15-16.

Carnevale, G., Landini, W., Sarti, G., 2006. Mare versus Lago-mare: marinefishes and the Mediterranean environment at the end of the MessinianSalinity Crisis. Journal of the Geological Society of London 163, 75–80.

Carnevale, G., Longinelli, A., Caputo, D., Barbieri, M., Landini, W., 2008. Didthe Mediterranean marine reflooding precede the Mio-Pliocene boundary?Paleontological and geochemical evidence from upper Messinian sequencesof Tuscany, Italy. Palaeogeography, Palaeoclimatology, Palaeoecology 257,81–105.

Castradori, D., 1998. Calcareous nannofossils in the basal Zanclean of theEastern Mediterranean Sea: remarks on paleoceanography and sapropelformation. In: Robertson, A.H.F., Emeis, K.C., Richter, C., Camerlenghi, A.

M. Roveri et al. / Geobios 41 (2008) 669–675 675

(Eds.) Proceedings of the Ocean Drilling Program, Scientific Results 160,pp. 113-123.

Cipollari, P., Cosentino, D., 1995. Miocene unconformities in the centralApennines: geodynamic significance and sedimentary basin. Tectonophy-sics 252, 375–389.

Clauzon, G., Suc, J.-P., Popescu, S.M., Marunteanu, M., Rubino, J.L., Marinescu,F., Melinte, M.C., 2005. Influence of Mediterranean sea-level changes on theDacic Basin (Eastern Paratethys) during the late Neogene: the MediterraneanLago-Mare facies deciphered. Basin Research 17, 437–462.

Coradossi, N., Romagnoli, A., Vaselli, O., Bertini, A., 1991. Mineralogy andgeochemistry at the Miocene/Pliocene boundary in the Maccarone strati-graphic sequence (Marche, Central Italy). Mineralogica et PetrographicaActa 34, 13–25.

Corradini, D., Biffi, U., 1988. Étude des dinokystes à la limite Messinien-Pliocène dans la coupe Cava Serredi, Toscane, Italie. Bulletin des Centres deRecherches Exploration Production Elf-Aquitaine 12, 221–236.

Di Stefano, E., Sprovieri, R., Scarantino, S., 1996. Chronology of biostrati-graphic events at the base of the Pliocene. Paleopelagos 6, 401–414.

Gennari, R., 2007. Cause e modalità della formazione messiniana ‘‘Lago-Mare’’ e passaggio al Pliocene. Ph. D. Thesis, University of Parma.

Gennari, R., Iaccarino, S.M., Zermani, A., Manzi, V., 2006. Messinian/Zan-clean boundary in Northern Apennines. Acta Naturalia de ‘‘L’AteneoParmense’’ 42/2, p. A20.

Gliozzi, E., Ceci, M.E., Grossi, F., Ligios, S., 2007. Paratethyan ostracodimmigrants in Italy during the Late Miocene. Geobios 40, 325–337.

Iaccarino, S., Castradori, D., Cita, M.B., Di Stefano, E., Gaboardi, S., McKen-zie, J.A., Spezzaferri, S., Sprovieri, R., 1999. The Miocene/Plioceneboundary and the significance of the earliest Pliocene flooding in theMediterranean. Memorie della Società Geologica Italiana 54, 109–131.

Londeix, L., Benzakour, M., Suc, J.-P., Turon, J.L., 2007. Messinian palaeoen-vironments and hydrology in Sicily (Italy): The dinoflagellate cyst record.Geobios 40, 233–250.

Lourens, L.J., Hilgen, F.J., Laskar, J., Shackleton, N.J., Wilson, D., 2004. TheNeogene Period. In: Gradstein, F., Ogg, J., Smith, A. (Eds.), A GeologicalTime Scale. Cambridge University Press, Cambridge, pp. 409–440.

Mazzei, R., 1985. The Miocene sequence of the Maltese Islands: biostrati-graphic and chronostratigraphic reference based on nannofossils. Atti dellaSocietà Toscana di Scienze Naturali, Serie A, 92, 165–197.

Popescu, S.M., Melinte, M.C., Suc, J.P., Clauzon, S., Quillevéré, F., Süto-Szentai, M., 2007. Earliest Zanclean age for the Colombacci and uppermostDi Tetto formations of the ‘‘latest Messinian’’ northern Apennines: Newpalaeoenvironmental data from the Maccarone section (Marche Province,Italy). Geobios 40, 359–373.

Raffi, I., Mozzato, C., Fornaciari, E., Hilgen, F.J., Rio, D., 2003. Late Miocenecalcareous nannofossil biostratigraphy and astrochronology for the Medi-terranean region. Micropaleontology 49, 1–26.

Riforgiato, F., Aldinucci, M., Donia, F., Foresi, L.M., Mazzei, R., Salvatorini,G., Sandrelli, F., Sturiale, G., 2006. The Miocene/Pliocene boundary inTuscany (Italy): micropaleontological evidence from the Serredi Quarrysuccession. Acta Naturalia de ‘‘L’Ateneo Parmense’’ 42/2, p. A43.

Rio, D., Sprovieri, R., Raffi, I., 1984. Calcareous plankton biostratigraphy andbiochronology of the Pliocene-lower Pleistocene succession of the CapoRossello area, Sicily. Marine Micropaleontology 9, 135–180.

Roveri, M., Bassetti, M.A., Ricci Lucchi, F., 2001. The Mediterranean Messi-nian salinity crisis: an Apennine foredeep perspective. Sedimentary Geo-logy 140, 201–214.

Roveri, M., Landuzzi, A., Bassetti, M.A., Lugli, S., Manzi, V., Ricci Lucchi, F.,Vai, G.B., 2004. The record of Messinian events in the Northern Apenninesforedeep basins. In: Guerrieri, L., Rischia, I., Serva, L. (Eds.), ExcursionGuidebook of the 32nd International Geological Congress, Florence 20-28August 2004, Memorie descrittive della carta geologica d’Italia 63, fromB16 to B33, Apat, Roma, B19, pp. 1–44.

Roveri, M., Manzi, V., 2006. The Messinian salinity crisis: looking for a newparadygm? Palaeogeography, Palaeoclimatology. Palaeoecology 238, 386–

398.Roveri, M., Manzi, V., Bassetti, M.A., Merini, M., Ricci Lucchi, F., 1998.

Stratigraphy of the Messinian post-evaporitic stage in eastern-Romagna(northern Apennines, Italy). Giornale di Geologia 60, 119–142.

Sprovieri, R., Di Stefano, E., Bonomo, S., Tamburini, F., McKenzie, J., 2007.The Messinian-Pliocene boundary in the north Italy regions. GeophysicalResearch Abstract 9, 06041.

Suc, J.P., Clauzon, G., Gautier, F., 1997. The Miocene/ Pliocene boundary;present and future. In: Montanari, A., Odin, G.S., Coccioni, R. (Eds.),Miocene Stratigraphy–An Integrated Approach. Elsevier, Amsterdam,pp. 149–154.

Sütone-Szentai, M., 1988. A microplankton zone of organic skeleton in thePannonian S 1, stratum complex and in the upper part of the Sarmantianstrata. Acta Botanica Hungaria 34, 339–356.

Vai, G.B., 1988. A field trip guide to the Romagna Apennine geology – TheLamone valley. In: De Giuli, C., Vai, G.B. (Eds.), Fossil Vertebrates in theLamone valley, Romagna Apennines. Field Trip Guidebook, InternationalWorkshop: Continental Faunas at the Mio-Pliocene Boundary. Faenza,March 28-31, 1988, pp. 7-37.

Van Couvering, J.A., Castradori, D., Cita, M.B., Hilgen, F.J., Rio, D., 2000. Thebase of the Zanclean Stage and of the Pliocene Series. Episodes 23, 179–

186.