Rivista ltaliana di Paleontologia e Srratigra *.".-*oo, l THE CASTAGNONE SITE (CERRINA VALLEY, MONFERRATO HILLS, N\I ITALY): EARLY PLEISTOCENE SEDIMENTARY RECORD AND BIOCHRONOLOGY CARLO GIRAUDI" ALBERTO MOTTURA" BENEDETTO SALA], M. STELLA SIORI'& DANIELE, BORMIOLI, Receit,ed May 6, 2002; accepted April 24, 2003 Key u^ords: earlv Pleistocene, Jaramillo Subchron, Mammal fauna, Italv. Abstract. Geological researches carried out near the Castagnone hamìet in the Cerrina Valle.v (Northern Monferrato Hills, Piedmont, N\l Italy), have brought to light a post-Messinian succession whose sedimentary record starts with a Lower Complex of pedogenized col- lur.ial materials and with two superimposed Alluvial Units (I and II). The low-er one of these units contains a Galerian macrofauna associated with n-ricrotine vole teeth (Mimomys satini, Mimomys pusillus, Unga- romys cl. U. nanus, Microtus (Allophaiomys) sp.), n-hile the upper one yields only scarce faunal remains. Most of this sediments were depos- ited durine a normal palaeomagnetic phase. The I Alluvial Unit, due to its biochronologìcal correlation, must be referred to the Jaramillo Subchron, between 1,070,000 and 990,000 years ago. The II Alluvìal Unit, being both unconformable with and younger than the first one, might be best referable to the Brunhes Chron. Or.erall, the bed dip- ping across the reported succession shos's a progressive syn-sedimen- taru tiltìng, with accelerated deformation durìng the I Alluvial Unit deposition. This tectonic stress over rhe Castagnone area is seemingly related to the uplift of the north-easternmosr ridge of the Monferrra- to Hills and appears to have been nearly exhausted before the II Al- luvial Unit deposition. Riassunto. Le ricerche geologiche condotte nei pressi dell'abi- teto dì Castagnone di Pontestura, in Val Cerrina (Monferraro Setten- trionale, Piemonte), hanno evidenzixro una successione sedimentaria post-messiniana che inizia con un Complesso Inferiore cosrituiro cla colluvi pedogenrzz:tr e da due unità sovrapposre formate da sedimenti alluvionali. tlUnità Alluvionale inferiore (I A.U.) contiene una macro- fauna Galeriana associata con denti di micromammiferi (Mimomys sa.oi- ni, Mimomys pusillus, Ungaromys cf . U. nanus, Microtus (Allophaìomys1 sp.), mentre ia superiore (lI A.U.) contiene solo scarsi resti faunìstici. Tuttr l:r serie sedimentaria risulta deposta in periodi a polarità magnetica normale. Per il suo conrenuro in fossili la I A.U. deve essere attribuita al Subchron Jararnillo, datato tra 1.OZO.OOO-99O.OOO anni fa. La II A.U. è più recente e discordante sulla I A.U. e sembra at- tribuibile al Chron Brunhes. Linclinaz.ione decrescenre degli strati che formano la successione sedimentaria indica una deformaz-ione sinsedi- mentaria progressir.a, parricolarmenre evidente all'interno della I A.U. Lo stress tettonico dell'area di Castaenone sembra in relazione con il sollevamento del margine nord-orient:le del Monferr.rto ed appare quasi completamete esaurito prima della sedimentazione della II A.U. Introduction The early Pleistocene sediments located in the Cer- rina Valley (Northern Monferrato Hills, Piedmonr, N\f Italy) preserve both lithological and fossil record which document palae oenvironmental variations, tecronic evo- lution and biochronology. The Cerrina Valley. is in a hi1ly area, mainly formed by Cenozoic marine sediments, with elevations of up ro 455 m a.s.l. The Monferrato and the relared Turin Hills form a central relief isolated from the main wesrern A1- pine ridge and are surrounded by the alluvial plains of two major rivers: the Po (N) and the Tanaro (S). The reported Pleistocene succession locally and in patches outcrops on the left slope at the end of the Cer- rina Valley (Fig. 1). Mrjor exposures are preserved near Castagnone (Pontestura, Alessandria) : on rhe local hills- lope the sediments are covered by a thin aeolian layer, while in the valley bottom the lower succession merges under the recent alluvium. In the late'ZOs fossiliferous gravelly-sandy alluvial sediments yrere first located by one of us (C.G.) during preliminary studies in a now disused quarry (Peratore quarry: C1 in the text and Fig. 1). There, some macrofau- nal remains suggested as possible an eariy middle Pleis- tocene age (Giraudi 1981), though the severe sediment deformation and the lack of arvicolid finds biased the biochronologic interpretation. From the Peratore quarry other fossil remains were collected in ensuing years in î ENEA., Centro Ricerche.Casaccia", Via Anguillarese,30l, OOO6O - S. Maria di Galeria (Roma). C.P 2400 - OO1O0 Roma A.D., Italy f -mail: qirrudiq/ cr\rccir.ener.il. Department of Animal and FIuman Biology, Via Accademìa Albertina, 17, 10123 - Torino, Itely. E-mail: siori@,dba.unito.ir Department of Natural and Environmental Resources, C.so Ercole Ii d'Este, 32,441AA - Ferrara, Ital,v. E-mail: b.sala(g_ìdns.unife.it Department of F,arth Sciences, Via Valperga Caluso, 35, 10125 - Torino, Italy. E-mail: geobormioli@,ìlibero.it 51u C. Girau.di, A. Mottura, B. Sala, M. S. Siori k D. BormioÌi sandy sllt aÌluv al val ey floor eol aî si r (Late Dleislocene' or sandy s lr of rhe lrd.2nd 1st alluvial terraces covering the I and I Alluvial Units E',1:;":':l:';"1':""::1""a l-lneoroc' \- îe.acescarps @'"'uo" still doubtful stratigraphic context (Giraudi 1983; SITEC 1984). More recentl,v, archaeological finds referable to dif- ferent levels of the same succession (not examined in this report) led to new extensive and detailed researches. Recent studies in a survey of the whole area have made it possible to localize new outcrops of the Castagnone sedimentary succession. In particulaq in a newly opened quarry (Cascina Nuova: CA1), a network of trenches, with an overall extent of about 300 mt and a depth of up to 5 m, were cut for better control of the stratigraphy, along with archaeological, palaeontological and magnetostratigraphic sampling. The following pre- liminary updating results allow the overall geological framework to be better known. Due to the small number of European locations with late early Pleistocene faunas, the reported Castag- Fìg. 1 - Geolosical sketch nap of thc lon er Cerrina Valle1,. 1: sandy silt of the alluvial vallev floor (Huluccrìeli 2: '.rnJ1 .ilr of tl-re .ltl-r elluvial terrace (late Pleistocene); 3: aeolian silt (late Pleistocene) or, local- 11; sandl' silt of the 3rd, 2nd and 1st alluvial terraces (mid- dle Pleistoccne?), w-hich cover the I ancl II Alluvial Units; 4: bedroch; 5: terrlce scarps. none succession improves the resolution of the European biochronological scale. The Castagnone/Pontestura sequence as exposed in the sections of the CA1 quarry The CA1 quarry (Fig. t) is in the upper part of the Castagnone hillslope, lear the top (193 m) of the local relief. The bottom bed of the sedimentary succession lie s with a slight unconformity over a bedrock of laminated clayey greylyellow sediments with hardened (chalky or silicized) interbedded levels. Such sediments are evidence of a brackish water facies with malacofauna (Girotti O., Esu D., pers. comm. 20OO) and were referred to a late Miocene .Messinian' formation (C.G.I. 1969). tL bed roc k lAU.+ Lithostratigraphv and strati graphicel units of the Castasnone/Pontestura oriented). L: loess cover; Il A.U.: pedogenized siltv gravelly sand (a) and silt end sandy silt rvith carbonatic concretions (b); sand (c) u'ith thin ancl (d); L. C.: consolidated silt and cla,v (a); sand, gravel and siìty cìa1'rvith N.ith (a) hardened levels (Messinian). succcssion. Cross section at the CA1 quarry' (N-S: ìeft- right send,v silt, silty sand and cla,v (b); I A.U.: sand and graveJ (a); discontinuous calcrete la1'ers (c1); gravelly sand and dark clal- a discontinuous calcrete layer (b-c); bedrock: laminrted ch1 Fio ) The early Pleistocene Castagnone site 519 The overlying succession starts, from the borom upwards, with the nl-ower Complex" unir, rwo ailuvial units, and ends with alayer of 1oess, as below (Fig. 2) . The Lower Complex (LC). It is formed by: a) clayey, massive, light to dark-grey, consolidated and cracked silt of colluvial origin, up to 40 cm thick; b) grey sand, gravel and silty clay of colluvial origin, with patchy rust oxidation, up ro 20-30 cm thick; c) at the top lies a pedogenic, lnore or less continuous calcrete, up to 10-15 cm thick. The first Alluvial Unit (I AU), This unit, about 10 m thick, lies in slight unconformity over rhe Lower Complex, and is formed by: a) grey-yellowish sand and gravel containing borh mac- ro- and micromammal remains; b) massive or poorly stratified grey-yellowish silt and sandy silt, with embedded carbonaric concretions, sized 1 to 15-20 cm, and some faunal remains; c) yellowish and grey unconsolidated sand, with thin and discontinuous calcreres and leaf marks; this sand is interbedded in the middle part of the silt and sandy silt layer; d) a heteropic facies of gravelly sand and dark clay over- lies the above sandy silty sedimenrs: rhe coarser frac- tions which increase ar rhe rop of the unit are fairly rich in fossil macro- and microfauna (the reported Elepbas sp. remains from C 1 were also in a correlated level: Giraudi i983). The largest raw clasrs which occur in the I AU are constituted mainly by silicized rvood or cherts. Actually, while other fluvial clasts (quartzites, jaspers and other alpine rocks) seldom exceed 6-7 cm in diameter, the sili- cized wood can atrain here 50-60 cm in length and weigh more than 1O kg; likewise, the largest cherty cobbles may 1 ^^ ^-be up to 20-25 cm rn diamerer. In the same unit, bony and cartilagineous marine fish teeth, echinoderm radiola and silicized inner casts of marine gastropods have been found, due to an intensive redeposition from the surrounding bedrock. The second Alluvial Unit (II AU). This enrire unir is weathered by a soil. It is 4-5 m thick, with definite un- conformity with the former unit. It is formed by: a) basal silty-gravelly sand with diffused yellowish clay chips: these sediments are weathered and contain badly preserved macrofaunal remains; b) sandy silt and silt with interbedded scanty lenses of siÌty sand and clay. During the Palaeomagneric srud- ies a marked increase in the total amount of mag- netic minerals with respecr to the I A.U. has been observed. The loess cover. In the uppermost slope of the hills, the topsoil whìch caps the above alluvial levels is buried in its turn by a loessic bed, everywhere thinner than Z0-80 cm. Discontinuous concenrrations of fresh, unpatinated Mousterian arrifacrs were embedded in this layer. Down the slope, the loessic covering and the under- lying alluvial unirs occur in different relationships. As the latter were cur by the subsequent drainage evolution and before the former occurred, the aeolian deposition took place on terraced slopes (mostly erosional morphologies in the CA1 relief). Hence there, the loess is directly in contact with the lower I AU or with the Tertiary bedrock, while the upper II AU is altogether lacking. The dip of the above different units progressive- ly decreases from the bottom upwards: the undulate "Messinian" sediments of the bedrock have an areal in- clination of about 14o-12", from the base to the top; the Lower Complex, formed by coliuvia and the calcareous crust, dips 12'to S (down the local slope); the I AU dip decreases from 1O'at rhe botrom ro 4o ar the top; final- ly the II AU is nearly horizontal. Two major erosional breaks in the Pleistocene sequence (i. e. rhose between the Lower Complex and I AU, and between I and II AU) correspond actually ro unconformable limits where the aggradation/tilting trend was also uncombined. As seen so far, the units of the above successìon are differently preserved along the hillslopes, both verti- cally and laterally: this fact is due to the mutual overlap and respective thickness and dip. Thus, in addition to its greater depth, the I AU is also more extensively preser-ved than the upper II AU: the latter, as it is nearly horizontal, is preserved only between the slope elevations of 193 m lthe top) and 185 m. . Discussion and interpretation The Lower Complex was characrerized by a soil strongly affected by the calcrete and carbonatic concre- tions. These pedogenic features have so far been recorded in the Piedmont region only in the "Villafranchian, type succession (Boano & Forno 1996).They are related usu- ally to high evaporarion rates mosrly controlled by rhe temperarure (Birkeland 1984) and so indicate an fairly warm but chiefly dry climate. The thin and discontinuous calcreres interbedded in the I AU dip S like the sand containing them, and there- fore must be nearly coeval: otherwise in . plrce affected by tilting their dipping would be different. The presence of calcareous concrerions and calcretes in the lower and middie part of the I AU, not affected by evident pedog- enetic weathering, must be due to the rapid ev.rporation above the water table. Towards the top of the I AIJ, a coarser deposition reappears with the energy renewal while the carbonatic precipitation decreases and stops: therefore a ciimatic change occurred during the sedimentation of this unit. Regarding the presence of silicized wood and raw clasts of cherts larger than other alluvial clasts in the I AU, it should be noted also that today, in the whole Cerrina Valley basin as well as in the Monferrato Hills, no other known formatìon contains so many cherts lboth as to va- 524 Specimen: l" A.U. - 1150 I 1E 02 1E-03 1 E-04 + 1 E-03 1É-A2 TE 01 TE+00 TE+01 Field B (T) Fig. 3 - Acquisitìon cun'e of the isothermal remnant nr:gnctizltion (IRM). riety and overall amount) as this unit. Therefore' during the I AU deposition, fluvial erosion exposed some (Ceno- zoic?) cherty formation which is now entirely eroded or covered by more recent sediments. A comparable redepo- sition of cherts was not active in the older Plio/Pleistocene regional bedloads: e.g., in the "Villafranchian" ones such pebbles (far smaller in size) rarely occur. Thus, during the I AU deposition a sudden important suppl;' of silex began in the regional clastic background. Silicized woods are frequent in the Cenozoic bedrock below the Castagnone sequence: their presence in the I AU is probably derived from the fluvial incision near this area, as inferable from the size range and minor reworking of such clasts. Also the bony and cartilagineous marine fish teeth, echinoderm radiola and silicized inner casts of marine gastropods were reworked from the bedrock. About the II AU, in respect to the I AU, a litho- logical change may be observed mainly in the contents of large siiicized wood, cherts and clastic load mineral- ogy; the content of reworked Tertiary fossils is smaller. These variations suggest possible drainage shifting and/ or climate changes compared with the I AU deposition. The absence of calcareous concretions and calcrete levels shows a different climate from that of the I AU. Magnetostratigraphy at CA1 The magnetostratigraphic study of the sections which were opened at CA1 allowed the palaeom..rgnetic trend across the entire ourcropping succession to be re- corded. The weakly consolidated sediments exposed in the cuts were previously cleaned from more superficial fractions in order to obtain optimized surfaces. The spe- cimens were collected in commercial plastic boxes (inner volume : 2x2x2 cm) gently pushed into the soft rock keeping the front face vertical, .rt 25 cm ìntervals (from the Messinian sediments up to the top) leaving out a few levels with too much incoherent sand1. lithologies; all the C. Giraudi, A. Mottura, B. Sala, M. S. Siori k D. Bormtoli specimens were oriented with bubble level and compass. The sampled section was about 16 m high and 62 speci- mens were taken in all. The magnetic mineralogy was investigated by iso- thermal remnant magnetization (IRM) using a Bussi pulse magnet. The resulting IRM acquisition graphs were char- acterrzed by a quick increase up to full saturation for magnetic field values of o.z-0.+ T (Fig. 3). The remnant coercive force (Hrc) maintains vaiues in the range of 40- 50 mT. These results suggest that the magnetite is the main ferromagnetic mineral in the tested lithologies; nev- ertheless, since thermal tests were not possible bec:ruse of the low coherence of the rock, the presence of iron sulphides cannot be excluded. The natural remnant magnetization (NRM) was measured with aJR-5 spinner. Intensity stays :rround the value of iO'A/m, with the exception of the specimens collected over 13.40 rn (from the I AU top upwards) whose intensity ranged between 1O'and 10'A/m. Al1 the directions, across the entire sampled succession, have a variable declination, ranging between N and E, and a positive inclination. Stepwise alternating field (Af) dem:rgnetization on pilot specimens, up to 1OO mT peak-field, showed the presence of a viscous component which is removed after the first steps; successively the direction remains stable up ro rhe field values,of 30-50 mT (Fig. 4). Sometimes the situation was more complicated: the presence of dif- ferent components whose cc,ercive spectra overlap made impossible to recognize the primary comPonent. The systematic demagnetization of all the collect- ed specimens made possible to detect the char.rcteristic remnant magnetization (ChRM) direction in 50% of the cases. Fisher's statistics were used to calculate the mean ChRM directions for each of the sedimentary units in the CA1 succession; the resuits are reported in Tab. I and Fig. 5. The precision k-parameter and the half-angle con- fidence cone show a good grouping for the data. The mean directions were then corrected for the respective inclination of the strata: L1"-L2" for the Messi- nian sediments; 12o for the Lower Complex; 1O' and 4' for the base and top, respectively, of the I AU; no cor- rection was made for the II AU which has maintained sub-horizontal strata. The inclination correction reduced the differences among all the recorded palaeomagnetic directions and so, in particular, the last three units probably assumed such corrected directions during the primary sediment deposition. Al1 the corrected directions for the entire succession at CA1 correspond to a norn'ral polarity time interval since they show an N-oriented declination and inclination ol 57o, well compatible with the local co-or- dinates lTab. I). The magnetic intensity values were correlated with those of the magnetic susceptibility for all the san-rpled $pecinnen: 1 'N50 ca 7A ta +J 1a JJ 25 2t) 15 ,lr0 Down Fig. I - AF demagnetization dìagram (specimen from LC + I AU). Svmbols: full dots - declination; open dots : apparenr rn- clination; figures : peak field value ìn mT. specimens (Fig. 6). This graph shows a good correlation among all of them and evidences the narrow grouping of data for the Messinian sediments. One can observe also that the II AU shows a marked increase in the total àmount of magnetic minerals, probably because of the more intensive clastic load discharges in this unit. This fact might be correlated with a more appreciable climate variation andl or with greater drainage changes since the final I AU deposition. Tbe earfu Pleistocene Crlstagnone site The palaeontological cA1 521 record and the biochronology at The vertebrate fauna remains from the Castagnone/ Pontestura succession exclusively come from the two al- luvial units and are best preserved in the lower one, ac- cording to the top soil weathering gradient. Indeed, the II AU is almost poor in such findings because of the pedogenic demineralization. On the contrary, the I AU is by far the richest in well mineralized fossiis, although often fragmented and rolled by rhe srream. The diagnostic macrofauna from the CA1 excava- tions is in fact mostly represented by isolated teerh, as more or less sorted within the sediments of the I AU (Fig. Z) . Diagnostic (at the genus level) non-denrary re- mains are only a distal fragment of the perone of a Hip- Popot.tmus sp. and a basioccipital skul1 rvith the horns of a Bison sp., coming from the base and rop, respectively, of the I AU. A whole macromammal association, as has now been recognized in the lower unit, is represented by Hippopotamus sp., Sus sp., Pseudodama sp., Stephanorbi- nus cf.S. hundsheimensis (Toula, 1l9a2), Bison sp. and oth- er Proboscidae, Bovini and Cervidae indet. In the Pera- tore quarry (C1), the previously reported finds (Girau- di 1981) and other unpublished ones (Ambrosettì pers. comm.) were from basal levels of the same I AU: also Dicerhorinus hemitoecbus'(recte: Stephanorhinus hunrlshei- mensis), Megaceros cf. verticornis (recte: Megaceroides cf. verticornis ) (Dawkins, 1872) and [Jrsus deningeri Re- ichenau,1906. Al1 these old and new findings from the I AU may indicate a generic Galerian (macro)Mammal age (sensu Gliozzi et al. 1997). An arvicolid microfauna has been samoled ar CA1 from two gravelly sandy levels of rhe I AU, at the base and the top respectively (Fig. 8). At the base of the unit Mimomys saoini Hinton, 1910, Mìmomys pusillus (Me- hely, 1914), Microtws (Allopbaìomys) sp. advanced form, and Ungaromys cf . U. ndnus a.re recorded. Together with these, other rodents and Insectivora occur: Sciurus sp., Apodemus sp. and Talpa cf . T. fossilis Pétenyi,1864. Due to the presence of Mimomys pusillus this microfauna is referable to the early Biharian (Rodent age), as defined by Fejfar 8e Horacek (1 990) . Since -M. pusillus is accom- panied also by an advanced form of Microtus (Allophaio- mys),the possible range must be constrained ro tr late or final time of the early Biharian. Besides M. (Allophaio- mys), the (Ìngaromys genus is a further biochronological marker in \le stern Europe. Either or both of these taxa occur rardily in rhe Collecurti fauna (Torre et al. nSeT and the Vallonnet fauna (Chaline 1988). By their own palaeomagnetic data both localities are calibrated in the Jaramillo Subchron as well as, in spite of some biogeo- graphic differences, the northern locality of IJntermass- feld (Kahlke 1992). Jr (A/m) NRM n k rr,,.DI ChRM D,lD,lnk0,, II'AU |.19 10 148',49' 12 18 6 357",57' 357"57. 8 1 /0 1. LC+I" AU 3.15 10 r J39",76' 26 5 11 6",57' 360',57" 1 1 27 9" Messinian bedrock 2.69 10 , 10",5e" 2e l0 9' I 1",59' 3s3',57" 11 59 5. Jl1 Tab. I C. Gìraudi, A. Mottura, B. Sala, M. S. Siori &. D. Bormiolt Paleomagnetic results from sediments of cAl successìon. Jr : NRM intensitlr NRM - nat- ural remn:nt mrqnerìz:rion. ChRM . chrr.rcrcri.ric r.n,,ìrn, rn:gnerization af,"rìi,r.r,- ment. D, I : declination, inclination. Dt, lt - declìnation, inclìnation after tilt correcrion. n : number of specimens. k - Fisher's precision paramerer. 95' = semi-angle of confide'ce. The microfauna ar rhe rop of the I AU is less represent- ed because of the soil effects: apart from Orcto/agus c[. /acosti tPomel, 1853), MimomIS saoini and a .ingle M' of Microtus sp. have been recorded. This latter (Fig. 8, n. 6) shows poorly differ- entiated enamel and closely cor- responds to homologous teerh of Micrcttws (Allophaiomys) from the same I AU base (Fig. 8, n.7) and from Monte Peglia (Meulen van der 1973). This material by itself cannot permit a sufficiènt biochronologic assessment. Nev- erthelesr. .rlong with a persi\rence ol normal mrgneric po- larity up to rhe top of the unit, the cf. M. (Attophaiomys) occurrence (but taking it with prudence) keeps here the correlation within the Jaramillo Subchron as well as, see m- ingly, the whole lithostratigraphic evidence (Fig. 9). At CA1 the ar-vicolid elements collected from the II AIJ are limited to some incisor reerh rhar do nor supporr any determination. The macrofauna there is represented only by fragmentary teerh of Stepbanorhiuzs sp. and of herbivorous indet. This upper unir, because of the top soil, lacks exhaustive biochronologic markers and remains at moment unresolved in its own correlation with the pos- sible palaeomagnetic optioht: the Jaramillo Subchron or alternatively the Brunhes Chron. The evolution of the setting and environmental infer- ences The above lithostratigraphic, palaeomagneric and biochronologic data allow us ro assume that at leasr the whole I AU was deposited during the Jaramillo Subchron (1.07 - 0.99 m.y., according to the .orbitally tuned time" scale of Cande & Kent (1995)',. The Lower Complex was deposired beneath the I AU under normal p.m. polarity but, due to the slightly unconformable contact between these units, one cannot exclude that the former deposition occurred well before the Jaramillo Subchron. The chronology of the II AU is by far more prob- lematic. Though this unit has a normal p.m. polarity, its unconformable contact over rhe I AU misht be in favour of a somewhar younger age: ìf not in the same Jaramillo Subchron rhen necessarily within the Brunhes Chron. The age of this upper unit hence is possibly iower than 0.78 Mr. In spite of the above gaps in the data, a prelimi- nary reconsrrucrion of the geologic evolution in the area is suggested as below. After the ma1'or erosional phase supported by the top of the "Messiniano 5sjimsnts (not yet deformed) Due to its normal magneric polarity, at least the base of I AU must fall necessarily inside the minor re- versal of the Jaramillo Subchron (in the Matuyama inverse age) and nor in the Brunhes Chron, since the arvicolid evolution level there precludes it. Likewise, the associa- tion of M. savini, Mimomys small size form, Microtus (Al- lophaiomys) advanced form, and (Jngaromys cf. (J. nanus is very similar to rhat from Monte peglia (Meulen van der 1973), a site which has been recenrly located not lons be_ fore the Jaramillo Subchron (Masini et al. 199g). Fig. 5 - Equal areaprojection ofthe ChRM directions (square = site mean value with alfa-95 ellipse of confidence). N-- -a- -- -.. N-'- -? r,qf t- Mcssinian bedrock LC + IO ATi IIO AU The early Pleistocene Castagnone site 1360 al I tgeo .tt . :î^L1Fi^-' :. I I I J (AJm) 1 E+01 '1E-04 Fig. 7 - Dental remains of large Mammals îrom the I AU ler.cls. 1: C, fragment oi Hippopotamus sp.; 2: rjght M. ol Stephanorhinus cl. huntlsheìmezsis, a) lingual viel., b) occlusal vieu'; 3: left P. of Pseudodanta sp., e) lingual vìeu-, b) occlusal vicu'; 4: right Nl, oÍ Pseudodama sp., a) buccal viel'. b) occlusal vìer'; 5: rìght P,ol Bíson sp., a) lingual vieq b) occlusal r.ìer.. Scale bar: 3 cm. Fig. 6 523 Relation between magnetic in- tensity (J) and magnetic sus- ceptibiìity (K). a) - Il'A.U. A : Messinian bedrock T : I. A.U, + L,C. Numbers refers to the position of the specimen in the succes- sion. and during a subsequent time span under stable condi- slow colluvi.rl .rccrerion with soil co-evolurion occurred. tions, more probably before the Jaramillo Subchron, a This soil was characterized by the calcrete and carbonat- ic concreîions which heavily af- fect the Lower Complex. Such pedogenic features have so far been recorded in the Piedmont region only in the "Viliafranchi- an> tyPe succession (Boano 8r Forno 1996). After a further small ero- sional phase, the I AU sedimen- tation took place along a water channel wirh high e nergy. In Lhe -i,lJl" .,rt ^f tLe I A U the en- ergy decreased and a meandering pattern prevailed, with lateral mi- grations. Due to the fast evapo- ration, along with the sand and silt aggradation, thick calcretes and concretions formed above the water table. Towards rhe top ol the I AU. à coarser deposition re- appeared with energy renewal while the carbonrtic precipita- tion decreased and then stopped. The floodplain which developed during the Jaramillo Subchron was at the same time tectonically deformed, even though the pro- r-re\sive tiltin- rrr" remainedb^r""'' compatibÌe with the continued aggradation. This major land- scape vrrirtion is hard to speci- fy here at local detail but, on the whoie, it is certainly linked with the uplift of the North-eastern- most Monferrato ridge. w", @n, al W*, ót /'\ l"?ts\\ Aa<\> n ffM8ÚB W 6789 f-\ /^\/ e) /\ U\, D a ?(,,>>R<"7\4"-r ,--K< \-1ì -..-v ==za"'z -\\ è )/t4 4 €. é= @[>è r-A>> ili \l 10 Jl+ C. Giraudi, A. Mottura, B. Sala, M. S. Siori t< D. Bormiolt Fig. 8 - An'ìcolid molar teeth from the I AU levels at CA1. 1-5: M/1, Microtus (AlLophaiomysl sp.; 6: M3/, Microtus (?A/- lopbaiomls); 7: M3/, Microtus (Allophaìontys) sp.; 8: M/2, Llngaromys cf . Ii. nanus;9:Ml 1, Mimomys pusìLlus; 10: M/ 1, Mimomys sa'c,itù (1,orng); 1 1 : M/ i, ùlimomys sacini (adùt) . Scale bar: 1 mnt After a second erosional hiatus, the aggradation of the II AU is more recent than the offset of the major tec- tonic deformation event. Therefore the hiatus itself con- tains the end of the deformation. The following desicca- tion of the II AU floodplain was possibly due to climatic changes or to a final but altogether different tectonic re- newal. A similar Pleistocene incision of the land surface is recorded from adjacent areas of central Piedmont (Car- raro 8. Valpreda 1991) in the late middle Pleistocene. At this moment, such an age might reasonablv be the case also oI the reported situ.rtion. Summary and Conclusions The post-Messinian sedimentary successron Pre- served in the hi11s near Castagnone (Pontestura) has been studied again by means of a network of suitable cuts car- ried out expressly for the purpose of lithostratigraphic, magnetostratigraphic and biochronologic assessment. This succession is formed by a Lower Complex of colluvial sediments with calcretes and carbonatic concre- tions of pedogenic origin, by a I Alluvial Unit of fluvial sediments with vertebrate macro- and microfauna (most significant is the arvicolid association with two Mimomys species together wrrh Ungaromys cf . U. nanus and Micro- tus (Allophaiomys) sp. and by a II Alluvial Unit of fluvial sediments with scarce faunal remains, in its turn covered by a thin loessic layer with embedded Mousterirn rrti- facts of "Vùrmiàn" age. The same succession, though broken off by ero- sional hiatuses, is entirely formed by sediments with normal palaeomagnetic polarity. A reliable chronologi- cal framework has been obtained for the I AU which contains most of the faunal remains allowing for a direct correlation with the Jaramillo p.m. Subchron, between l,O70,OOO and qgO,OOO years ago. No chronological infor- mation results from the ol-ower Complex" whose true age still remains uncertain. Likewise, on the basis of the lithostratigraphic evidence alone, the upper and younger parts of the succession are apparently best referable to the Brunhes Chron. The sedimentary facies in the I AU indicate a fluvial setting with energy decreasing from the bottom upwards, and a renewal of the flood energy at the top. The progressive tilting of the sedimentary beds in the lowermost succession indicates also a syn-deposition- al tectonic deformation, more marked during the I AU de- position. Such a deformation which tilted the fluvial strata southwards would not 3èerii to involve the II AU by itself. The tectonic stress over the Castagnone area is seemingly related with the uplift of the North-eastern Monferrato Hills and, even though already noticeable in the earlier phases, it strengthened during the Jaramillo time span then becoming nearly exhausted before the II AU deposition. As regards the time interval for the I A.U. mam- mal associations - with four arvicolid species - this must be limited to the late early Pleistocene. The cross calibration between palaeomagnetic and palaeontological data makes it possible to refer the fossiliferous unit of the succession to the Jaramillo Subchron. Together with the Vallonet, IJntermassfeld and Collecurti localities (Chaline 1988; Kahlke 1997:.Torre et aL. 1996), the Castagnone site is one of the few Eu- ropean sites with a crucial and still poorly known early Pleistocene tirne span. During the late early Pleistocene, indeed, the ma- jor Quaternary faunal turnover took p1ace, along with the climate cooling towards the coldest middle Pleistocene glacial phases (Gliozzi et al. 1997; Markova & Kozha- rinov 1998; Rekovets 8r Nadachowski 1995; Sala et al. 1992:Turner 1992). The reported succession thus represents rn im- provement of the whole European biochronological time scale. The early Pleistocene Castagnone site Fig. 9 - Biochronolog,v of the Castagnone sitc in tbe frame of the Italian Pleistocene faunas (modified from Masini er al. 1998) 525 Acknou^ledgemen ts. \íe wish to thank P Guarnero, owner of the reported quarries, for permission ro carry out the field excava- tìons and support, O. Girottì and D. Esu ("La Sapienza" University of Rome) for palaeontological assessments of thc Messjnian bed- rock, Dr. G. Muttoni and an anonimous referee for the improve- ment of the manuscript. \ù/ork supported by the Italian C.N.R (n. 96.003 /0.CTO5;97 .a4fi1.CTa5;98.00297 .CT05) and MURST grants to the authors. REFE,R Birkeland P\( (1981) - Soils and Geomorphology. Oxford (UK): Oxford University Press. Boano P & Forno M. G. (1996) - La successione tipo villa- franchiana. Litostratigrafia. In: (F. Carraro Ed.) - Re- visione del Villafranchiano nell'area-tipo di Villafranca d'Asti. 1/ Quaternario, g(1): 38-62, Roma. Cande S.C. & Kent D.V (1995) - Revised calibration o{ the geo- *^--^':^ ^^l^-:'-- time scale for the late Cretaccous andl-,urdrrr) Cenozoic./. Geophys. Res., 100: 6A%-6a95,'Washington. Carraro F. & \lilpreda E. (1991) - The Middle-Upper Qua- ternary of the Asti Basin. 1/ Qwaternario, 4: 151-172, Roma. C.G.I., 1969 - Carta Geologica d'Italia (1:100.000): Foglio n'57, "VerceÌ1i". Sen'izio Geologico d'Italia - I.G.M, Firenze. ENCES Chalinc J. (1988) - Les Rongeurs de la Grotte du Vallonet (Alpes-Maritimes), l'environnemenr er l'age du site. LAnthropologie, 92 (2), 497 -499, Paris. Fejfar O. 8c Horacek L (1990) - Review of fossil arvicolids (Mammalia, Rodentia) of the Pliocene and Quarernarv of Czechoslovakra. Acta Int. Symp. E,Lol. Phyl. Biostr. Aruicolids, pp. 125-132, Praha. Giraudi C. (1981) - Presenza di depositi medio-pleistocenici in- . tensamente deformati in Val Cerrina (Monferrato Setten- trionale). Geogr. Fis. Din. Quat.,4: 69-71, Torino. Giraudi C. (1983) - Pontestura, Loc. Roletto (prov. Alessandria). Scavo paleontologico. Quaderni della Soprintendenza Ar- cheologica del Piemonte 2: 1.43-1.44, Torino. Gliozzr E., ,\bbazzi L., Argenti P., Azza.roh A., Caloi L., Ca- 6 18 o ("/*) composite curve Shackleton, 1995 5.0 4-5 4.0 3.5 3.0 F*,.i" 7 9 11 B 10 2- 14+ ;_=,, - <>f7 U: 20 -- 1e ------3' zt,ì-=-23 )ó {> 2s-w: :+,€ 33 qs --è 35-lflPun" 4>:=- 1r^.4^1=__)i ?"s-;: 5B 60- 62- o+ 6 68- 70- 72 +sàq6 >__)f, €_"- -- ,?9:-<,%" <*_Y' (] 0 c a'r e c i! c e 0 c € c0 U L- u 0,0 0.1 0.2 0.3 {, {, 0.5 I Mammal I Rodent lFaunall Calibrated Jaramillo I ltalian Localities Ages I Zones I Unitrl italian 'i:" l'î'' I .,i,'i.ì lirturvll ti"r!'fl' Subchron I with callDrateo I Microtus rf.|îÎ|:,f,:' I GuonnaiomYs) c e a f l,:,:*1',:,, UJ z U O J È c .9or'. 0 I T 6 a 0 q 0.: oE (? T ro 0 C .!g Cr: FO ! \ AwiCOla I rur rtr il r -. I Preffacanabnusl i'i,' F0nla na uccí F, U. Isernia F, U. Slivia FU, Colle Cu*i F.U. sauni I pirro F,U. Fa rneta puaais I F U F.U, ".. 1 Olivola F.U, Minwmys prrocasrbs vitÍnia Ìsotopic stage 7 Torre in Pìetra icnr..i. d:^È q F. Ranuccio 0.458 KlAr Blddittu etal. 1979 Isernìa 0.730+/-0.04 K/Ar Cotort etal. 1981 Vitinia F.U, 0.6 0.7 0.8 v =1.0 É ru 1.I eC r.2 : 1.3 s {E rE> # ," rg'" E 1.7 1.8 l! : 1.e ! n 2,0 ll:?riíZ Castagnone (l A,U). M. (A.) sp. Colle Curti M, {A,) sp. M. Peglia A M.(A.)nutiensis lvl. Peglia B M,(A,)buryondíae Soave C. Sud M, {A,) ruffai Pirro Delì Erba M, (A.) cf. rutroi Pietrafittd M.{A)d,ntfui M. {A') chalínei Pl. La N4esa M,{A)6.pliffie,1íc:/6 Torre etal. 1996 Le Vallonet Chaline 1988 U ntermassfeld Kah ke 1997 ul z r.t i U f-l È u) u J 526 C. Giraudi. A. Mottura, B passo Barbato L., Di Stefano G., Esu D., Ficcarelli G., Girotti O., Kotsakis T., Masini F.,Mazza P, Mezzabotta C., Palombo M.R., Petronio C., Rook L., Sala B., Sardella R.,Zanalda E., Ec Torre D. (1997) - Biochronology of se- lected mammals, molluscs and ostracods from the Middle Pliocene to the Late Pleistocene on Italy. The status .of the art. RitL. Ital. Paleont. Strat., 1.03:369-388, Milano. Kahlke R.D. (1997) - Das Pleistozàn von Untermassfeld bei Meiningen (Thuringen). Monogr. Ròmiscb-Germanisches Zentralmuseum - Forschwngsinstitut fùr oor- und frùhges- chichte,40/1.418 pp, Bonn: Rudolf Habelt GMBH. Masini F., Abbazi L., Lippi P, Sala B. & Torre D. (1998) - Re- view and new finds of Microtws (Allophaiomys) (Roden- ti. A ^.i.^ril".\ {"om the Early Pleistocene of the italian peninsula. Paludicola, 2(1):75-9a, New York Markova A. & Kozharinov A. (i998) - Allophaiomys oî the Southern Russian plain. Paludicola, 2(1): 62-69. New York. Meulen A.J., van der, (1.973) - Middle Pleistocene smaller mam- Sala, M. S. Siori k D. Bormioli mals from the Monte Peglia (Orvieto, Italy), with spe- cial reference to the Phylogeny of Microtus (Arvicolidae, Rodentia). Qwaternaria,LT: 1-144, Roma. Rekovets L.I. Ec Nadachowski A. (1995) - Pleistocene voles (Arvicolidae) of the Ukraine. Paleontologia i evolucio, 28-29: 145-245. Sabadell. Sala B., Masini F., Ficcarelli G., Rook L. Ec Torre D. (1992) - Mammal dispersal events in the Middle and Late Pleis- tocene of Italy and western Europe. Courier Forsch. Senckenberg, 153: 59-68, Frankfurt a M. SITEC (1984) - Stwdio geologico e geomorfologico dell'Area PO 1. ENEL: Internal Report. Torre D., Albianelli 4., Bertini A., Ficcarelli G., Masini F. & Napoleone G. (1996) - Paleomagnetic calibration of Plio- Pleistocene mammmal localities in central Ialy. Acta Zo- ol. Cracov., 39 (1) : 559 -570, Krakow. Turner A. (1992) - Villafranchian-Galerian larger carnivores of Europe: dispersion and estinctions. Courier Forsch. Senckenberg, l53: 153-160, Frankfurt am M.