Rivista Italiana di Paleontologia e Stratigrafia volume 104 numero 2 a:oìne )9.7-)9L Agosto 1998 NOA BREVry REMARKS ON THE STRATIGRAPHY AND BIOCHRONOLOGY OF THE LATE PLEISTOCENE DEPOSIT OF INGARANO (APULIA, SOUTHERN ITALY) CARMELO PETRONIO* E. RAFFAELE SARDELLA*,I Receiaed Decenrber 10, 1997; accepted March 19, 1998 Key-oorcls: Biochronologn Stratigraphy, Late Pleistocene, Ver- tebrates, Southern Italy. Riassunto. Ulteriori considerazioni sulla stratigrafia del deposi- to di Ingarano (Foggia) consentono di rpotrzzare la presenza di tre associazioni faunistiche riferibili ad un intervallo ben definito del Pleistocene Superiore. La associazione pirì antica (Ingarano a) può essere riferita allo stadio 4 delle paleotemperature, quelle più recenti rispettivamente allo stadio 3 (Ingarano b) e allo stadio 2 (Ingarano c). I dati faunistici si accordano con I'alternanza delle diverse condizioni climatiche nel corso del Pleistocene Superiore. Abstract, New field data on the Late Pleistocene deposit of Ingarano (Foggia) allow us to hypothesise the occurrence of three fau- nal assemblages (Ingarano a, b, c) respectively referable to isotopic stages 4, 3 and 2 of the palaeotemperature scale. The palaeontological data match with the alternating palaeoclimatical conditions during Late Pleistocene times. Introduction. The Late Pleistocene karst cave deposit of Ingara- no is exposed at 270 m a.s.l. along the railway of Garga- no) near Apricena (Foggia, Southern ltaiy). The first studies started in the second half of '8Oties, but only sínce 1992 the complete stratigraphical succession has been exposed by quarry works. These ac- tivities and the previous quarry ril/orks partially disturbed or hid the stratigraphy of the cave succession, which ap- pears chaotic. From this succession a rich vertebrate fau- na and some lithic tools have been recorded and descri- bed (Capasso Barbato er. aI., 1992; Petronio et aI., 1996). The study o[ the fossils allow us to point out rwo possible biochronological hypothesis: a single or some distinct faunal assemblages. In order to test these two hypothesis and to point out a more detailed stratigraphy of the cave succession new geological field surveys have been carried out in the area. The vertebrate fauna has been considered following the updated biochronological framework for the late Aurelian Mammal Age (Glíozzr et aI., 1997). These new researches allow us to point out a more detailed stratigraphy and biochronology for the locality of Ingarano, with the definition of three distinct faunal assemblages and the general lines of the paleoeco- logical evolution of the area. Stratigraphical framework. Two fossiliferous deposits can be identified (Fig. 1): 1) a sandy-clays deposit (b), includingvery big calca- reous blocks, which is exposed in external position to the cave deposit;2) a successiofl oÍ 12 m, which rcstifies Fig. 1 - Stratigraphical sketch of the Late Pleistocene fossiliferous deposit of Ingarano (Foggia, Southern Italy): a) calcareous bedrock (Calcari di Sannicandro Fm.); b) silty-clays with calcareous blocks; c) alabastrine and encrusted phosphatic layer; d) conglomerate with round and flattened calcareous pebbles of smrll size (less than one cm) with silty reddish matrix; e) conglomerate with calcarenitic large sized peb- bles (1 m length), cemented by a calcareous matrix; f) con- glomerate with calcareous pebbles scarcely rounded (sized more than 10 crn), cemented, with a prevalently calcareous -"r,;*. ol i."^h".".î ";lr- ."..1;-"., Sapienza" Ptazza Aldo Moro 7 , Roma, Italy. e-mail: petronio@axrma.uniroma.it Università di Roma "La Sapienza". 'f Dip. Scienze della Terra, Università di Roma "La 'r'r Borsa Post-Dottorato - Dio. Scienze della Terra. 288 C. Petronio & R. Sardella different phases of the filling of an ancient cave, today destroyed by the quarry works. The sandy-clays sediments (b) are in unclear strati- graphical relationship with the cave succession and con- tain remains of several fossil mammals: Stephanorhinus hernitoechus, Coelodonta antiquíatis, Hippopotamus ampbibius, Elepbas a.ntiquus, Pantbera spelaea, Vulpes oul- pes, Apodemus syhtaticus. The Calcare di Sannicandro (a) constitutes the bed rock of the cave, in which an intense karst activity for- med stalagmites and a thick alabastrine layer. These layers are encrusted with phosphatic material (c). A geo- F;o 2 - l.'..,'^ /F^..;' s^"rL.'.^ ,b. " Italy). Layer b: 1) Stephanor- hinus bemitoecbus: fragmenta- ry maxillary bones with teeth (occlusal view); 2 and 3) Coe- lodonta antiquitatls: [ngmen- tary humerus (posterior and anterior view). All the figures ''" ...'^"i-"r.ì., a(or^ .'r"- ral size. .L^-: .-r .- -r...: . with thellÌcrlrrLdr 4rr4IyJrù rrrTh/2r4U merhod on differenr samples of the speleothem gave an age of 40.000 +- 2.000 for the phospatic concretion due to the guano of the birds which lived in the area, while the speleothem stopped its growing àt about 58.000 +- 2.000 (Petronio et al. 19961 From lrwer r come the vertebrate'bones represenîing se- veral taxa: Aves - Aquila cbrysae- tos, Columba livia, Pyrrhocorax graculus; Mammalia - Cervus elaphus, Dama dama dama, Lynx lynx, Canis lupus, Canis ex gr. ar- nensis-mosbachensis, Vulpes oul- pes, Martes sp., Mustela ni,ualis. Above layer c a conglome- rate with round and flattened calcareous pebbles of small size (less than one cm), with silty reddish matrix, poorly cemented is exposed (d). This conglomera- te gradually evolves upv/ards into another conglomerate with cal- careous scarcely rounded pebbles o[ larger size (more than 10 cm), more cemented, with a prevalen- tly calcareous matrix (e). A grear number of fossil bones (repre- sented often by some isolated re- mains) comes from these two layers: Equus Ì.rydruntinus, Bos primigeniws, Cerous elapbus, Capreolus capreolus, Rupica- pra sp., Felk silaestris, Lynx lynx, Panthera pardus, Crocu- ta crocuta, Canis lupus, Canis ex gr. arnensis-mosbachen- sis, Vulpes oulpes, Ursus arctos. At the top of the succession, with a thickness of almost 3 m, a .o.rg1o-..rte with calcarenitic large sized boulders (1 m length), cemented by a calcareous matrix (f), is exposed, testifying the phase of closing of the cave, when the vauit fell down. The passage between the layers e and f is charac- terised by a different inclination of the deposits; twelve lithic implements of Levaliois technique comes from the Stratigraplty and biochronology of tbe Late Pleistocene of Ingarano 289 Ingarano (Foggia, Southern kaly). Lryer c: l) Lynx lynx - skull (at the top) (palatal view); emimandible (at the bot- tom) (labial view). Al1 the figures are approximately 70o/o natural size. top of layer e and their state of preservation allow to hypothesise a very low degree of transportation (Petro- nio er aI., 1.996). Some smail galleries, probably connected together, exist into the levels d and e. They are partially filled with a silty incoherent sediment (g), with severai fossil bones of micromammals and birds (Petronio et al., 1996). Some bones of mammals and birds are partially covered by phosphatic encrustation and for this reason are clearly coming from layer c. The taxa surely coming from Iayer g are: Aves - Aquila chrysaetos, Falco peregrinws, Falco tinnunculus, Alectoris graeca, Circus nov. sp., Perdix perdix, Columba lioia, Nyctea scandiaca, Pyrrbocorax graculus, Pynhocorax sp., Corvus corax, Corvus rnonedula; Mammalia - Erina' ceus europaeus, Myothis blythi, Oryctolagus cuniculus, k- pus europdeus, Arvicolidae indet., Microtus (fenicola) sp., Microtus ex gr. aroalis-dgrestis, Apodemus syhtaticus, Eliorrrys quercinus. Fauna. Layer b. The fauna assemblage coming from the sandy- clays sediments is characterised by the presence of the "warm" species Stepbanorbinws hemitoechus and the "cold" Coelodon ta an L iqu ita Lis. Stephanorhinus hemitoecbzs is represented by skull fragments with a complete series of teeth - Fig. 2.1-, a complete tibia and other parts of limb bones, probably belonging to the same juvenile specimen. Even if it is a juvenile specimen some dental features allow a quite sure taxonomical determination. Premolars and molars show the peculiar undulate profile of the ectoloph and the tendency to develop quite ipsodont teeth. These characters are typícally ol Stephanorhinus bemitoechus and are shared with the "etruscoid" forms. Coelodonta dntiquitatis is represented by a frag- mentary upper tooth (Capasso Barbato et al. t992), a paftial humerus (fig 2.2, 2.3), carpal bones and some metacarpal in anatomical connection, radius and tibia). The fauna is completed by Hippopotamus amphibius (a fragmentary molar and a proximal epiphysis of a fe- mur); Elephas antiquus (a fragment of a molar), Panthera spelaea (part of a very large sized tibia), Vulpes oulpes (some limb bones) and Apodernws syloaticus (a complete emimandible). Cave succession. Layer c. Bird remains are represented by several limb bo- nes, with Pyrrhocorax graculus the most common spe- cies. Among mammals, the boreal lynx is the best repre- sented taxon with skull fragments, mandibles and limb bones referable to almost five specimen (Fig 3). A study still in progress is evidencing some peculiarities of these lynxes. The skulls are comparable in size to the actual Scandinavian Qnx lynx, but show proportionally less developed teeth, intermediate in size between the north- ern lynxes and the Pieistocene samples of Lynx pardina spelaea (\ferdelin 198 1). Canids and mustelids are quite common in this layer. The occurrence of Canis lupus ts testified by seve- ral limb bones, while also a small wolflike dog occurs. The presence of a middle sized canid in Late Pleistocene deposits from Southern Italy have been considered as a persistence of a taxon of Galerian origin (Di Stefano et al., 1992; Capasso Barbato &. Glíozzí, 1996). The syste- matic position of this dog is still unclear. The Mediter- ranean Galerian deposits are characterised by the occur- rence of Canís aff. arnensis (Rook & Torre 1996), while the small wolf Canis mosbachensis is a peculiar element 294 C. Petronio G R. Sardella Fìg.4 - Ingarrno (Foggia, Southern Italy). Layer c: Vulpes vulpes - 1) emimandible (labial tiew); 2) fragmentary skull (palatal view); 3) skull (sagittal vrew); Canis lupus - 4) emimandible (labìal view); 5) skull (palatal view). All the figures are ap- proximately 607o natural size. of the coeval European localities. Until new studies will clarify the relationships between these middle sized ca- nids and their dispersal during Pleistocene times, we re- fer the fossils to Canis ex gr. "arnensis-mosbachensis". Vulpes vulpes (Fig. 4.1, 4.2, 4.3) is a very common ele- ment and the skulls, the mandibles and the limb bones testifies the occurrence of at least six individuals. Martes sp. is represented only by an os penis, whtle Mustela ni- aalis occurrence is testified by four almost complete skulls. Among cervids an adult specimen of Ceraus elaphus have been recorded (mandible, vertebrae and limb bones of), while Dama dama dama is represented only by few fragmentary remains (Fig. 4.4, a.5). The modern subspecies of the fallow deer is an important biochronological marker. The modern fallow deer ap- pears in ItaIy at the beginning of the Aurelian mammal age in correspondence of the isotopic sfage 7 with the subspecies Dama dama tiberina (Di Stefano 8c Petronio 1997). This cervid evoives in Dama dama dama in isoto- pic stage 5 and it characterises the Late Pleistocene cie- posits. Despite the fragmentary remains the evolved fea- tures in premolars and molars allow to refer the fossils recorded to the modern fallow deer. Layers d and e: These layers are characîerised by the abundance of the carnivores remains (in particular almost complete skulls or cranial fragments). Ursus arctos is represented by a complere skull with mandible - Fig. 5 -, partially crushed, atlas and epistropheus; among fehds Lynx lynx is the most common taxon (several skulls, mandibles and limb bones), whiie the occurrence of Panthera par- dus and Felis sih,estris is recorded only by fragmentary remains. Canis lupus and Vulpes vw/pes are quite com- mon (at least three specimens for every taxon), while seems there is no evidence of the presence of the middle sized canid; Crocuta crocutA, usually very common in cave deposits, is really poorly represented (a fragmenta- ry cubitus and a metapodial). Herbivores remarns are less frequent; Equus lrydruntinus is represented by a complete mandible (Capasso Barbato et aI., 1992) and some jugal isolated teeth, while the occurrence of Cer- ous elaphws, Bos primigenius, Capreolus capreolus, Rupica- pra sp. is testified only by fragments of mandibles and isolated jugal teeth. Layer g: Among the birds, to which belong a very high percentage of the fossil bones of this layer, the most pa- leoecologically significant taxon is Nyctea scandiaca. The avifauna is characterised most of al1 by the abundance of ràptors and corvids. Several micromammals come from this layer. The best represented species rs Microtus ex gr. agrestis, while Apodemus sylaaticus, Eliomys quercinus and \èrricola sp. are quite rare. Probably micromammals may be conside- red as prey of the raptors which inhabit the area. Discussion. In the first study on this deposit (Capasso Barbato et al., 1992), the presence of two distinct faunal assem- biages, respectively referred to the stage 4 of the isotopic scale (the fauna with "pachyderms') and to the stage 2 Stratigraplry and biochronolog of the Late PLeistocene of Ingarano (in particular the avifauna, with the "cold" taxon Nyctea scandiaca), have been hypothesised" In a further work, (Petronio et a1., 1.996) the hypothesis of a single faunal assemblage, referred to the isotopic stage 3, was considered as more probable for the coexistence of "cold" and "warm" taxa. New field data allow us to test these different hypothesis showing a more articulated stratigraphical framework outlined in this paper. At the present time, there is no evidence of direct stratigraphical relationships with the cave succession for deposit b, with the "pachyderm" remains. No fragments of "pachyderms" come from the other layers. The layer b may probably be older than the cave succession. The coexistence of two species of rhinos, ecologically di- stinct, of the hippo and of the forest elephant may be explained, if we consider the faunal assemblage coming from the layer b as homogeneous, referring it to the iso- topic stage 4 (Ingarano a, Fig. 6). In particular the chronology of this fauna is defi- ned by the occurrence of Coelodonta antiquitarls, which was widespread in Italy in Late Pleistocene since the iso- topic stage 4, while the other iarge mammals are fre- quent in the isotopic stage 5 and survive until the isoto- pic stage 3 (Late Aurelian Mammal Age, Gliozzi et a1., 1997). However the coexistence oí Coelodonta and Ste' phanorhinus, even if rare, has been recorded in some Eu- ropean localities as Achenheim (Alsace, France) and Ba- lauzière (France) (in Guerin, 1980, with bibliography). The occurrence of the woollv rhino, in association with I-ig.5 - Irrgrrrno (Foggir, Southern Italy). Layer d: Ursus arctos - skuil and mandible (approxì- rnately 45% naturel size). r ho ^r ho, ^,.L-,,1.,*- -.L;^1"'-rlc uLrlLr PdLlr)uqr lll)r wlllLll show.r wider ecological srgnifi- c:ìnce. may restify rhe begrnnrng of the climaric detenoration re- ferred to the isotopic stage 4. Considering the cave suc- cession, the geochemical analysis wirh the 2laTh/234IJ method of rhe phosphatic encrustation g,rve :rn.rhsolr:re,rpe of 40.000 +- 2.OOO (Petronio et a1., 1996). in consequence this is the age of the fossil bones coming from level c. For this reason it is possible to refer the faunal assemblage to the isotopic stage 3 (see Gliozzi et al., 1992). Also rhe vertebrate fauna coming from the layers d and e can be referred to the iso- tonic sr.rpe 3. hrrr ir resti[ies the progressive change toward more temperate-cold climatic conditions (reiative abundance of Cervus elaphus, occur- rence of Equus hydruntìnus and Rupicapra sp.)- The medium sized dog, even if rare, coexists with the wolf in layer c, while doesn't occur in layers d and e, where there is the predominance of Cants lupus. A similar tendency seems to be present in the Grotta Ro- manelli succession with the Canis ex gr. "arnensis-mo- sbachensis" more frequent than wolf in "terre rosse" and very rare in the younger "terre brune" with the predo- minance of Canis lupus Qagliacozzo, personal comm.). As a matter of fact, the maintenance of delicate anatomical structures as nasal coanes (Vulpu aulpes, Ca- nis lupus, Felis silaestris from layer d and e) and encepha- 1ic casts (Lynx lynx from layer c), besides some examples of fossil bones in anatomical connection (Ursus arctos from layer d), suggest a 1ow degree of transportation for the fossils and a quite high degree of the sedimentation ratio. Also the study of the lithic implements coming from the top of leve1 e match with both the deposition setting than with the chronological considerations (the Levailois technique is not used after approximately 35- 37.000 B. P.) (Petronio et a1., 1996). The fossils coming from layers c, d and e mat be included in the faunal assemblage Ingarano b (Fig 0)" The fauna coming from g (Ingarano c) is younger than the other faunal assemblages, probably referable to the isotopic srage 2, with cold climatic conditions testi- fied by the predominance of the snow-owl. lhronostratigraphy Absolute ^ge Nfagneto- stratigraphy Mammal Aqes Isothopic scale Sclcctcd localities from Southcrn ltaly HOLOCENE 0.05 0,1 0.1 5 0,2 I c O lr l (t) l'r ì o rlt ; :z F-l t\/. frl F ,l z J l'rl - Crotta Romanelli ("terre brune") Grotta B di Spagnoli (upper levels) Grotta di Cardamone I n g a r a t o Ingarano Ingarano Grotta del Cavallo, Grotta del Sarcofago Grotta B di Spagnoli (lower levels), Grotta Romanelli (K-G), Grotte delle Striare, Melpignano, S. Sidero, Grotta Uluzzo C rTì z frl U F (t) rrì J rrì F J f r'ì .t trì J Grotta Lina Cerveteri, Sedia del Diavolo, Torre in Pietra(upper levels), Vitinia 292 Fio A Palaeoecological notes. From the fossiliferous deposit of Ingarano have been identified three faunal assembiages (a, b, c) refera- ble to the late Aurelian mammal age. The study of the assemblages allow us to stress some considerations about the palaeoenvironmental evo- lution of the area. The fauna wíth Coelodonta aîd Ste- pbanorhinus (Ingarano a) is probabiy to refer to the be- ginning of the climatic deterioration occurred in isoto- pic stage 4. The presence of these two genera of rhinos testifies the existence of plains and open areas. The fauna Ingarano b is characterised by the oc- currence of the modern subspecies of the fallow deer in layer c, which may be referred to warm-temperate clima- tic conditions. The fauna of the layers d and e testify instead the changing of the climate (occurrence of Equus lrydrwntinus and Rwpicapra sp.), with the deve- lopment of more temperate-cold conditions and the re- duction of the forested areas. The coldest phase, related to the last Pleniglacial (isotopic stage 2), is testified by the fauna Ingarano c, characterised by the abundance of the snow-owl Nyctea scandiaca, which today is wide- spread in Northern latitudes and is specialised in preying lemmings. C. Petronio & R. Sardella Revised chronological framework of Ingarano faunal assemblages in comparison with some late Middle and Late Pleistocene selected Iocalities from Southern Italy (from Petronio et a1.7996, modified). Conclusions. The analysis of faunal assemblages Ingarano a, b and c allow to implement the knowledge on the late Aurelian mammal faunas (Gliozzi et al., 1997). The Late Aurelian, is referable to a time span included between the Eemian and the end of the last Glaciation. No Fau- nal Units have been defined because the late Aurelian faunas are characterised by the disappearance of large and medium mammals. In the corresponding time span some climatic events occur. These events, in particular in the Adriatic coast, have a great influence in the fau- nal composition and in the causes of their disappearan- ces. Flowever, another important element to consider is the rule of Homo neandertalensls and Homo sapiens. Fí- nally, the great amount of data referable to late Aurelian faunas if on the one hand gives the possibility to define a paleoecological framework of the microclimatic and environmentai conditions for the different areas of the peninsula, on the other hand do not allow to choose a single fauna to define a Faunal Unit. For this reason the Ingarano faunal assemblages testify the disappearance of the large "pachyderms" in an earlier moment than 35.000 B.P. (Ingarano a), a qui- te progressive climatic deterioration (Ingarano b) corre- sponding to the isotopic Stage 3, toward the cold phase of the isotopic Stage 2 (Recent Wurm) with the disap- pearance of several taxa of mammals and the abundance of cold taxa as Nyctea scandíaca. 293 Acknotaledgernmts. 'Sfe wish to thank prof . M. Gaetani (lvlilano) rnd t-o anonymous reviewers for the useful suggestions, L. Codebò for the field observations, G. Di Stefano, P Ferraro and E. Squazzini for suggestions and technical support. Photographs by L. Spinozzi e G. D'Arpino. This work was supported by MURST 40% grants. Stratigraplry and biochronology of the Late Pleistocene of Ingarano REFERENCES Capasso Barbato L., Cassoli P.F., Minieri M.R., Petronio C., Sardella R. & Scarano M. (1992) - Note preliminari sul- la fauna pleistocenica di Ingarano (Apricena, Foggia). Boll. Soc. Paleont. 1t.,v.31, n. 3, pp" 325-334, Modena. Capasso Barbato L. & Gliozzi E. (1,996) - Biostratigraphical and palaeogeographical implications of the late Middle Pleistocene well balanced fauna from Quisisana-Certosa (Capri, Southern Italy). Boll. Soc. PaLeont. It., v. 34, n. 2, pp. 235-261., Modena. Di Stefano G., Petronio C., Sardella R., Savelloni V. & Squaz- ziti E. (1992) - Nuove segnalazioni di brecce ossifere nella costa fra Castro Marina e Otranto (Lecce). Il Qu* ternario, v. 5, n. 1, pp. 3-10, Roma. Di Stefano G. & Petronio C. (1997) - Origin and evolution of the European fallow deer (Dama, Pleistocene). N. /&. Geol. Palaont. Abh. v.203, n. 1, pp. 57-75, Stuttgart. G|iozziE., Abbazzi L., Argenti P., Azzarolí A., Caloi L., Ca- passo Barbato L., Di Stefano G., Ficcarelli G., Kotsakis T., Masini F., MazzaP., Mezzabotta C., Palombo M.R., Petronio C., Rook L., Sala B., Sardella R., Zanalda E. & Torre D. (1997) - Biochronology of selected mammals, molluscs and ostracods from the Middle Pliocene to the Late Pleistocene in ltaly. The state of the art. Rlz. It. PaL Strat., v. 103, n. 3, pp" 369-388, Milano. Guerin C. (1980) - Les Rhinocéros (Mammalia, Perissodacty- ia) du Miocéne terminal au Pleistocène superiéur en Eu- rope occidentale comparaison avec les espèces actuelles. Docum. Lab. Géol. Lyon, v.79, n. 3, pp. 785-1185, Lyon. Petronio C., Billia E., Capasso Barbato L., Di Stefano G., Mussi M., Parry S.J., Sardella R. & Voltaggio M. (1996) - The Late Pleistocene far.rna of Ingarano (Gargano, Italy): biochronological, paleoecological, paleoethnolo- gical end geochronological implications. Boll. Soc. Pa- leont. It., v. 34, n. 3 (1995), pp. 333-339, Modena. Rook L. & Torre D. (1996) - The latest Villafranchian-early Galerian small wolves in the Mediterranean area. Acta zool. Cracoo., v. 39, pp. 1-7, Cracovia. Verdelin L. (1981) - The evolution of lynxes. Ann. Zool. Fen- nici. v. 18. pp. 37-7 1. Helsinki.