Imp.Palombo BIOCHRONOLOGY OF PLIO-PLEISTOCENE MAMMALIAN FAUNAS ON THE ITALIAN PENINSULA: KNOWLEDGE, PROBLEMS AND PERSPECTIVES Maria Rita Palombo Dipartimento di Scienze della Terra, Università "La Sapienza CNR- Istituto di Geologia Ambientale e Geoingegneria Piazzale A. Moro, 5 I-00185 ROMA, Italy ABSTRACT: Palombo M.R., Biochronology of Plio-Pleistocene Mammalian Faunas on the Italian Peninsula: Knowledge, Problems and Perspectives. IT ISSN 0394-335, 2004 The biochronological setting thus far proposed for mammalian faunas from the Italian peninsula (constructed on the basis of classic criteria: the absence/presence of particular taxa, faunal turnovers, the “evolutionary stage” of taxa belonging to a well-defined phyletic lineage or “typical taxa associations”) provides a highly-detailed categorization of Italian assemblages. Nevertheless, the more detai- led the bioevent sequence becomes, the more difficult it is to establish correlations on a larger scale. Multivariate analysis (the reco- gnition of successive, non-overlapping, ecologically-adjusted assemblages of taxa living together in a given space and time) may pro- vide a useful tool for defining faunal complexes having broader biochronological significance. As far as Middle Pliocene- Middle Pleistocene large mammalian faunas from the Italian peninsula are concerned, on the basis of our results, we operationally consider the following complexes as possible high-rank biochrons: Villafranchian+ early Galerian (Middle, Late Pliocene +Early Pleistocene local faunal assemblages) = “Archaic Faunas Complex ” (AFC), and middle, late Galerian+early Aurelian (Middle Pleistocene) = “Modern Faunal Complex” (MFC). Moreover, the following groups may be regarded as possible lower-rank biochrons: I = early Villafranchian; II = middle Villafranchian + late Villafranchian; III = latest Villafranchian + early Galerian; IV = middle and late Galerian; V = early Aurelian. The separation between the two main clusters highlights the faunal renewal occurring during the latest Early Pleistocene, as confirmed by mammalian species turnover and richness trends. The important reorganization of faunal complexes, characterising the transition from the late Villafranchian plus early Galerian (Early Pleistocene) to the middle-late Galerian plus early Aurelian (Middle Pleistocene) mammalian faunas, can be linked to worsening climate during the last part of the Early Pleistocene (from about 1.1 to about 0.8 Ma). This evidence encourages reconsideration of the definition of Galerian Land Mammal Age (LMA) previously proposed for Italian mam- malian faunas. The definition of biochronology (the method most widely-applied to the terrestrial realm in order to relate biological events to the geolo- gical time scale), whose application has often been tainted by reference to and inference from biostratigraphy, is briefly reconsidered. RIASSUNTO: Palombo M.R., Biocronologia dei complessi faunistici a grandi mammiferi del Plio-Pleistocene della penisola italiana: problemi, conoscenze, prospettive. IT ISSN 0394-3356, 2004. L’assetto biocronologico fin ad ora proposto per i complessi a grandi mammiferi dalla penisola italiana si base essenzialmente su bioe- venti di comparsa e scomparsa e quindi sui classici criteri di assenza/presenza di definiti taxa, rinnovi faunistici, “grado evolutivo” di taxa appartenenti a ben definite linee filetiche, riconoscimento di “associazioni tipiche”. Gli schemi attualmente in uso forniscono una scansione biocronologici dettagliata dei complessi faunistici italiani, tuttavia, quanto maggiore è la quantità di dati di nuovo acquisizio- ne e quanto più dettagliata diviene la sequenza bioeventi riconosciuti, tanto più difficile risulta operare correlazioni a grande scala. Le analisi multivariate (finalizzate alla definizione di “gruppi” di faune locali tra loro ben discriminabili, ciascuno dei quali costituente un insieme di taxa presumibilmente vissuti in uno stesso intervallo temporale in definite aree geografiche, coerente dal punto di vista eco- logico) possono fornire un mezzo utile di indagine per una prima definizione di complessi faunistici che abbiano una più ampia valenza biocronologica. I risultati ottenuti per complessi faunistici a grandi mammiferi della penisola italiana dal Pliocene medio al Pleistocene medio, con analisi di somiglianza e di ordinamento dei beventi, suggeriscono di considerare quali biocroni di più alto rango due com- plessi faunistici: il complesso “ Villafranchiano + Galeriano inferiore” (local faunal assemblages, LFAs, del Pliocene medio e superiore e del Pleistocene inferiore), qui denominato “complesso faunistico arcaico” (AFC), e il complesso “Galerian medio e superiore + Aurelian inferiore” (LFAs del Pleistocene medio), qui denominato “complesso faunistico moderno” (MFC). Inoltre, possono essere considerati quali possibili biocroni di rango inferiore i seguenti gruppi: I = Villafranchiano inferiore; II = Villafranchiano medio + Villafranchiano supe- riore; III = tardo Villafranchiano superiore + Galeriano inferiore; IV = Galeriano medio e superiore; V = Aureliano inferiore. La separa- zione fra i due gruppi principali evidenzia il profondo rinnovamento faunistico che si realizza alla fine del Pleistocene inferiore. La rior- ganizzazione dei complessi a grandi mammiferi, che caratterizza la transizione tra il complesso che riunisce le LFAs del tardo Villafranchiano e del Galeriano inferiore (Pleistocene inferiore) e quello delle LFAs del Galeriano medio, superiore e dell’Aureliano infe- riore (Pleistocene medio), può essere correlata con il deterioramento climatico del tardo Pleistocene inferiore (da circa 1,1 a circa 0,8 Ma). Su queste basi, sembrerebbe, quindi, auspicabile una ridefinizione dell’Età a mammiferi (Land Mammal Age, LMA) ” Galeriano”, come a suo tempo proposta per le faune a mammiferi della penisola italiana. Vengono anche brevemente discussi la definizione di biocronologia ed i suoi criteri di applicazione. Keywords: Biochronology, Pliocene, Pleistocene, Large Mammals, Italy. Parole chiave: Biocronologia, Pliocene, Pleistocene, Grandi Mammiferi, Italia. Il Quaternario Italian Journal of Quaternary Sciences 17(2/2), 2004, 565-582 566 M.R. Palombo 1. FOREWORD As claimed by Lindsay (2003), chronostratigraphy (the method generally applied to the oceanographic- marine realm) and biochronology (that most commonly applied to the terrestrial realm), are the prime concep- tual methods for relating biological events to the geolo- gical time scale. Nevertheless, to date, the organisation of geological time on the basis of evidence supplied by continental mammal faunas has continued to be a field plagued by interpretative and semantic confusion. The divergence of opinions on biostratigraphic/biochronological matters may often be attributed to interference between operational (based on stratigraphic records like biostratigraphy) and theo- retical (based on bioevents linked to taxa evolution and palaeocommunity transformation - biochronology) methodological approaches (see e.g. Tedford, 1970; Mein 1975, 1990, 1998; Woodburne, 1977, 1987, 1996; Berggren and Van Couvering, 1978; Guerin, 1982, 1990; De Giuli et al., 1983; Daams & Freudenthal, 1981; Flynn et al., 1984; Lindsay, 1985, 1997, 2001, 2003; Archibald et al., 1987; Emry et al., 1987; Krishtralka et al., 1987; Lindsay & Tedford, 1990; Pickford, 1990; Fernandez-Lopez, 1991; Brujin et al., 1992; Lister, 1992; Lucas, 1992; Salvador, 1994; Berggren et al., 1995a, 1995b; Prothero, 1995; Opdyke et al., 1997; Sen, 1997; Van Dam, 1997, 2001, 2003; Daams et al., 1998; Fejfar et al., 1998; Walsh, 1998; Agusti et al., 2001; Azanza et al., 2002; Koufos, 2003; Ravazzi, 2003 etc). Moreover, even if biochronology is a fundamental tool in correlating terrestrial deposits, “biochronology has never been discussed in any stratigraphic code because of that loose and ambiguous application” (Lindsay, 2003 p.216). However, it is worth noting that a biochronological unit is a time span defined by palaeo- biological events, while a biostratigraphic unit is a body of rock strata defined by its fossil content (stratigraphic datum) (Schoch, 1989; Salvador, 1994). Indeed, discon- tinuity in the continental sedimentary record, the rarity of deposits formed in a regime of virtually continuous sedimentation, the anisotropy of palaeoenvironmental conditions and taphonomic and sampling biases are responsible for the fact that the stratigraphic order of the highest and lowest occurrences (stratigraphic datum) of taxa remains within a given geographical area does not necessarily reflect the temporal order of their actual first/last appearances (palaeobiological events) in time. Therefore, as far as large continental mammals are concerned, given that: 1) the actual occurrence in time of a palaeobiological entity (both in general or in a given geographical area) cannot be unquestionably established by its fossil evidence; 2) the relative or absolute deposition dates for rock-bodies in which the first/oldest and last/youngest local appearances of fos- sil remains (lowest and highest stratigraphic records) are documented, respectively correspond only to the “ante quem” or “post quem” time of actual origination and extinction bioevents; 3) boundaries (specific surfa- ces within a sequence of rock strata selected as the standard for defining and recognising biostratigraphic units, Salvador, 1994) are not indicative “because of the impossibility of defining them with precision” (Gliozzi et al., 1997, p.369) and “biozones” ( “bodies of rock stra- ta that are defined or characterised on the basis of their contained fossils”, Salvador, 1994, pag. 53) cannot be defined, even if “biochrons” have previously been regarded as similar to “concurrent range” or “assem- blage” zones (Gliozzi et al., 1997). Ultimately, biochro- nology can be defined as “ the organization of geologic time according to the irreversible process of organic evolution” (Lindsay, 2003). “Biochrons” represent lapses of time during which faunas demonstrate clear taxonomic homogeneity (see also Alberdi et al., 1997); from an operational point of view, they might be consi- dered as “blocks of coordinated stasis” (Brett et al., 1996) during which no turnover is expected. Faunal complexes belonging to biochrons may be regarded as non- overlapping and " ecologically adjusted groups of animals with specific geographic limits and chronologic range" (Tedford, 1970, pag. 602). Nonetheless, if on the one hand first/last appea- rance bioevents (palaeobiological events on which bio- chronology is based) have been the principal basis for establishing the chronological setting of continental mammalian faunas, on the other hand these bioevents are documented by the lowest/highest occurrences of fossil remains in fossiliferous levels of various continen- tal stratigraphic successions (stratigraphic datum, the basis of biostratigraphy). Indeed, mammalian paleonto- stratigraphy (sensu Walsh, 1998) is actually founded on fossil records available in continental rocks, and new discoveries can substantially alter pre-established bio- chronological schemes. The biochronological settings thus far proposed for mammal faunas on the Italian peninsula have been based on the faunal complexes named faunal units (FUs), further classified in a number of Land Mammal Ages (LMAs). The definition of FUs derives from bioe- vents, such as the first appearance of one or more taxa, and/or the possible disappearance of others on the evolutionary stage, displayed by taxa belonging to a well-established phyletic lineage, or on typical taxa associations. This substantially agrees with Lindsay’s definition (2003) for LMAs, defined as relatively short intervals of geologic time that can be recognized and distinguished from earlier and later units (in a given region or province) by a characterising assemblage of mammals. Since initial introduction of these biochrons (Azzaroli, 1977), Italian palaeontologists have proposed some new FUs and/or new Mammal Ages (MAs) or revi- sed others (see e.g. Torre, 1987; Sala et al., 1992; Caloi & Palombo, 1996; Gliozzi et al., 1997; Petronio & Sardella, 1999; Palombo et al, 2004). The introduction of such new biochrons often depends on acquiring new data and discovering new fossiliferous assemblages that enable better discrimination of the temporal suc- cession of bioevents by means of the highest and lowe- st occurrences in local sedimentary sequences. Indeed, the complex physiography of the Italian Peninsula, the geographical and ecological barriers represented by the Alpine and Apennine chains, together with its remarka- ble latitudinal rather than longitudinal extension, as well as the great influence of the Mediterranean Sea on cli- matic conditions, supply a unique opportunity to analy- se the influence of various factors influencing the struc- ture of mammal assemblages in the course of time. It is thus evident that the current biochronological sche- me provides very detailed resolution for Italian assem- blages, but is less useful for comparing and correlating other European mammal faunas. The aim of this paper is: 1) to provide a concise picture of our knowledge of biochronology of Italian large mammals; 2) to update previous biochronological schemes in light of the latest discoveries; 3) to define adjusted homogeneous groups of taxa having biochro- nological significance using multivariate analyses; 4) to compare these “groups” with previously-established biochrons. 2. BIOCHRONOLOGICAL FRAMEWORK: AN OVERVIEW To define the chronological framework of Italian mammal faunas, a number of schemes have been pro- posed by several authors (for instance Ambrosetti et al., 1972; Azzaroli, 1977, 1982, 1983, 1991; De Giuli et al, 1983; Torre, 1987; Azzaroli et al., 1988; Torre et al., 1992, 1996; Caloi & Palombo, 1990, 1996; Gliozzi et al., 1997; Petronio & Sardella, 1999). Initially, Azzaroli (1977) proposed a subdivision of six FUs for mammal faunas belonging to the Villafranchian LMA. Subsequently, the discovery of new faunas, adding to knowledge of faunal renewal phases and the transitio- nal character of some local assemblages (LFA), led to the introduction of new faunal units and/or new propo- sed mammal ages. In 1997, Gliozzi et al. suggested a biochronological scheme for Middle Pliocene to Late Pleistocene Italian continental faunas (large and small mammals, fresh-water molluscs and brackish ostra- cods) that has recently been widely utilised by Italian palaeontologists. Three LMAs were defined for the time span between the Middle Pliocene and Late Pleistocene: Villafranchian (late Middle Pliocene to early Early Pleistocene), Galerian (late Early Pleistocene to early Middle Pleistocene), and Aurelian (late Middle Pleistocene to Late Pleistocene), respectively including eight (Triversa, Montopoli, Saint Vallier, Costa San Giacomo, Olivola, Tasso, Farneta, Pirro), four (Colle Curti, Slivia, Isernia, Fontana Ranuccio) and more than two FUs (Torre in Pietra, Vitinia and non-defined). Moreover, the transitional character of some local fau- nas was pointed out (see discussion in Gliozzi et al., 1997). Subsequently, Petronio & Sardella (1999) propo- sed Ponte Galeria local fauna as representative of a distinct middle Galerian FU, transitional to Isernia FU. 2.1 The Villafranchian Land Mammal Age (LMA) The Villafranchian was originally defined as an Age/Stage, since Pareto (1865) designated the lower boundary of the Villafranchian stratotype in Villafranca d’Asti (Italy) at the bottom of the vertebrate bearing flu- vio-lacustrine sequence overlying Pliocene marine beds. Subsequently, the “Villafranchian” question has been extensively debated (see Carraro, 1996 for a review and discussion). However, subsequent authors considered the Villafranchian as a biochron/biochrono- zone defined by dispersal events, and not as a geoch- ron/geochronozone, formally defined by golden spikes. For example, Heintz (1970) considered the Villafranchian as a biochron including the Middle and Late Pliocene. This French scientist divided his 567Biochronology of the ... Villafranchian into four biochrons (early, middle, late and latest Villafranchian) on the basis of taxa belonging to phyletic lines of cervid taxa. On the other hand, Azzaroli (1977) extended the Villafranchian to the Early- Middle Pleistocene boundary. More recently, Gliozzi et al. (1997) considered the Villafranchian MA a biochron beginning with faunal turnover marked by the appea- rance of newcomers such as “Pseudodama”(= Axis according to Di Stefano & Petronio, 2003), Leptobos stenometopon, and Stephanorhinus elatus (=S. jeanvi- reti), and ending before the Jaramillo Subchron. 2.1.1 Early and middleVillafranchian FUs Based on faunal renewals, the Villafranchian LMA was divided into early, middle and late sub-biochrons (Azzaroli et al., 1988; Gliozzi et al., 1997). The early Villafranchian is represented by the Triversa FU, established on a number of local faunas from the Villafranca d’Asti area (Piedmont) and from central Italy (e.g. Lower and Upper Valdarno, Arcille, Spoleto). Palaeomagnetic analysis at Fornace R.D.B. (Villafranca d’Asti) ascribes the sequence to Gauss Chron (Lindsay et al., 1980). Nevertheless, according to Lindsay et al. (1995), the intervals of reversed magnetic polarity separated by a normal magnetozone previously identified at Fornace R.D.B. cannot be correlated with the Mammoth or Khaena episodes, but with the late Gilbert Chron. Nevertheless, this hypothesis disagrees with geological palynological evidence (Albianelli et al., 1995,1997; Bertini & Roion, 1997) as well as with the paleomagnetic calibration (Torre et al. 1996) carried out on a long sedimentary sequence containing typical early Villafranchian fossil assemblages appearing at the Santa Barbara quarry (lignitic lacustrine silty clays and sands, Meleto clays from the Castelnuovo dei Sabbioni sequence, Upper Valdarno). The results attest a long normal polarity interval interrupted a few meters above the fossiliferous lignite level by a reversed polarity inter- val, respectively attributable to the late Gauss age and the short Kaena event. Consequently, the beginning of the early Villafranchian can be dated at about 3.3-3.2 Ma. B.P. The Triversa faunal complex (Table 1), although renewed with respect to Ruscinian faunas, maintained quite a damp forest character: pachyderms and small- middle sized Ruscinian Perrissodactyla and Artiodactyla, mainly browsers or mixed-feeders inhabi- ting dense or clear woodland, were still present, along with arboreal omnivores such as Ursus minimus. On the other hand, the disappearances of primates and carni- vores inhabiting humid and forest environments, as well as the appearance of some taxa dwelling in more or less open environments, such as Leptobos stenometo- pon, testified to moderate structural renewal, which can be correlated with the climatic cooling taking place at about 3.2/3.1 (Palombo, 2004). The following Montopoli FU is based on the rich fauna from a site near Montopoli (Lower Valdarno), found in a fossiliferous level at the top of a marine sequence, calibrated with the Gauss/Matuyama transi- tion (Lindsay et al., 1980). The Montopoli local faunal assemblage (LFA), (dated to about 2.6 Ma), as well as the Montopoli F. U., have long been considered Early Villafranchian in age. Nevertheless, the noticeable fau- nal turnover characterising the transition from the 568 M.R. Palombo Table 1 - Biochronology of selected large mammals of the Italian peninsula, dating from the late Pliocene to the Late Pleistocene. Distribuzione cronologica dei principali taxa di grandi mammiferi della penisola italiana dal Pliocene medio al Pleistocene superiore 569 Triversa F.U to the Montopoli F.U, delineated by the cooling event taking place about 2.6/2.5 Ma, may be regarded as a bioevent marking the beginning of the Middle Villafranchian (Alberdi et al., 1998, Caloi & Palombo 1999, Palombo 1995, 2004, Palombo et al., 2003). The reorganization of "palaeocommunities” from the early to middle Villafranchian (the so-called “Elephant-Equus event” sensu Lindsay et al., 1980), corresponds to a true turnover and can be regarded as the starting point of a dispersal phase. In the Montopoli FU, forest taxa greatly diminished and several new her- bivores appeared, including large-sized ones and pachyderms, such as a primitive Mammuthus (see Palombo & Ferretti, 2005 for a discussion). Moreover, the occurrence of the large Equus livenzovensis was consistent with the increasing extension of grassy areas, characterised by herbaceous vegetation of the steppe type, suggesting arid conditions (Suc et al. 1995; Torre et al. 2001; Palombo, 2004). Italian faunal assemblages belonging to the midd- le Villafranchian (“Saint Vallier” FU) are scanty, with weak stratigraphic constraints. For example, the Colle Pardo LFA (Anagni basin, Cassoli & Segre Naldini, 2000), despite the scarcity of faunal remains, has been considered intermediate in age between the Montopoli and Costa San Giacomo complexes by several authors (Caloi & Palombo, 1996; Gliozzi et al., 1997). According to the faunal list, only the occurrence of a Caprinae (possibly similar to Hemitragus orientalis recorded at the Late Pliocene Bulgarian site of Slivnitsa, Cregut- Bonnoeur & Spassov, 2003) might exclude attribution to an older faunal complex (?Montopoli FU). Consequently, using only large mammals, a faunal complex characterising the “Saint Vallier” FU cannot be adequately defined in Italy, unless running through taxa are considered. In the following Costa San Giacomo FU, a dog, similar to the late Villafranchian species Canis etruscus, and Sus strozzii were first recorded (Costa San Giacomo LFA, Anagni basin, Biddittu et al. 1979). Moreover, typical middle Villafranchian taxa, such as “Pseudodama” lyra (Axis lyra, according to Di Stefano & Petronio, 2003), Eucladoceros tegulensis, Leptobos merlai, Gazella borbonica and Gazellospira torticornis were still present, along with the Ruscinian- middle Villafranchian taxon Anancus arvernensis1. The diffusion of the genus Canis in Europe, the so called “wolf event” (sensu Azzaroli, 1983) [Canis sp. is also recorded at Chilhac (Monguillon-Douillet, 2000), a site dating more than 1.9 Ma (Bœuf, 1997)], possibly a more gradual phenomenon than previously supposed, crossed the Plio-Pleistocene boundary (sensu Aguirre & Pasini, 1985)2 and culminated at the beginning of the Pleistocene, at the time of the Tasso FU. 2.1.2 The Beginning of the late Villafranchian The faunal renewal taking place at the end of the Late Pliocene involved both carnivore and herbivore guilds, which clearly became more diversified. Most Pliocene species disappeared, particularly among her- bivores, and new carnivores and herbivores progressi- vely spread across the Italian peninsula (Table 1). Thus this renewal may be regarded as an increasing disper- sal phase rather than an actual turnover (Palombo, 2004). According to Azzaroli (1977), three FUs (Olivola, Tasso and Farneta) belong to the late Villafranchian, at that time considered to coincide with the early Pleistocene. Subsequently, a fourth FU, Pirro, more advanced than Farneta, was added (Caloi & Palombo, 1996), and the end of the late Villafranchian was later brought forward, before the Jaramillo Subcrhon (Gliozzi et al., 1997). The LFA from Olivola (Val di Magra) is characteri- sed by the lowest occurrence of Pachicrocuta breviro- stris, Panthera gombazsoegensis, ”Pseudodama” nestii (Axis nestii according to Di Stefano & Petronio, 2003) and Eucladoceros dicranios; scanty remains of Procamptoceras brivatense were also found, whereas Canis etruscus was represented by several individuals. Two late Villafranchian LFAs, Matassino and Casa Frata, were retrieved from the Upper Valdarno, from fossiliferous levels cropping out along the Montevarchi sequence (Borselli et al., 1980; De Giuli & Masini, 1987). Matassino and Casa Frata LFAs were ascribed respec- tively to Olivola and Tasso FUs (Azzaroli et al., 1988; but see Gliozzi et al, 1997), due the occurrence of new taxa such as Lycaon falconeri (see Martinez Navarro & Rook, 2003, for a discussion) at Casa Frata, Equus stehlini,? Lepobos vallisarni and Praeovibos sp. Moreover, Matassino and Casa Frata LFAs have been respectively calibrated with the β event and with the top of the Olduvai Subchron (Torre et al., 1996). Recently, a new faunal assemblage was collected at Poggio Rosso, in sediments of the second deposition phase called the Montevarchi Succession (Argille del Torrente Ascione, Albianelli et al., 1995; Mazza et al. 2004), just below the Plio-Pleistocene boundary, but about 30 m above the Matassino LFA (Napoleone et al., 2001; Napoleone & Azzaroli, 2002; Napoleone et al., 2003). The taxonomi- cal composition of the Poggio Rosso LFA is “interme- diate” between that of Olivola and Tasso FUs due to, for example, the occurrence of both Canis arnensis and Chasmaporthetes lunensis (Mazzini et al., 2000; Mazza et al., 2004). Accordingly, on the basis of magnetostra- tigraphic calibration of the Matassino and Poggio Rosso localities, the late Villafranchian theoretically started during the latest Pliocene, undoubtedly before the Plio-Pleistocene boundary. According to Napoleone Biochronology of the ... 1 It is worth noting that Zanchetta & Mazza (1996) proposed an Early Pleistocene age for a fragmentary skeleton of Anancus from Monte Castello (Lower Valdarno); in any case, this datum requires confirmation. 2 The profuse changes in flora and fauna during this event have prompted several authors (Alberdi et al 1997; Suc et al. 1997; Morrison & Kukla, 1998 and references therein) to propose placing the Plio-Pleistocene boundary here. [We have to note that, as pointed out by Kolfschoten & Gibbard (1998, p. 16) “the fact that the Vrica boundary was only notified by the Moscow IGC in 1984, should not be a reason for not reconsi- dering the boundary position”. Thus, the lowering of the Pliocene/Pleistocene boundary to around the Gauss/ Matuyama palaeomagnetic reversal (MIS 104, approximately 2.6 Ma) is being debated by several authors, as well as the INQUA Commission of Stratigraphy. In this paper, we provisio- nally maintain the Plio-Pleistocene boundary proposed by Aguirre & Pasini (1985), even if the global change occurring at MIS 104 should be considered a more appropriate boundary. 570 et al. (2003, p. 308), faunas belonging to the Olivola FU could be “inferred as pre-Olduvaian, not younger than 2.0 Ma.” Moreover, it is worth noting that the structure of faunal complexes ascribed to Olivola and Tasso FUs are not very dissimilar, the second differing essentially in the appearance of Lycaon falconeri, Leptobos valli- sarni and Equus stehlini, whereas Gallogoral meneghinii and Gazellospira torticornis were not recorded. Moreover, the Hippopotamus antiquus from Valdarno, whose appearance was previously considered as belonging on the Tasso FU, was probably retrieved from fossiliferous levels younger than those of the Montervarchi succession (Napoleone et al., 2003). Accordingly, the Olivola and Tasso faunal complexes seem to express the same progressive faunal renewal culminating with the “Tasso” faunal complex. Conse- quently, the hypothesis that the Olivola and Tasso fau- nal complexes may represent a single FU cannot be definitively ruled out. 2.2 The End of the Villafranchian LMA and the Dawn of the Galerian LMA During the following Early Pleistocene (“Postgalerian” sensu Caloi & Palombo, 1996), most of the typical Villafranchian taxa disappeared or became rare, and some new species, phylogenetically linked to taxa characterising early Middle Pleistocene faunas, progressively appeared. On the whole, during the Early Pleistocene (late Villafranchian pro parte and early Galerian), herbivores inhabiting open environments became more abundant, and forest dwellers noticeably reduced their diversity (Table 1). This is consistent with the Early Pleistocene climatic trend, characterised by a progressive decrease in mean temperature, although climatic oscillations were not accentuated. Moreover, several data suggest the spread of grassland/steppe areas (Suc et al., 1995; Bertini, 2000 and references therein). Equus stenonis was replaced by the slender- limbed Equus altidens and by the heavier, larger equid already ascribed to the problematic group of Equus bressanus/Equus suessenbornensis. Among pachy- derms, rhinoceroses, perhaps linked to Stephanorhinus hundsheimensis, seem to have been ecological substi- tutes for Stephanorhinus etruscus. Moreover, ovibovi- nes, Megalovis sp., and Bison-like bovines appeared locally (Bison (Eobison) degiuli and perhaps another taxon (?Bison sp.) may be linked to the more advanced middle Galerian Bison schoetensacki). Nevertheless, forest areas were still rather abundant and, little by lit- tle, woodlands spread during more temperate and humid climatic phases. Woodlands were mainly inhabi- ted by cervids belonging to the small-medium sized species of the “Pseudodama” genus and by the larger mixed-feeder “Praemegaceros” obscurus3, which repla- ced the browser Eucladoceros (Table 1). Furthermore, a number of data seem to confirm the transitional character of typical late Villafranchian and Galerian mammal assemblages (Azzaroli et al., 1988, Caloi & Palombo, 1996; Palombo, 2004 and refe- rences therein). According to Gliozzi et al. (1997), the appearance of “Praemegaceros” verticornis is the bioevent conven- tionally marking the beginning of the Galerian LMA (Colle Curti FU). The lowest species occurrences of this taxon in Italy is in the Colle Curti LFA, calibrated with the Jaramillo Subchron (Coltorti et al., 1998). However, Colle Curti LFA shares several taxa (carnivo- res, rhinoceros, hippopotamuses) with older LFAs belonging to the Pirro Nord FU (Table 1). Thus, despite the appearance of “Praemegaceros” verticornis as well as the innovative characteristics demonstrated by arvi- colids from Colle Curti LFAs (Coltorti et al., 1998), early Galerian LFAs seem more closely related to the latest Villafranchian faunas than to ensuing true Galerian ones (Palombo et al., 2003; Palombo, 2004; Palombo & Valli, 2004). The end of the Early Pleistocene (end of the Villafranchian event sensu Azzaroli et al., 1988) actually represented the most significant period of striking fau- nal renewal on the Italian peninsula, giving rise to remarkable palaeocommunity reorganization involving faunal migrations and extinction events (see Alberdi et al. 1997, Gliozzi et al. 1997; Palombo, 2004; Palombo et al., in press). 2.2.1 Middle and late Galerian Faunal Complexes The middle Galerian (early Middle Pleistocene) came into being during great climatic cooling (stages 24 and 22 of the marine oxygen isotopic signal, MIS) characterising the beginning of the Middle Pleistocene (sensu Cita & Castradori, 1995)4. Faunal renewal derives from two distinct trends: 1) the progressive reduction in richness characterising late Early Pleistocene faunas (from Farneta to Colle Curti FUs), 2) the subsequent dispersal and progressive diffusion in Italy of taxa from Eastern and Central Europe, leading to an important and relatively rapid increase in diversity (Slivia FU) (Palombo, 2004). M.R. Palombo 3 The name Megaceroides (type species “Cervus" algericus Lydekker 1890) was proposed by Joleaud in 1914, as a Cervus subgenus for a fragmentary maxillary; subsequently, Arambourg (1932, 1938) ascribed a skull from the Late Pleistocene of Algeria and Morocco to this species. Similarities in skull features between specimens from the Magreb and the megacerines (= Megacerini tribus, sensu Viret 1961) from Europe, first pointed out by Azzaroli (1952), have subsequently been widely debated (see e.g. Hadjoudis, 1990; Azzaroli & Mazza, 1992; Croitor, 2004; Abbazzi, in prep.). All in all, it seems more correct to maintain the name Megaceroides only for the North African species. Nevertheless, the problem con- cerning the nomenclature of the genera referring to the tribe Megacerini is still unresolved. Assuming that the species ascri- bed to the so-called “verticornis” group and the “giganteus” group belong to two distinct lineages, what is the correct generic name among those used thus far for the giant deer belonging to the “verticornis” group and the “giganteus” group belong to two distinct lineages, what is the correct generic name among those used thus far for the giant deer belonging to the “verticornis” group? The name “Praemegaceros”, thou- gh not formally correct (see e.g. Azzaroli, 1979; Caloi & Palombo, 1996), has been frequently employed, whereas other names, such as “Ortogonoceros” or “Psekupsoceros”, have not been de facto utilised in more recent times. For this rea- son, to avoid additional confusion, we provisionally use the generic name “Praemegaceros” for these giant European deer. 4 It is worth noting that the Early- Late Pleistocene boundary is currently placed informally at the Matuyama/Brunhes palaeo- magnetic inversion (Richmond 1996), but Cita & Castradori (1995) proposed a lower horizon at Marine Isotope Stage (MIS) 25. In the course of time (Table 1), the appearances of several pachyderms (Elephas (Palaeoloxodon) anti- quus, Mammuthus trogontherii, Stephanorhinus kirch- bergensis, Stephanorhinus hemitoechus), cervids (small-sized Capreolus capreolus, medium-sized Cervus elaphus and Dama clactoniana, large-sized Megaloceros savini and “Praemegaceros” solilhacus) and large and middle-sized bovines (Hemibos galeria- nus - see Martinez Navarro & Palombo 2004, for a discussion-, Bos primigenius, Ovis ammon antiqua and Hemitragus bonali), brought about a major change in herbivore guilds. Among carnivores, Crocuta crocuta was surely recorded for the first time in the Ponte Galeria 2 LFA (Sardella, 2004), whereas the lowest occurrence of some other large carnivores (active hun- ters that were both flesh-eaters and scavengers, such as Ursus deningeri, Panthera pardus, Panthera leo fos- silis, and “Hyaena prisca”) was even later, in LFAs belonging to the Isernia FU. According to Gliozzi et al. (1997), two FUs, Slivia a n d I s e r n i a , b e l o n g t o t h i s m a m m a l s u b L M A . Subsequently, Petronio & Sardella (1999) proposed a new FU, Ponte Galeria FU, based on specimens found in HST deposits of the PG2 sedimentary sequence cropping out along the Latium coast (= Ponte Galeria 2, see Milli et al., 2004). The above-mentioned authors believed the age and taxonomical composition of the Ponte Galeria FU to be intermediate between the Slivia and Isernia LFAs. If the three middle Galerian FUs were valid, they would differ essentially in the persistence of M i m o m y s s a v i n i , P a c h y c r o c u t a b r e v i r o s t r i s a n d Panthera gombaszoegensis (Slivia FU), the exclusive occurrence of Hemibos galerianus (Ponte Galeria FU) and the appearance of Bos primigenius (Isernia FU). Nevertheless, it is possible that Slivia and Ponte Galeria 2 LFAs belong to the same FU due to the uncertain identification of some herbivores from Slivia local fauna, as well as limited knowledge of the carni- vores and micromammals from the Ponte Galeria 2 LFA (Palombo, 2004). Moreover, an arvicolid similar to Mimomys savini has recently been found in Pagliare di Sassa LFA (L’Aquila, Abruzzo) (Palombo and Di Canzio, unpublished data) where “Praemegaceros” verticornis, Megaloceros savini (as in Ponte Galeria) and Dama cf D. clactoniana were also recorded (Palombo et al., 2001). As far as the transition from the middle (Isernia FU) to the late Galerian (Fontana Ranuccio FU) is con- cerned, we can place it around 500 ka, in view of the age proposed for the LFAs ascribed to the Isernia FU (Notarchirico, estimated age about 600 ka, Lefèvre & Raynal 1999, Lefèvre et al. 1999; Isernia, estimated age about 600 ka; Coltorti et al., 2000), and the possible age of the Fontana Ranuccio LFA (estimated age about 458 ka, Biddittu et al. 1979). During the late middle Galerian (Fontana Ranuccio FU), “Villafranchian” taxa had not been recor- ded and the number of very large herbivores diminished (Table 1), leaving the occurrence of “Galerian” equids and megacerines in doubt. Moreover, the specialised and slightly larger H. antiquus was replaced by Hippopotamus amphibius and, among carnivores, Ursus arctos appeared, as well as Felis silvestris silve- stris (Table 1). 2.3 Early Aurelian Faunal Complexes Gliozzi et al. (1997) proposed the Aurelian LMA (divided into early, middle and late Aurelian) for late Middle (from 10 to 6 MIS) and Late Pleistocene mam- mal faunas (from 5 to 1 MIS). The authors separated the early and middle Aurelian on the basis of faunal diffe- rences between Torre in Pietra and Vitinia FUs (Caloi & Palombo 1990, 1996), whereas no characteristic FUs have been designated for the late Aurelian. The Vitinia FU was considered dissimilar to the Torre in Pietra FU, essentially due to the abundance of a primitive fallow deer (Dama dama tiberina) and the appearance of Cervus elaphus, with features similar to those of modern forms of red deer and Equus hydrunti- nus (Caloi & Palombo, 1996; Gliozzi et al. 1997). Both these FUs have been defined on the basis of palaento- logical and stratigraphic data collected in the Campagna Romana area (cfr. Caloi et al., 1998 and references therein). Nevertheless, the structure and taxonomical composition of Torre in Pietra and Vitinia faunal complexes are similar (Palombo & Mussi, 2001; Palombo 2004), so the division into two FUs seems to be based on weak palaeontological evidence. Moreover, it is worth noting that the LFA of Sedia del Diavolo (Rome) referred to the Vitinia FU, since the abundance of Dama dama “tiberina” remains were found in alluvial deposits overlaying “Tufo lionato” (dated 355 ka by Karner et al., 2001), where Equus hydruntinus was also found (Caloi & Palombo, 1995). Caloi et al. (1998) correlated the Sedia del Diavolo deposits containing Dama dama “tiberina” remains with the Vitinia Formation (MIS 7). Nonetheless, on top of the deposits previously correlated with MIS 7, an ash flow deposit, dated 285 ± 1 ka by Karner et al. (2001), has been recognised (Marra & Rosa, 1995). In confor- mity with this new chronological data, the deposition of the fluvio-lacustrine sediments overlaying “Tufo lionato” at Sedia del Diavolo occurred approximately between 355 and 285 ka (Palombo et al., 2004). Accordingly, Dama dama ”tiberina” and Equus hydruntinus were already present during MIS 9, and thus the Vitinia FU cannot be considered a valid biochron. The LFAs pre- viously ascribed to the Torre in Pietra and Vitinia FUs must be included in a single FU (named Torre in Pietra FU for reasons of priority). Several bioevents characteri- sed the beginning of this FU, as well as the beginning of the Aurelian LMA (subdivided into early and late Aurelian): Canis lupus and Felis silvestris silvestris first appeared, as well as Ursus spelaeus, substituting U. deningeri, and E. hydruntinus, Megaloceros giganteus, Dama dama “tiberina”, and Bubalus (Palombo, unpubli- shed data), whereas Ovis ammon antiqua and Bison schoetensacki were not yet recorded (Table 1). 3. IN SEARCH OF NEW BIOCHRONOLOGICAL ASSESSMENTS In recent years, several multivariate approaches have been developed in an attempt to supplement the similarity and association concepts. Actually, multiva- riate analysis, if not conclusive in itself, can be a useful tool in initially defining faunal complexes and could be used to limit the biochronological framework based on classic methodologies (Azanza et al., 1997, 1999; 571Biochronology of the ... Palombo et al., 2003 and references therein). In order to quantify the similarities and biochronological relation- ships among selected local mammal assemblages and establish a faunal succession, we have applied two dif- ferent kinds of multivariate analysis: "appearance event ordination" (AEO; Alroy, 1994) and similarity cluster analysis (see e.g. Hazel, 1970). The former method uses the various relationships between the first and last appearances of taxa (actually the highest and lowest occurrences of their fossil remains in local faunas) as basic data. Chronological information is inferred by detecting which taxa appear before the last occurrence of another one. The "best" solution in this procedure involving correspondence analysis is to search for a parsimonious arrangement of events that minimises the overall number of implied first and last appearance relationships. "Appearance event ordination" is based on parsimony, as only relationships between first and last “appearances” in several LFAs are taken into account, while occurrences of common taxa are not considered (see Palombo et al., 2003, for a discussion). In the second method, the presence of common taxa provides basic data for evaluating similarities between local faunas grouped according to their taxo- nomic homogeneity. Conversely, the absence of a spe- cies, highly dependent on taphonomic and random fac- tors, supplies no precise information. We evaluated similarities on the basis of the Jaccard binary coeffi- cient and cluster analysis performed using the UPGMA method, as according to this method each member of a cluster has equal weight at all levels of clustering (cfr. Hazel 1970; Shi 1993; Azanza et al., 1997,1999; Palombo et al., 2003 and references therein) (NTSYS- PC program, version 2.0, Rohlf, 1998). Clusters do not necessarily supply chronological information or neces- sarily imply recognition of successive, non-overlapping, ecologically-adjusted assemblages of taxa living together in a given space and time. For this reason, the Q-Mode dendrogram, particularly suitable for biochro- nologically-oriented studies (cfr. Hazel 1970), was pre- ferred. 3.1 “Appearance Event Ordination" Calibration of the appearance event sequence (Fig. 1) introduces a temporal distortion that becomes greater as available dating diminishes, as in the case of Italian faunas. The distortion between the actual and estimated ages of localities indicates a contraction in time for Ruscinian and early+middle Villafranchian LFAs (from 4.5 to 2.5 Ma) and quite good correspondence for late Villafranchian and Galerian (2.0 to 0.400 Ma), whe- reas early Aurelian LFAs are dated more recently (between 0.400 and 0 Ma) (see Azanza et al., 2004). After obtaining chronological relationships for Italian mammal faunas, two groups can be identified, corresponding to the Pliocene + Early Pleistocene and to the Middle Pleistocene LFA (Fig. 1). Within the first group, early Villafranchian LFAs appear equivalent in age, whereas the middle Villafranchian are distributed in a span of less than 1 Ma, in accordance with the faunal renewal characteri- sing this faunal complex. Colle Pardo LFA is shown to be older than the Montopoli LFA, probably due to the absence of Mammuthus and Equus in the former loca- lity. The distribution of LFAs ascribed to Olivola and Tasso FUs confirms the homogeneity within each fau- nal complex, as well as the intermediate evolutionary stage of the Poggio Rosso LFA. Matassino cannot be clearly positioned, given the scarcity of biochronically- relevant taxa. Successive LFAs are arranged according to the progressive renewal characterising the late Early Pleistocene. Within the group including Middle Pleistocene LFAs, Slivia and Ponte Galeria LFAs appear to be the most archaic among middle Galerian faunas, whereas Fontana Ranuccio and Visogliano (late Galerian) under- go the greatest change, along with Valdemino, which actually belongs to the middle Galerian, but where some taxa (such as Panthera pardus and Bos primige- nius) were recorded for the first time on the Italian peninsula (Nocchi & Sala, 1997). The reduced biochronological significance of some late Middle Pleistocene (early Aurelian) LFAs makes it impossible to define their relationships. We can note that Paglicci (external beds) and Torrente Conca LFAs set a part, probably because of the occur- rence of Rupicapra and Capra ibex (the latter only in the Paglicci LFA). 3.2 Similarities between LFAs Among large mammal faunas from the Middle Pliocene to the late Middle Pleistocene, 74 LFAs were selected for a new similarity analysis, improving and updating those already performed on Italian mammal LFAs (Palombo et al., 2003; Palombo 2004; Palombo & Valli, in press). The faunal list for each locality was criti- cally reviewed, including up-to-date records and perso- nal observations. Since Italian FUs are in some cases based on the appearance of subspecies, these taxa were also included in the data matrix. The Q-mode dendrogram (Fig. 2) separates local faunal assemblages (LFAs) into two main groups and reveals a major separation between Pliocene-Early Pleistocene (cluster A = Villafranchian LMA and early Galerian, Colle Curti FU) and Middle Pleistocene LFAs (cluster B = middle, late Galerian - Slivia, Isernia and Fontana Ranuccio FUs-, and early Aurelian, Torre in Pietra FU), the latter group having a rather higher simi- larity coefficient. The major separation between A and B clusters underlines the faunal reorganisation which occurred during the climatic crisis taking place at the Early to Middle Pleistocene transition (sensu Cita & Castradori, 1995). This renewal phase seems to have been more important than those marking transitions from the early to the middle Villafranchian (the so-called “Elephant- Equus event” sensu Lindsay et al., 1980) and from the middle to the late Villafranchian (the so-called “wolf event” sensu Azzaroli, 1983). A and B clusters share a low number of taxa; con- sequently, we can hypothesize that each one makes up part of “an ecologically adjusted group of animals with specific geographic limits and chronological range” (Tedford 1970: 602). Alberdi et al. (1997) have informally named these kinds of aggregates “Superages”, consi- dering them equivalent to “Chronofauna” (sensu Tedford 1970, Emry et al., 1987). We informally name these groups “Archaic Faunal Complex” (AFC = cluster A) and “Modern Faunal Complex” (MFC = cluster B). 572 M.R. Palombo 573Biochronology of the ... F ig .1 - C h ro n o lo g ic a l r e la ti o n sh ip s fo r It a lia n m a m m a l f a u n a s o rd e re d a c c o rd in g t o a “ p a rs im o n y- b a se d a p p ro a c h ” (A p p e a ra n c e e ve n t o rd in a ti o n , A lr o y 1 9 9 2 , 1 9 9 4 ). O rd in am e n to b io c ro n o lo g ic o d e lle f au n e lo c al i d e lla p e n is o la it al ia n a d al P lio c e n e m e d io a l P le is to c e n e m e d io o tt e n u to u til iz za n d o il m e to d o “ A p p e ar an c e e ve n t o rd in at io n ” d i A lr o y (1 9 9 2 , 1 9 9 4 ). Both clusters are further divided into subclusters; within A, A1 includes early Villafranchian (Triversa FU) LFAs as well as the Ruscinian Val di Pugna LFA (inclu- ded to indicate the age pole in the analysis, see Bianucci et al. 2001), whereas A2 includes middle and late Villafranchian LFAs, as well as early Galerian ones. Accordingly, the separation between Triversa and Montopoli FUs seems to be the most significant within the AFC, emphasising the importance of the transition from early to middle Villafranchian (an important turno- ver phase) with respect to that from the middle to late Villafranchian (prevalently a dispersal phase) (Palombo, 2004). Within the A2 subcluster, a gap separates the A2.1 group, including middle + early late Villafranchian LFAs (from Montopoli to Tasso FUs), from the A2.2 group including the latest Villafranchian + early Galerian LFAs (from Farneta to Colle Curti FUs). Actually, early late Villafranchian complexes (Olivola and Tasso FUs) were relatively static and quite homogeneous; moreover, the presence of long-persi- stent and frequent middle Villafranchian taxa (such as Mammuthus meridionalis, Stephanorhinus etruscus and Equus stenonis) can explain their greater resemblance to middle Villafranchian LFAs than to subsequent latest Villafranchian ones. An extinction phase followed, during which last occurrences surpassed new ones. This phase culminated at the Pirro/Colle Curti FU tran- sition with a moderate turnover, due to the appearance in the course of time of some new taxa typical of the Middle Pleistocene (Table 1). A minor gap separates the group of LFAs belonging to the Farneta FU from the group of Pirro and Colle Curti FUs, thus pointing out the strong similarities between latest Villafranchian and early Galerian Italian LFAs. Actually, Pirro Nord and Colle Curti LFAs show a relatively high similarity coeffi- cient (Jaccard index = 0.3), due to the presence of several common taxa (carnivores, rhinoceros, hippopo- tamus) in both localities. This fact markedly reduces the importance of the occurrence of “Praemegaceros” ver- ticornis in the Colle Curti LFA. Within cluster B, including Middle Pleistocene LFAs, group B1 stands apart, probably because the FLAs included share few taxa with other middle and late Galerian LFAs due to low richness (Borgo Nuovo, Monte Oliveto, Pagliare di Sassa) or the occurrence of peculiar taxa, such as Mammuthus meridionalis in the Monte Tenda LEA and Hemibos galerianus in the Ponte Galeria LFA. Within B2, group B2.1 includes middle and late Galerian LFAs. The Valdemino LFA shows a low simila- rity coefficient and stands apart from the other LFAs, probably due to the occurrence of Panthera pardus, not recorded in other Italian Galerian LFAs, and the generic identification of some taxa such as Ursus sp. Group B2.1 includes LFAs attributed to the Torre in Pietra FU (e.g. La Polledrara, Castel di Guido, Malagrotta, Sedia del Diavolo, Torre in Pietra lower level, correlated with MIS 9 and Vitinia, Casal de’ Pazzi, Torre in Pietra upper level correlated with MIS 7), although it is difficult to justify the clustering of groups of minor rank, probably because of the basic homoge- neity of this faunal complex. On the basis of our results and taking into account both the biochronological significance of some rare taxa and the influence of long-lived ones, we will operationally consider clusters A and B as high rank biochrons (A = “Archaic Faunas Complex ”AFC, and B = “Modern faunal complex” MFC) and the following groups as possible biochrons of lower rank: I = early Villafranchian (A1); II = middle Villafranchian + late Villafranchian (A2.1); III = latest Villafranchian + early Galerian (A2.2); IV = middle and late Galerian (B1); V = early Aurelian (B2) (Fig. 2). In addition, the two groups of middle (A2.1.1) and early late Villafranchian (A2.1.2) should also be regarded as possible biochrons due to their low similarity coefficient. 4. REMARKS The transition from Early to Middle Pleistocene faunas (from early to middle Galerian MAs, sensu Gliozzi et al. 1997), coinciding with the onset of 100 ka climate cyclicity and vegetational changes, represents major community reorganisation in the Western Mediterranean area (see e.g. Azanza et al., 1999, in press; Palombo & Valli, in press; Palombo et al., in press and references therein). The importance of this renewal and ecological reorganisation is highlighted by the clear separation between clusters of “archaic” (Villafranchian + early Galerian, Middle Pliocene and Early Pleistocene) and “modern” (middle, late Galerian + early Aurelian, Middle Pleistocene) Italian mammal faunas, as well as by the AEO. This change results from a moderate turnover, taking place at the Pirro-Colle Curti FU transition, followed by marked faunal renewal at the beginning of the Middle Pleistocene. This evidence encourages reconsideration of the definition of Galerian Land Mammal Age (LMA) previou- sly proposed for Italian mammalian faunas. Indeed, as mentioned above, according to Gliozzi et al. (1997), the appearance of “Praemegaceros” verticornis, first recor- ded in Italy in the Colle Curti LFA, conventionally marks the beginning of the Galerian LMA. However, this fauna seems more similar to Villafranchian than to ensuing “true” Galerian LFAs, and the renewal between Colle Curti and Slivia FUs was more significant than that between Pirro and Colle Curti FUs (Table 1; Fig. 3). Moreover, and as far as large Italian mammals are con- cerned, analysis results confirm that during the late Pliocene and Pleistocene, the most important faunal renewals (due both to originations /immigrations and extinctions) are linked to major global climatic changes. Faunal renewal (“elephant-Equus event”, Lindsay et al., 1980; Azzaroli 1983; Azzaroli et al., 1988) from the early to middle Villafranchian faunal complexes, linked to Pliocene climate worsening, is responsible for the slight similarity between A1 and A2 subclusters. This renewal may be regarded as a true turnover phase due to the high percentage of last and new appearan- ces and the important ecological structural change in faunal complexes, involving both herbivore and carni- vore guilds. Moreover, this event may be considered the starting point of a dispersal phase leading to a pro- gressive standing richness increase during the following Pliocene and up to the beginning of the Early Pleistocene, given that new occurrences exceeded last occurrences (Fig. 3). On the other hand, even if the so-called “wolf- 574 M.R. Palombo 575Biochronology of the ... Fig. 2 - Dendogram Q- Mode for 74 Italian local faunal assembla- ges (Q-mode) based on un-weighted data for 136 taxa. CCC = 0.92936. Abbrevia- tions: Pugna= Val di Pugna; Ponzano= Ponzano di Magra; Gav/Barb.= Gaville Santa Barbara; P. Fosciana= Pieve Fo- sciana; C.Toppetti= Cava Toppetti; C.Gia- como= Costa San Giacomo; V.Caten= Valle Catenaccio C. Pardo= Colle Pardo; T.Picchio= Torre Pic- chio; Mt.Riccio= Mon- te Riccio; C.Sgherri= Casa Sgherri; Val- d a r n o s 1 = U p p e r Valdarno 1; Poggio Ros= Poggio Rosso; Valdarnos2= Upper Valdarno 2; B. Tibe- rino= Bacino Tiberino; F.Acetosa= Fontana Acetosa; M.Strada = Madonna della Strada; Val Chiana = Val di Chiana; Pirro = Pirro Nord; C.Curti = Colle Curti; Mt.Tenda = Monte Tenda; P. Ga- leria = Ponte Galeria; Sassa = Pagliare di Sassa; MtOliveto = Monte Oliveto; Bor- gon. = Borgonuovo; Isernia = Isernia la Pineta; Ranuccio = Fontana Ranuccio; Notarch. = Notar- chirico; Loreto = Ve- nosa-Loreto; S.Diavolo = Sedia del Diavolo; CapriQ. = Capri- Quisisana; Vitinia = Vitinia, upper levels (MIS 7); C.Mont. = Contrada Monticelli; Cverde = Campo Ve- rde; T.Pietra1 =Torre in Pietra , lower levels (MIS 9); Gr.Rom. = Grotta Romanelli, lo- wer levels; Mt.Gioie = Monte delle Gioie; CPazzi = Casal de’ Pazzi; Malagr. = Malagrotta; Polled. = La Polledrara di Cecanibbio; T.Pietra2 = Torre in Pietra, upper levels (MIS 7); GrLina = Grotta Lina; CGuido = Castel di Guido; PFiscali = Prati Fiscali; GrMBern. = Grotta Maggiore di San Bernardino; Pontec. = Pontecorvo; T.Conca = Torrente Conca; FSabina = Fara Sabina. Dendogramma Q-Mode ricavato per 74 faune locali italiane (Q-mode) sulla base dell’analisi di 136 taxa. CCC = 0.92936. Abbreviazioni: Pugna= Val di Pugna; Ponzano= Ponzano di Magra; Gav/Barb.= Gaville Santa Barbara; P. Fosciana= Pieve Fosciana; C.Toppetti= Cava Toppetti; C.Giacomo= Costa San Giacomo; V.Caten= Valle Catenaccio C.Pardo= Colle Pardo; T.Picchio= Torre Picchio; Mt.Riccio= Monte Riccio; C.Sgherri= Casa Sgherri; Valdarnos1=Upper Valdarno 1; PoggioRos= Poggio Rosso; Valdarnos2= Upper Valdarno 2; B.Tiberino= Bacino Tiberino; F.Acetosa= Fontana Acetosa; M.Strada = Madonna della Strada; Val Chiana = Val di Chiana; Pirro = Pirro Nord; C.Curti = Colle Curti; Mt.Tenda = Monte Tenda; P.Galeria = Ponte Galeria; Sassa = Pagliare di Sassa; MtOliveto = Monte Oliveto; Borgon. = Borgonuovo; Isernia = Isernia la Pineta; Ranuccio = Fontana Ranuccio; Notarch. = Notarchirico; Loreto = Venosa-Loreto; S.Diavolo = Sedia del Diavolo; CapriQ. = Capri-Quisisana; Vitinia = Vitinia, upper levels (MIS 7); C.Mont. = Contrada Monticelli; Cverde = Campo Verde; T.Pietra1 =Torre in Pietra , lower levels (MIS 9); Gr.Rom. = Grotta Romanelli, lower levels; Mt.Gioie = Monte delle Gioie; CPazzi = Casal de’ Pazzi; Malagr. = Malagrotta; Polled. = La Polledrara di Cecanibbio; T.Pietra2 = Torre in Pietra, upper levels (MIS 7); GrLina = Grotta Lina; CGuido = Castel di Guido; PFiscali = Prati Fiscali; GrMBern. = Grotta Maggiore di San Bernardino; Pontec. = Pontecorvo; T.Conca = Torrente Conca; FSabina = Fara Sabina. 576 M.R. Palombo Fig. 3 – Biochronology, standing richness and turnover changes of Italian large mammal faunas from Middle Pliocene to Late Pleistocene. Standing richness value was calculated according to Harper’s method (Harper, 1975) (Nsr = Nbda + Nrt + 1/2 (Nf + Nl– No) where Nsr = number of taxa that potentially occur at a given time interval, Nbda = number of species present before-during and after the faunal unit, Nrt = number of species present before and after but not in the faunal unit, Nf = number of first appearances, Nl = number of last appearances, No = number af taxa present only in the faunal unit). Faunal renewal was evaluated considering all occurrences at the transition between two successive FUs (turnover index, TI =% FA + % LA / 2 as in Torre et al., 1999). Biocronologia, ricchezza standarizzata e rinnovi faunistici delle faune a grandi mammiferi della penisola italiana dal Pliocene medio al Pleistocene superiore. La ricchezza standardizzata è stata calcolata secondo la formula proposta da Harper (1975): Nsr = Nbda + Nrt + 1/2 (Nf + Nl– No); Nsr = numero pesato dei taxa presenti nell’intervallo considerato, Nbda = numero di specie presenti prima, durante e dopo l’intervallo considerato, Nrt = numero dei taxa presenti prima e dopo, ma non durante l’intervallo di tempo considerato, Nf = numero delle prime comparse, Nl = numero delle ultime comparse, No = numero dei taxa presenti solo nell’intervallo considerato. L’indice di rinnovo è stato calcolato valutando il numero totale di taxa, le prime comparse e le scomparse alla transizione tra due suc- cessivi intervalli, secondo la formula TI =% FA + % LA / 2 come in Torre et al. (1999). event” (Azzaroli 1983, Azzaroli et al., 1988), seems to be a more gradual phenomenon than previously suppo- sed, middle Villafranchian LFAs and early late Villafranchian LFAs form separate groups within A2.1, clearly separated from the group including late Early Pleistocene LFAs, in which taxa phylogenetically rela- ted to or identical to Galerian Middle Pleistocene ones progressively appeared, while “Pliocene species” were no longer present. Finally, similarity analysis results confirm the basic homogeneity of late Middle Pleistocene LFAs. Consequently, without stratigraphic constraints it is dif- ficult to arrange these assemblages in a chronological sequence that is ascribing them to a specific MIS. Given these considerations, the biochronology of Plio-Pleistocene Italian mammal faunas should be re- evaluated, taking the following factors into account: - “biochrons” represent time lapses during which fau- nas have a degree of taxonomic homogeneity; the “faunal complex” belonging to each biochron might be regarded as non- overlapping and an "ecologically adjusted group of animals with specific geographic limits and chronological range" (Tedford, 1970, pag. 602); - actually, biochronology, a theoretical discipline, ope- rationally depends on the empirical documentation of the stratigraphic ranges of fossils in superimposed sections; - relative or absolute deposition dates for rock-bodies in which fossil remains are first/last documented respectively correspond only to the “ante quem” or “post quem” time of actual origination and extinction bioevents; - the more detailed the bioevent sequence becomes, the more difficult it is to make correlations on a larger scale; Moreover, the application of multivariate analysis may be a useful tool in defining faunal complexes having a high degree of similarity. Since separation between clusters corresponds to faunal renewal, clu- sters themselves should have biochronological signifi- cance and should be taken into consideration for bio- chronological correlations. On the basis of the results obtained, the faunal complex corresponding to LFAs previously ascribed to FUs from Farneta to Colle Curti should be considered as the most advanced biochron in the Villafranchian LMA; accordingly, the beginning of the Galerian LMA must correspond to the faunal reorganization occurring at the Early to Middle Pleistocene transition. Moreover, the validity and applicability of lower rank biochrons resulting from our multivariate analysis (I = early Villafranchian, Triversa FU; II = middle Villafranchian + late Villafranchian, from Montopoli to Olivola + Tasso FUs; III = latest Villafranchian + early Galerian, from Farneta to Colle Curti FU; IV = middle and late Galerian, from Slivia to Fontana Ranuccio FU; V = early Aurelian, Torre in Pietra FU) need further confirmation. ACKNOWLEDGEMENTS The author expresses her gratitude to Prof. B. Sala for his comments and suggestions. The English version has been revised by Dr. Mary Groeneweg, English Language Lecturer at Cagliari University. Work supported by MIUR 60% grants 2003, Prof. M.R. Palombo ("Evoluzione degli ecosistemi continen- tali dell’Italia peninsulare durante il Plio-Pleistocene: un approccio multidisciplinare"). REFERENCES AGUIRRE E. & PASINI G. (1985) - The Pliocene-Pleistocene Boundary. Episodes, 8 (2), pp. 116-120. AGUSTÍ J., CABRERA L., GARCÉS M., KRIJGSMAN W., OMS O. & PARÉS J.M. (2001) - A calibrated mammal scale for the Neogene of Western Europe. State of the art. Earth-Science Reviews, 52, pp. 247-260. ALBERDI M. T., AZANZA B., CERDEÑO E. & PRADO J. L. (1997) - Similarity relationship between Mammal faunas and biochronology from latest Miocene to Pleistocene in Western Mediterranean area. Eclogae Geologicae Helvetiae, 90, 115-132. ALBERDI M.T., CALOI L., PALOMBO M.R. (1998) - Large mammals associations from the Early Pleistocene: Italy and Spain. Medelingen Nederlands Inst. Toegepaste Geowetenschappen TNO, 60, pp. 521-532 ALBIANELLI, A., AZZAROLI, A., BERTINI, A., FICCARELLI, G., NAPOLEONE, G. & TORRE D. (1997) - Paleomagnetic and palynologic investigations in the upper Valdarno basin(Central Italy): calibration of an early Villafranchian fauna. Riv. Ital. Paleont. Stratigr., 103 (1), pp. 11-118 ALBIANELLI A., BERTINI A., MAGI M., NAPOLEONE G. & SAGRI M. (1995) - Il bacino Plio-Pleistocenico del Valdarno Superiore: eventi deposizionali paleoma- gnetici e paleoclimatici. Il Quaternario, 8 (1) , pp. 11-18 ALROY J. (1992) - Conjunction among taxonomic distri- butions and the Miocene mammalian biochrono- logy of the Great Plains. Paleobiology, 18(3), pp.326-343 ALROY J. (1994) - Appearance event ordination: a new biochronologic method. Paleobiology, 20 (2), pp.191-207, Lawrence. ALROY J. (1996) - Constant extinction, constrained diver- sification, and uncoordinated stasis in North American mammals. Palaeogeography, Palaeocli- matology, Palaeoecology, 127, pp. 285-311, Amsterdam. AMBROSETTI P., AZZAROLI A., BONADONNA F.P. & FOLLIERI M. (1972) - A scheme of Pleistocene chronology for the Tyrhenian side of Central Italy. Bollettino della. Società geologica italiana, 91, pp. 169-184. ARAMBOURG C. (1932) - Note preliminare sur une nouvel- le grotte à ossements des environs D’Alger. Soc. Hist. Nat. Afrique du Nord, 23 , pp. 154-162 ARAMBOURG C. (1938) - Mammiferes fossiles du Maroc Soc. Sci. nat. Maroc, Mem., 46, pp. 1- 74 ARCHIBALD J.D., GINGERICH P.D., LINDSAY E.H., CLEMENS W.A., KRAUSE D.W. & ROSE K.D. (1987) - Fist North American Land mammal Ages of the Cenozoic Era. In Woodburne, M.O. (Ed.): Cenozoic Mammal of North America: Geochronology and Biostratigraphy, pp. 24-76; University California 577Biochronology of the ... Press, California. AZANZA B., ALBERDI M. T., CERDEÑO E. & PRADO,J. L. (1997) - Biochronology from Latest Miocene to Middle Pleistocene in the Western Mediterranean Area. A Multivariate Approach. In: AGUILAR, J. L., L EGENDRE, S. & M ICHAUX, J. (Eds) Actes du Congrès BiochroM’97, Mémoires des Travaux de l’Institut de Montpellier, Montpellier, pp. 567-574. AZANZA B., ALBERDI M.T. & PRADO J.L. (1999) - Large mammal diversity and turnover patterns during the Plio-Pleistocene in Western Mediterranean Area. Revista de la Sociedad Geológica de España, 12 (1), pp. 113-122, Madrid. AZANZA B., PALOMBO M.R. & ALBERDI M.T. (2002) - Mammalian Biochronological Framework of the European Plio-Pleistocene: some remarks. 3rd French Symposium on Stratigraphy, July 8- 10/2002, Lyon, France AZANZA B., PALOMBO M.R., & ALBERDI M.T. (2004) - Large Mammals fauna turnover in Italy from Pliocene to Late Pleistocene. Rivista Italiana di Paleontologia e Stratigrafia, 110 (2), pp. 531-545 AZZAROLI A. (1952) - La sistematica dei Cervi Giganti e i Cervi Nani delle isole. Atti Soc. Tosc. Sci. Nat., Mem.(A), 59 , pp. 119-127 AZZAROLI A. (1977) - The Villafranchian Stage in Italy and the Plio-Pleistocene Boundary. Giornale di Geologia, 41, pp. 61-79. AZZAROLI A. (1979) - Critical ramarks on some Giant Deer (genus Megaceros Owen) from the Pleistocene of Europe. Palaeontografia Italica, n.s. 41, 71, pp. 5- 16 AZZAROLI A. (1982) - Remarques sur les subdivisions chronologiques du Villafranchien. Actes Colloque "Le Villafranchien Mediterranéen", Lille 9-10 décembre 1982, 1, pp. 7-14. AZZAROLI A. (1983) - Quaternary mammals and the “End-Villafranchian” dispersal event - A turning point in the history of Eurasia. Palaeogeography, Palaeoclimatology, Palaeoecology, 44, pp. 117- 139. AZZAROLI A. (1991) - Some problems of mammalian stra- tigraphy for the Pleistocene of Western Europe. I.U.Q.S. Special Proc. Rev. Rep., XIII INQUA con- gres, August 2-9. 1991, Beijing China, 130-143 AZZAROLI A., DE GIULI C., FICCARELLI G. & TORRE D. (1988) - Late Pliocene to Early Mid-Pleistocene mammals in Eurasia: Faunal succession and dispersal events. Palaeogeography, Palaeoclimatology, Palaeoecology, 66, pp. 77-100. AZZAROLI A. & MAZZA P. (1992) - The cervid genus Eucladoceros in the Early Pleistocene of Tuscany. Palaeont. Ital., 79, pp. 43-100 BERGGREN W.A., HILGEN F.J., LANGEREIS C.G., KENT D.V., OBRADOVICH J.D., RAFFI I., RAYMO M.E. & SHACKLE- TON N.J. (1995a) - Late Neogene chronology: New perspectives in high-resolution stratigraphy. BulletinGeology Society American, 107, pp. 1272- 1287. BERGGREN W.A., KENT D.V., AUBRY M.P. & HARDENBOL J. (1995b) - Geochronology, Time Scales and Global Stratigraphic Correlations. SEPM Special publica- tion, 54, pp. 1-386. B ERGGREN W.A. & J.A. V AN C OUVERING (1978) - Biochronology. In COHEE G.V., GLAESSNER M.F. & HEDBERG H.D. (Eds.): Contribution to the Geologic Time Scale. American Association of Petroleum Geologists Bulletin, 50, pp. 1487-1500. BERTINI A. (2000) - Pollen record from Colle Curti and Cesi: Early and Middle Pleistocene mammal sites in the Umbro-Marchean Apennine Mountains (central Italy). Journal of Quaternary Science, 15, 825-840. BERTINI A. & ROIRON P. (1997) - Evolution de la végéta- tion et du climat pendant le Pliocène moyen en Italie centrale: apport de la palynologie et des macroflores à l’étude du bassin du Valdarno Supérieur, (coupe de Santa Barbara). Comptes. Rendus de l’Académie des Sciences, Paris, Sér. 2a, 324, pp. 763-771 BIANUCCI G., MAZZA P., MEROLA D., SARTI G. & CASCELLA A. (2001) - The early pliocene mammal assembla- ge of Val di Pugna (Tuscany, Italy) in the light of calcareous plankton biostratigraphicaldata and paleoeological observations. Riv. Ital. Paleont. Stratigr., 107 (3), pp. 425-438. BIDDITTU I, CASSOLI P.F., RADICATI DI BROZOLO F., SEGRE A.G., SEGRE-NALDINI E., VILLA I., (1979) - Anagni a K/Ar dated Lower Middle Pleistocene site, Central Italy: preliminary report. Quaternaria, 21. 53-71 B ŒUF O. (1997) - A propos de Chilhac, Senèze, Blassac-la-Girondie (Haute-Loire, France), gise- ments du Pliocène terminal, leur intérêt biochro- nologique. In AGUILAR J.-P., LEGENDRE S., MICHAUXI J. (Eds): European Neogene Mammal Chronology, pp. 661-668, Mém. Trav. E.P.H.E., Inst. Montpellier. BORSELLI V, DE GIULI C., FICCARELLI G. & MAZZINI M. (1980) - Casa Frata: una località fossilifera del Villafranchiano Superiore presso Terranuova Bracciolini (Arezzo) nel Valdarno Superiore. Boll.Soc. Pal.It., 19, pp. 245-258, BRETT C.E, IVANY L.C. & SCHOPF K.M (1996) - Coor- dinated stasis: an overview. Palaeogeography, Palaeoclymatology, Palaeoecology, 127, 1-20 BRUIJN H. DE, DAAMS R., DAXNER-HÖCK G., FAHLBUSCH V., GINSBURG L., MEIN P. & MORALES J. (1992) - Report of the RCMNS working group on fossil mammals, Reisensburg 1990. Newsletter Stratigraphy, 26, pp. 65-118. CALOI L. & PALOMBO M.R. (1990) - Osservazioni sugli equidi italiani del Pleistocene medio inferiore. Geologica Romana, 26 (1987), pp. 187-221, CALOI L. & PALOMBO M.R. (1995) - Biocronologia e paleoecologia delle mammalofaune del Plei- stocene medio dell’Italia centrale. Studi Geologici Camerti, vol. spec. 1994 (B), pp. 503-514 C ALOI L. & P ALOMBO M.R. (1996) - Latest Early Pleistocene mammal faunas of Italy, biochronolo- gical problems. Il Quaternario, 8 (2), pp. 391-402 CALOI L. & PALOMBO M. R. (1999) - Rapporti tra carnivori ed erbivori, prede e predatori nel Pliocene medio- Pleistocene medio della penisola italiana. Geologica Romana, 34 (1998), pp. 115-129 CALOI L. & PALOMBO M.R. & ZARLENGA F. (1998) - Late Middle Pleistocene mammals faunas of Latium; stratigraphy and environment . Quaternari International, 47/48, pp. 77-86 CARRARO F. (1996) - Revisione del Villafranchiano nell’a- rea -tipo di Villafranca d’Asti. Il Quaternario, 9 (1), 578 M.R. Palombo pp. 5-120 CASSOLI P. & SEGRE NALDINI E. (2000) - Les gisements du Pléistocène myenne ed inférieur du Latium meri- dional. Coll. Inter."Les Premieres habitants de l'Europe, Tautavel 10-15 avril 2000, Resumes de communications, pp. 34-35 CITA M.B. & CASTRADORI D. (1995) - Workshop on marine sections from the Gulf of Taranto (Southern Italy) usable as potential stratotypes for the GSSP of the Lower, Middle and Upper Pleistocene. Il Quaternario, 7 (2), pp. 677-692. COLTORTI M., ALBIANELLI A., BERTINI A., FICCARELLI G., NAPOLEONE G. & TORRE D. (1998) - The Collecurti Mammal Site in the Colfiorito area (Umbrian- Marchean Apennines): stratigraphy and palynolo- gical analysis. Quaternary International, 47/48, pp. 77-86 COLTORTI M., TON-THAT T., MARZOLI A., ARZARELLO M., BUOSO N., CORRADO S., DI BUCCI D., MINELLI A., NASO G., PERETTO C., THUN HOHENSTEIN U., VILLA I. (2000) - New Chronostratigraphic and Paleo- climatic Data from the “Isernia La Pineta” site, Molise, Italy. In: SubCommission on European Quaternary Stratigraphy (SEQS), 2000 Meeting The Plio-Pleistocene Boundary and the Lower/Middle Pleistocene transition: Type Areas and Sections: Abstracts. CREGUT-BONNOURE E. & N. SPASSOV. (2002) - Hemitragus orientalis nov. sp. (Mammalia, Bovidae, Caprinae), un nouveau taxon d’Europe orientale. Rev. Paléobiol. 21(2), pp. 553-573 CROITOR R. (2004) - Systematics and phylogeny of large- sized deer oh the genus Praemegaceros Portis, 1920 (Cervidae, Mammalia). In KAHLKE R-D. & MAUL L. (Eds.) Late Neogene and Quaternary bio- diversity and evolution. 18 th International Senckenberg Conference, Weimar 25th-30th April, 2004.Terra Nostra, 2004/2, pp 89-90 DAAMS R. & FREUDENTHAL M. (1981) - Aragonian: the stage concept versus Neogene mammal zones. Scripta geologica, 1, pp. 1-17. DAAMS R., ALCALÁ L., ALVAREZ-SIERRA M.A., AZANZA B., VAN DAM J., VAN DER MEULEN J., MORALES J., NIETO M., PELÁEZ-CAMPOMANES P. & SORIA D. (1998) - A stratigraphical framework for Miocene (MN4- MN13) continental sediments of central Spain. Comptes Rendus de l'Academie de Sciences de Paris, s. IIA, 327, pp. 625-631. DE GIULI C. & MASINI F. (1987) - Late Villafranchian fau- nas in Italy: the Casa Frata Local fauna (Upper Val- darno, Tuscany). Palaeontogr.Ital., 74 (1986), 1-9. DE GIULI C., FICCARELLI G., MAZZA P. & TORRE D. (1983) - Confronto tra successioni marine e continentali del Pliocene e Pleistocene inferiore in Italia e nel- l'area mediterranea. Bollettino della Società Paleontologica italiana, 22, pp. 323-328. DI STEFANO G. & PETROLIO C. (2003) - Systematics and evolution of the Eurasian Plio-Pleistocene tribe Cervini (Artiodactyla, Mammalia. Geologica Romana, 36 (200-2002), pp. 311-334. EMRY R.J., BJORK P.R. & RUSSELL L.S. (1987) - The Chadronian, Orellan, and Whitneyan North American Land Mammal Ages. In Woodburne, M.D. (Ed.): Cenozoic Mammals of North America, pp. 118-152; University of California Press, California. F EJFAR O., H EINRICH W-D. & L INDSAY E.H. (1998) - Updating the Neogene Rodent biochronology in Europe. Mededelingen Nederlands Instituut voor Toegepaste Geowetenschappen TNO, 60, pp. 533-551. FERNÁNDEZ LÓPEZ S. (1991) - Taphonomic concepts for a Theoretical Biochronology. Revista Española de Paleontología, 6(1), pp. 37-49. FLYNN J.J., MACFADDEN B.J. & MCKENNA M.C. (1984) - Land mammal ages, faunal heterochroneity, and temporal resolution in Cenozoic terrestrial sequen- ces. Journal of Geology, 92, pp. 687-705. GLIOZZI E., ABBAZI L., AMBROSETTI P., ARGENTI P., AZZAROLI A., CALOI L., CAPASSO BARBATO L., DI STEFANO G., ESU D., FICCARELLI G., GIROTTI O., KOTSAKIS T., MASI- NI F., MAZZA P., MEZZABOTTA C., PALOMBO M.R., PETRONIO C., ROOK L., SALA B., SARDELLA R., ZANALDA E. & TORRE D. (1997) - Biochronology of selected Mammals, Molluscs, Ostracods from the Middle Pliocene to the Late Pleistocene in Italy. The state of the art. Rivista Italiana di Paleontologia i Stratigrafia, 103 (3), pp. 369-388. GUERIN C. (1982) - Première biozonation du Pléistocène européen: principal résultat biostratigraphique de l’étude des Rhinocerotidae (Mammalia, Perisso- dactyla) du Miocène terminal au Pléistocène supé- rieur d’Europe occidentale. Geobios, 15, pp. 593- 598. G UERIN C. (1990) - Biozones or Mammal Units? Methods and Limits in Biochronology. In: LINDSAY, E.H., FAHLBUSCH, V. & MEIN, P. (Eds): European Neogene Mammal Chronology, pp. 119-130; Plenum Press, New York. HARPER C.W. Jr. (1975) - Standing diversity of fossil group in successive intervals of geologic time: a new mesur. Journal of Paleobiology, 4, pp. 181- 207 HAZEL J.E. (1970) - Binary Coefficients and Clustering in Biostratigraph. Bulletin geological American Society, 81, pp. 3237-3252. HEINTZ E. (1970) - Les Cervidés villafranchiens de France et d’Espagne. Bull. Mus. natn. Hist. nat., Paris, sér. C, 22, p.303 KARNER D.B., MARRA F. & RENNE P.R. (2001) - The history of the Monti Sabatini and alban Hills volcanoes: groundwork for assessing volcanic-tectonic hazards for Rome. Jour. Volcanol. Geother. Res., 107, pp.185-219. KOLFSCHOTEN T. VAN & GIBBARD P.L. (1998) - The Dawn of the Quaternary: an introduction. Medelingen Nederlands Instituut voor Toegepaste geowen- shappen TNO, 60, pp. 13- 17 KOUFOS G.D. (2003) - Late Miocene mammal events and biostratigraphy in the Eastern Mediterranean - In: REUMER, J.W.F. & WESSELS, W. (Eds.): Distribution and migration of Tertiary mammals in Eurasia. A volume in honour of Hans de Bruijn. Deinsea, 10, pp. 343-371 KRISHTRALKA L., WEST R.M., BLACK C.C., DAWSON M.R., FLYNN J.J., TURNBULL W.D., STUCKY R.K., MCKENNA M.C., BOWN T.M., GOLZ D.J. & LILLEGRAVEN J.A. (1987) - Eocene (Wasachian through Duchesnean) biochronologyof North America. In Woodburne, M.O. (Ed.): Cenozoic Mammal of North America. 579Biochronology of the ... Geochronology and Biostratigraphy, pp. 77-117; University Califronia Press, California. JOLEAUD L. 1914 - Sur le Cervus (Megaceroides) algeri- cus Leydekker (sic !) 1890. Rec. Soc. Archéol . Costantine, 49, pp.737-739 LEFÈVRE D. & RAYNAL J.P. (1999) - Etudes géologique à Notarchirico et dans le bassin de Venosa: premier bilan et perspectives. In PIPERNO M. (Ed.): Notarchirico. Un sito del Pleistocene medio inizia- le nel bacino di Venosa, pp. 253-258; Edizioni Osanna, Italia. L EFÈVRE D., R AYNAL J.P.& V ERNET G. (1999) - Enregistrements pléistocènes dans le bassin de Venosa. In PIPERNO M. (Ed.): Notarchirico. Un sito del Pleistocen medio iniziale nel bacino di Venosa”, pp. 139-174; Edizioni Osanna, Italia. LINDSAY E.H. (1985) - European Late Cenozoic biochro- nology and the magnetic polarity time scale. National Geograohy Society Research Repport, pp. 449-456. LINDSAY E.H. (1997) - Correlation of Eurasian Late ceno- zoic mammal Chronology with the magnetic Polarity Time scale. Palaeogeography, Palaecli- matology, Palaeoecology, 133, pp. 1-288. LINDSAY E.H. (2001) - Correlation of mammal biochrono- logy with the magnetic Geomagnetic Polarity Time scale. In ROOK L. & TORRE D. (Eds.): Neogene and Quaternary continental stratigraphy and mammal evolution. Papers in honour of Prof. Augusto Azzaroli’s outstanding contribution in Geology and Paleontology. Bollettino della Società Paleontologica Italiana, 40(2) , pp. 225-233 LINDSAY E.H. (2003) - Chronostratigraphy, Biochrono- logy, Datum Events, Land Mammal Ages, Stage of Evolution, and Appearance Event Ordination. Bulletin of the American Museum of Natural History: No. 279, pp. 212-230. LINDSAY E.H., OPDYKE N.D. & JOHNSON N.M. (1980) - Pliocene dispersal of the horse Equus and Late Cenozoic mammalian dispersal events. Nature, 287, pp. 135-138. LINDSAY E.H. & TEDFORD R. (1990) - Development and application of Land Mammal Ages in North America and Europe, A comparison. In LINDSAY E.H., FAHLBUSCH V. & MEIN P. (Eds.): European Neogene Mammal Chronology, pp. 601-624; Plenum Press, New York. LISTER A. (1992) - Mammalian fossils and Quaternary biostratigraphy. Quaternary Science Reviews, 11, pp. 329-344. LUCAS S.G. (1992) - Redefinition of the Duchesnean land mammal "age", late Eocene of western North America. In Prothero R. & Berggren W.A. (eds.): Eocene-Oligicene Climatic and Biotici Evolution, pp. 88-105; Princeton University Press. LYDEKKER R. (1890) - On cervine jaw from Algeria. Proc. Zool. Soc. London, 1890, pp. 602-604 MARRA F. & ROSA C. (1995) - Stratigrafia e assetto geolo- gico dell’area romana. Mem. Descr. Carta Geol. d’Italia, 50, pp. 49-118 M ARTINEZ-N AVARRO B. & P ALOMBO M. R. (2004) - Occurrence of the Indian genus Hemibos (Bovini, Bovidae, Mammalia) at the Early-Middle Plei- stocene transition in Italy. Quaternary Research, 61, 314- 317. MARTINEZ-NAVARRO B. & ROOK L. (2003) - Gradual evolu- tion in the African hunting dog lineage. Systematic implications. Compte Rendue Paleoevolution, 2, pp. 695-702. MAZZA P., BERTINI A. & MAGI M. (2004) - The Late Pliocene Site Of Poggio Rosso (Central Italy): Taphonomy And Paleoenvironment. Palaois. 19, pp. 217-248 MAZZINI M., BORSELLI V., CIOPPI E. & NAPOLEONE G. (2000) - Poggiorosso: un importate arricchimento delle faune villafranchiane a vertebrati del Valdarno Superiore, di età prossima al limite Plio- Pleistocene. Bollettino della Società Paleontolo- gica Italiana, 39, pp. 381-388 MEIN P. (1975) - Résultats du groupe de travail des vertébrés: bizonation du Néogène méditerranéen à partir des Mammifères. Report Activity RCMNS working groups, Bratislava, 78-81. MEIN P. (1990) - Updating of MN zones. In LINDSAY E.H., FAHLBUSCH V. & MEIN, P. (Eds.): European Neogene Mammal Chronology. Plenum Press, New York, Nato ASI Ser., 180, pp. 73-90. MEIN P. (1998) - Biochronologie et phases de disper- sion chez les Vertébrés cénozoïques. Bull. Soc. géol. France, 170 (2), pp. 195-204. MILLI S., PALOMBO M. R., PETRONIO, C. & SARDELLA S. (2004) - Sedimentary sequenze and mammal bio- chronology: the case of Middle Pleistocene of the campagna Romana area. Riv. It Paleont. Strat., 110, pp. MONGUILLON-DOUILLET A. (2000) - Les carnivores du site plio-pleistocène de Chilhac (Haute-Loire, France) - Etude paléontologique, biochronologique, paléoécologique et taphonomique. Mémoire de D.E.A. “Quaternaire : Géologie, Paléontologie hu- maine, préhistoire”, Paris, 95 p M ORRISON, R. & K UKLA. G. (1998) - The Pliocene- Pleistocene (Tertiary-Quaternary) boundary should be placed at about 2.6 Ma, not at 1.8 Ma. GSA Today 8 (8), p. 9. NAPOLEONE G., AZZAROLI A., ALBIANELLI A. & MAZZINI M. (2001) -The Poggio Rosso locality calibrated to the end-Pliocene and its significance for dating the late Villafranchian mammal faunas of the upper Valdarno, central Italy. Rivista Italiana di Paleontologia e Stratigrafia, 107 (2), pp.287-296 NAPOLEONE G. & AZZAROLI A. (2002) - A collection of fos- sil vertebrates from the Upper Valdarno (Central Italy) calibrated in the end Pliocene to span 220,000 years across the Olduvai magnetochron. Rivista Italiana di Paleontologia e Stratigrafia, 108, pp. 479-492. N APOLEONE G., A ZZAROLI A. & M AZZINI M. (2003) - Magnetochronologic significance of the end- Pliocene age of the new found locality of Poggio Rosso in the Upper Valdarno, for dating its late Villafranchian faunas . Rivista Italiana di Paleontologia e Stratigrafia, 108 (3), pp. 479-492 NOCCHI G. & SALA B. (1997) - The fossil rabbit from Valdemino cave (Borgio Verezzi, Savona) in the context of western Europe Oryctolagini of Quaternary. Palaeovertebrata, 26(1-), pp. 167-187 OPDYKE N., MEIN P., LINDSAY E., PÉREZ-GONZÁLEZ A., MOISSENET E. & NORTON V.L. (1997) - Continental deposits, magnetostratigraphy and vertebrate 580 M.R. Palombo paleontology, Late Neogene of Eastern Spain. Palaeogeography, Palaeclimatology, Palaeoe- cology, 133, pp. 129-148. PALOMBO M.R. (1995) - Le principali associazioni a mam- miferi del Pliocene dell’Italia centrale. Studi Geologici Camerti, vol. spec. 1994 (B) , pp. 429- 446 PALOMBO M.R. (2004) - Guild of large Mammals from the Pliocene to the Late Pleistocene in the Italian peninsula. In Baquedano, E. & Rubio, S. (eds). Homenaje a Emiliano Aguirre. Zona Paleonto- logica, Museo Arqueológico Regional, Madrid. PALOMBO M.R. & FERRETTI M. (2005 )- Elephant fossil record from Italy: knowledge, problems, and per- spectives. Quaternary International, 126-128, pp. 107-136 PALOMBO M.R. & MUSSI, M (2001) - Human presence and faunal evolution in the Middle Pleistocene of Italy. In ROOK,. L., TORRE, D. (Eds.): Neogene and Quaternary continental stratigraphy and mammal evolution. Papers in honuor of Prof. Augusto Azzaroli’s outstanding contribution in Geology and Paleontology. Bollettino della Società Paleontologica Italiana, 40 (2), pp. 257-267 PALOMBO M.R. & VALLI A. (2004) - Similarities between Large Mammal Faunas of the Italian Peninsula and France from the Pliocene to the Middle Pleistocene. Neues Jarbuch fur Geologie und Paleontologie, Abh, 233(1), 69-102. PALOMBO M.R., AZANZA B. & ALBERDI M.T. (2003) - Italian mammal biochronology from Latest Miocene to Middle Pleistocene: a multivariate approach. Geologica Romana, 36 (2000-2002), pp. 335-368. PALOMBO M.R, MILLI S. & ROSA C. (2004) - Remarks on the late Middle Pleistocene biochronology of the mammalian faunal complexes of the Campagna Romana. Geologica Romana 37 (2003) PALOMBO M.R., RAIA P. & GIOVINAZZO C. (in press) - Early- Middle Pleistocene structural changes in mamma- lian communities from the Italian peninsula. In Head M.J. & Gibbard P.L. (Eds.) Early-Middle Pleistocene transition: the land-ocean evidence, Geological Society, London. PALOMBO M.R., ABBAZZI L., AGOSTINI, MAZZA P. & MUSSI M. (2001) - Middle Pleistocene faunas and lithic implements from Pagliara di Sassa (L'Aquila, cen- tral Italy). In CAVARETTA G., GIOIA P., MUSSI M. & PALOMBO M.R. (Eds.) The World of Elephants, pp. 224-229, pubbl. CNR, Roma PARETO M. (1865) - Note sur la subdivision que l’on por- rait établir dans les terrains de l’Apennin septen- trional. Bulletin de la Societé Géologique de France, série 2, 22, pp. 210-277. PETRONIO C. & SARDELLA R. (1999) - “Biochronology of he Pleistocene mammal fauna from Ponte Galeria (Rome) and remarks on the Middle Galerian fau- nas”. Rivista Italiana di Paleontologia e Stratigrafia, 105 (1), pp. 155-164 PICKFORD M. (1990) - Dynamics of Old World biogeo- graphic realms during the Neogene: implications for biostratigraphy. In Lindsay E.H., Fahlbusch V. & Mein P. (eds.): European Neogene Mammal Chronology, pp. 413-442; Plenum Press, New York. PROTHERO D.R. (1995) - Geochronology and magneto- stratigraphy of Paleogene North American Land Mammal "Ages": an update. In Berggren W.A., Kent D.V. & Hardenbol J. (eds.): Geochronology, Time Scales and Global Stratigraphic Correlations, pp. 305-315; SEPM, Special publi- cation,p.54; Tulsa, Oklahoma. RAVAZZI C. (2003) - An overview of the quaternary conti- nental stratigraphic units based on biological and climatic events in Italy. Il Quaternario - Volume Speciale INQUA,16(1Bis), pp. 11-18 RICHMOND G. M. (1996) - The INQUA-approved provi- sional Lower-Middle Pleistocene boundary. In: Turner, C. (ed.) The Early-Middle Pleistocene in Europe. Balkema, Rotterdam, 319-326. ROHLF, F.J. (1998) - NTSYS-pc Numerical Taxonomy and Multivariate Analysis System. Version 2.0. Exeter Software, LTD., New York. SALA B., MASINI F., FICCARELLI G., ROOK L. & TORRE D. (1992) - Mammal dispersal events in the Middle and Late Pleistocene of Italy and Water Europe. Courier Forsch.-Inst. Seckenberg, 153, pp. 58-68 SALVADOR A. (1994) - International Stratigraphic Guide. A Guide to Stratigraphic Classification, Terminology, and Procedure, 214 pp., The Geological Society of America, Boulder SARDELLA R. (2004) - Late Villafranchian and Galerian carnivores (Mammalia) fron the Tyrrhenian coastal area of Central Italy. In Kahlke R-D. & Maul L. (Eds.) Late Neogene and Quaternary biodiversity and evolution. 18th International Senckenberg Conference, Weimar 25th-30th April, 2004.Terra Nostra, 2004/2, p. 228. SHACKLETON N.J. (1995) - New Data on the Evolution of Pliocene Climatic Variability. In: VRBA E.S., DENTON G.H., P ARTRIDGE T.C. & B URCKLE L.H. (Eds.): Paleoclimate and Evolution, with Emphasis on Human Origins. Pp. 242-248; Yale University Press (1996), New Haven and London. SCHOCH R.M. (1989) - Stratigraphy: Principles and methods. Van Nostrand Reinhold, New York, 375 p. SEN S. (1997) - Magnetostratigraphic calibration of the European Neogene mammal chronology. Palaeogeography, Palaeoclimatology, Palaeoe- cology, 133, pp. 181-204. SHI G.R. (1993) - Multivariate data analysis in palaeoe- cology and palaeobiogeography -A review. Palaeogeogr. Palaeoclimatol. Palaeoecol., 105, pp. 199-234. SUC J.-P., BERTINI A., COMBOURIEU-NEBOUT N., DINIZ F., LEROY S., RUSSO-ERMOLLI E., ZHENG Z., BESSAIS E. & FERRIER J. (1995) - “Structure of West Mediter- ranean vegetation and climate since 5.3 Ma. Acta Zoologica Cracoviense, 38(1), pp. 3-16, Polonia. SUC J.-P., BERTINI A., LEROY S.A.G. & SUBALLYOVA D. (1997) - Towards the lowering of the Pliocene- Pleistocene boundary to the Gauss-Matuyama reversal. Quaternary International, 40, pp. 37-42, Oxford TEDFORD R.H. (1970) - Principles and practices of mam- malian geochronology in North America. American Paleontology Convenction Chicago, 1969, Proc. F, pp. 666-703. TORRE D. (1987) - Pliocene and Pleistocene marine-con- tinental correlations. Ann. Inst. Geol. Publ. Hung, 581Biochronology of the ... 70, pp. 71-77 TORRE D., ROOK, L. & FICCARELLI, G. (1999) -The mammal fauna turnover in Italy at the Early to Middle Pleistocene transition”. In: GIBERT G., SANCHEZ F., GIBERT, L. & RIBOT, F. (eds.) The Hominids and their environment during the Lower and Middle Pleistocene of Eurasia: 541-548. Museo de Prehistoria y Paleontologia “J. Gibert”, Orce TORRE D., FICCARELLI G., MASINI F., ROOK L. & SALA B. (1992) - Mammal dispersal events in the Early Pleistocene of Western Europe. Courier Forschungsinstitut Senckenberg, 153, pp. 51-58. TORRE D., ALBIANELLI A., BERTINI A., FICCARELLI G., MASINI F. & NAPOLEONE G. (1996) - Paleomagnetic calibra- tion of Plio-Pleistocene mammal localities in cen- tral Italy. Acta Zoologica Cracoviensia, 39 (1), pp. 559-570 TORRE D., ABBAZZI L., BERTINI A., FANFANI F., FIICCARELLI, G., M ASINI F., M AZZA P., R OOK, L. (2001) - Structural changes in Italian Late Pliocene - Pleistocene large mammal assemblages. In ROOK L., TORRE D. (Eds.): Neogene and Quaternary con- tinental stratigraphy and mammal evolution. Papers in honuor of Prof. Augusto Azzaroli’s out- standing contribution in Geology and Paleon- tology. Bollettino della Società Paleontologica Italaliana, 40 (2), pp. 303-306 VAN DAM J.A. (1997) - The small mammals from the Upper Miocene of the Teruel-Alfambra Region (Spain). Paleobiology and Paleoclimatic recon- structions. Geológica Ultraiectina, 156, pp. 1-204. VAN DAM J.A. (2001) - The Upper Miocene Mamad record from the Teruel-Alfambra region (Sapin). The MN system and continental Stage/Age con- cepts discussed. Jounal of Vertebrate Paleontology, 21 (2), pp. 367-385. VAN DAM J.A. (2003) - European Neogene mammal chronology: past, present and future. In REUMER J.W.F. & WESSELS W. (Eds.): Distribution and migration of Tertiary mammals in Eurasia. A volu- me in honour of Hans de Bruijn - Deinsea 10, pp. 85-95 VIRET J. (1961) - Artiodactyla. In PIVETEAU J. : Traité de Paléontologie, 6 (1), pp. 887-1021 WALSH S.L. (1998) - Fossil datum and paleobiological event terms, Paleontostratigraphy, Chrono- stratigraphy and the definition of Land Mammal "Age" boundaries . Journal of Vertebrate Paleontology, 18 (1), pp. 150-179 WOODBURNE M.O. (1977) - Definition and characteriza- tion in mammalian chronostratigraphy. Journal of Paleontology, 51, pp. 220-234 WOODBURNE M.O. (1987) - Cenozoic Mammal of North America: Geochronology and Biostratigraphy. University Califronia Press WOODBURNE M.O. (1996) - Precision and resolution in mammalian chronostratigraphy: principles, practi- se, examples. Journal of Vertebrate Paleontology, 16, pp. 531-555 ZANCHETTA M. & MAZZA P. (1996) - Anancus arvernensis remains from the basal portion of the Arctica islandica-bearing marine deposits of Lower Valdarno. Bollettino della Società geologica Italiana, 115, pp. 105-113. 582 Ms. ricevuto il 22 giugno 2004 Testo definitivo ricevuto il 30 novembre 2004 Ms. received: June 22, 2004 Final text received: November 30, 2004