The European Descendants of Ursus Etruscus C. Cuvier (Mammalia, Carnivora, Ursidae)

Home , Cave bear, Cromer Forest Bed, Hemicyoninae, Plionarctos, Ursavus, Ursus (mammal)

Boletin Geologico y Minero. Vol. 103-4. Ano 1992 (632-642)

GEOLOGIA The European descendants of Ursus etruscus C. Cuvier (Mammalia, Carnivora, Ursidae).

Por T. DE TORRES PEREZHIDALGO (')

ABSTRACT

This paper deals with a review of the Pleistocene Bears origin, evolution, and stratigraphical distribution.

U. ruscinensis DEP. could be considered the common ancestor of all the European Pleistocene Bears, as the source of two evolutive lineages: one the more conservative, U. mecliterraneus F. MAJOR, and another the more evolved which starts with U. etruscus G. CUV. and gave origin o two evolutive trends: the today vanished speloid one (U. deningeri v. REICH, and U. spelaeus ROS.-HEIN) and the still living artoid one which is represented today in the true brown bear (U. arctos LIN.).

U. minimus DEV.-BOUILL. does not constitute a link between U. ruscinensis and U. etruscus: it is a lateral branch, in the general evolutionary schedule, more evolved than U. etruscus.

Recent findings of arctoid Ursidae remains in the Iberian Peninsula, ranging from Lower to the Middle (Upper) Pleis tocene, allowed us to think that the general migration southwards of an Asiatic Brown bear population during the Wurm glacial period beginning, superimposed on an authochtonous European Brown bear population (prearctoid): U. prearctos BOULE, which is the probable ancestor of the polar bear [U. maritimus PHIP.).

Key words: Ursidae, Phylogeny, Stratigraphy, Pleistocene, Europe.

R E S U M E N

En este trabajo se revisan el origen, evolucion y distribucion estratigrafica de los osos del Pleistoceno.

IS. ruscinensis DEP. puede considerarse como e! ancestro comun: de el derivaria un grupo muy conservador (U. mediterraneus F. MAJOR)) y otro mas evolucionado U. etruscus G. CUV. que sera el origen de las dos principales lineas evo- lutivas de Ursidos pleistocenos en Europa: la linea espeloide (U. deningeri V. REICH, y U. spelaeus ROS.-HEIN.), muy bien conocida, y la linea arctoide, cuyo representante actual es U. arctos LIN.

U. minimus DEV.-DEBOUILL. no constituye un estadio intermedio entre U. ruscinensis y U. etruscus, debe tratarse de un representante lateral dentro del esquema evolutivo general, mas evolucionado que U. etruscus.

Hallazgos recientes en la Peninsula Iberica, de osos arctoides con edades que van desde el Pleistoceno Inferior al Medio (Superior), permiten pensar que la gran migracion hacia el sur de parte de la poblacion asiatica de oso pardo, que tuvo lugar a comienzos del Wurm, se superpuso a una poblacion europea autoctona de oso pardo (prearctoide): U. prearctos BOULE, de la cual derivo el oso polar (U. maritimus PHIP.).

Palabras clave: Ursidos, Filogenia, Bioestratigrafi'a, Pleistoceno, Europa.

1. FOREWORD cies. To solve this will be the goal of this article. In spite of the fact that there are a very large amount of papers dealing with Pleistocene bears, there remain some gaps in the phylogeny and 2. PHYLOGENY stratigraphical distribution of the different spe- The genus Ursavus could be considered the com mon ancestor of the Ursus genus, the first re- (*) Dpto. de Ingenieria Geologica. Escuela Tecnica Su presentings of which appeared at the Upper perior de Ingenieros de Minas. Madrid. Pliocene.

12 a: IOUGOCENE! MIOCENE PUOCENE PLEISTOCENE r0. I .., ::l ;l: ~ iii C'l bl MINDEL RISS WURM ~ c: ~ 0 ~ ~ ";l: ;l: ~ ~ a 0 /; '"C'l ~ "z ~ LOW I MIDDLE I UPPER C'l " t"" "'" t"'

U rusclnensis - U. minutus ._

Uminlmus _

U. """",,""""'JS _ ••••••••••••••••

\ \ i \ 4-634 T. DE TORRES PEREZHIDALGO

The Ursavi were Carnivora with a typical enough with a heavily built skeleton and bunodont-sec- Ursoid dentition: there is a certain cusp dupli torial cheek-teeth morphology: it is the Deninger city in molars and the lower carnassials show bear, Ursus deningeri VON REICHENAU, which a cutting-like paraconid, while the metaconid is would be superseded by its descenent: the cave very small and placed backward. They were not bear, Ursus spelaeus ROSEN-MULLER-HEINROTH, too different from true bears, but smaller, like which finally vanished at the end of Wurm with the wolverine (Gulo gulo). This genus had a out further descendants. In this species those full Miocene development. morphological characteristics which were poin ted in the Deninger bear appeared exagerated. The first relatively well documented bear spe cies are Ursus ruscinensis DEPERET and Ursus In the Cromer, and until the Riss, there is a minimus DEVEZE-DEBOUILLET. Their first ap- more conservative evolutionary lineage represen peareance was during the Pliocene. ting: U. prearctos BOULE, which retained very close similarities with its ancestor U. etruscus. U. ruscinensis was found in the Ruscinian locali In the WCirm this species would be substituted ty of Perpignan (France).) Very close to this by the still living brown bear: Ursus arctos species, if not the same, are Ursus minutus GER- LINNEO. VAIS from the Pliocene of Montepellier (France); U. wenzensis STACH from the Polish site of Although they will not be studied in this paper Weze, of similar age to Montpellier, and Ursus it is interesting to briefly cite a group of small (Protursus) boecki SCHLOSSER from the Plio sized Pleistocene bears very closely related to cene lignites of Barot-Kopeck (Hungary) which the U. ruscinensis, also known as Plio narctos in FICCARELLi's opinion (1979) is a U. ruscinen group, of which the best known, and probably sis synonymous. the only one representative was U. mediterraneus FORSYTH-MAJOR, in STEHLIN (1933)'s opi Along the Lower ViHafranchian, Villarroya (Spain), nion, which KURTEN and POULI.ANOS (1977) there appeared, for the first time, Ursus etrus- called «european black bear», assuming a nar cus G. CUViER, which reached its highest eco row parallelism with the Thibet bear Ursus thi- logical development in the Middle ViHafranchian: betanus G. CUVIER. Val d'Arno (Italy), Tegelen (The Netherlands), Saint Vallier (France), La Puebla de Valverde The bear of Achenheim (France) Ursus schertzi (Spain) and a certain number of other minor lo DEHM, must be considerated as a U. mediterra- calities. This species disappeared at the Upper neus synonymous, in spite of MOTTL (1951 )'s ViHafranchian, Venta Micena (Spain), and it is opinion that it was an arctoid lineage relation, considered to be the ancest or of the main Euro like an intermediate stage between the brown pean Pleistocene bears. bear of Repolust and U. priscus: «Der Braun bar der Repolust Hole also zwischen Ursus Although it is possible that U. minimus and U. schertzi DEHM und Ursus priscus GOLDFUSS etruscus were related to U. ruscinensis, it seems der Steiermark...». difficult that they are both related, in spite of the fact that U. minimus has been considered Others representatives of this group of small an U. etruscus ancestor species, STEHLIN (1933), sized bears, and probably U. mediterraneus sy THENIUS (1959), KURTEN (1975), TORRES (1984, nonymous too, are: U. (Plionarcstos) Stehlini 1986). BERZI's analysis (1966) of Gaville U. mi KRETZOI from Gombasszog (Hungary) of Riss (?) nimus mandible morphology, and of this author's age, Ursus (Plionarctos) telonensis BON I FAY of own data on unpublished material from Layna, Cimay (France) of Mindel age and Ursus sack- proved that both species were unrelated, dillingensis HELLER from Sackdillinger Hole showing U. minimus mandibular characteristics (Austria)) of doubtfull age (Lower- Middle Pleis more evolved than in U. etruscus; FICCARELLI tocene?). (op. cit.) agrees on an early extintion of the former, but thought that this had more carnivo The Plionarctos genus can not be employed in rous characteristics that U. ruscinensis. European bear description, mainly because it was employed by FRICK (1926-1929) to describe In the Cromer, for the first time, a bear of un- an Ursus (Tremarctos) ornatus F. CUVIER an doubtable speloid characteristics was detected cestor: Ursus (Plionarctos) edensis FRICK, it

14 THE EUROPEAN DESCENDANTS OF URSUS ETRUSCUS C. CUVIER (MAMMALIA, GARNIVORA, URSIDAE) 4-635

was employed later by KRETZOI (1970-1971) as other, more conservative (arctoid)), was com «Plionarctosgroup» to describe some small sized posed of U. prearctos and U. arctos, showing a species that he throught similar, ERDBRINK similar, but quite less marked, tendency; U. ma (1953), to Ursus (Helarctos) malayanus RAFFLES. ritimus can be explained as result of an insulari- zation period effecting a prearctoid population In short: since the Upper Villafranchian there during an advance of the ice cap period. has been a relativelly net species distribution: some were big sized, U. etruscus derivated, A description and discussion of the six species coexisting with others clearly smaller and clo before mentioned will be the central objective sely related to its ancestor U. ruscinensis It of the following. is not necessary to invoque an U. etruscus in- vassion, as was suggested by FICCARELLI (op. cit., p. 1681: "Col Villafranchiano medio-superiore VUrsus etruscus invade larga parte dell 'Eu rasia. ..». This species was already all over 3. DESCRIPTION AND DISCUSSION since the Lower Villafranchian, although its de- OF THE SPECIES mographical boom took place at the moment Ursus etruscus G. CUVIER 1823 indicated by the author. Synonymous: Ursus cultridens NESTI 1826, There is also a co-existence of two evolutive Thalassarctos etruscus AIRAGHI 1922. lineages apart from the residual presence of As stated before, this species appeared at the U. mediterraneus: since the Cromer and until the Riss it was composed by U. deningeri and Lower Villafranchian (Viliarroya), reached its po U. prearctos, since the Riss it was composed of pulation peak at the Middle Villafranchian (Val U. spelaeus and U. prearctos. Later, during the d'Arno, Saint Vallier) and disappeared during the Wurm, U. arctos substituted U. prearctos. The Upper Villafranchian (Venta Micena, Spain). former was an asiatic inmigrant which at the It is anatomically characterised by a notably si end of Pleistocene times colonized the whole of zed skeleton when compared with U. ruscinensis. Europe, North America and the Circunmediterra- In the humerus it is possible to observe a still nean border of Asia and Africa: KOBY (1955), developed entepicondilar foramen, a characteris KURTEN (1969), TORRES (1988). This emigra tic that links this species to Ursavus. It is mo tion took place from a supposed species "ances derately brachipodial, there are small articular tral stock» represented in the Niowan and Chou- facets in carpus and tarsus bones and the meta- K'ou-Tien materials, ZDANSKI (1928) and TEIL- podial bones are slender. HARD DE CHARDIN (1938), although in the opi nion of PEI (1934, 1938) and ERDBRINK (op. cit.) This species usually shows a dental formula the Chou-K'ou-Tien bear was related to the De- identical to Ursavus: ninger bear. An opposed opinion to U. etrus- 1,2,3 1 1,2,3,4 1,2 cus-U. arctus relationship can be seen in ZAPFE I C P M (1946). 1,2,3 1,2,3 1,2,3,4 1,2,3 An European bear population development with There is a single cheek teeth morphology, and net carnivorous affinities could explain the ap- its cusps are strongly convergent towards the peareance of U. maritimus PHIPPS without need inner part of the teeth. In the first upper molar ing an asiatic brown bear ancestral stock, KUR there is a relativelly small transversal develop TEN (1969, p. 25): «The ancestral population ment of the talus. When compared with the probably lay within the Asiatic Brown Bear po anterior lobe of the teeth and the heel of the pulation of this time, perhaps in the vast Sibe second upper molar, it is relativelly short. The rian area». fourth lower premolar is single built with a pro- In short: from an ancestral species, U. etrus toconid rising in the center of the crown. The cus, two different evolutive lineages were deri first lower molar is also very simple and its ved: one, more evolved, changed into the paraconid looks like that of Ursavus, the meta- speloid group: U. deningeri and U. spelaeus. The conid is almost negligible and backward The

15 4-636 T. DE TORRES PEREZHIDALGO

third lower molar crown is almost circular sha and the metaconid has grown and moved towards ped, without lateral troughs. de anterior part of the tooth. There is also a net length augmentation of the third lower mo It seems highly possible that this species would lar, having frequent troughs on both sides of be opportunist as is suggested by its having the tooth as in U. etruscus and is covered by colonizated fluviatile regions (Valdarno, Italy), cusplets. In spite of the appearance of strong marshes (Tegelen, Holland), lacustrine areas speloid characteristics it is not infrequent, in (Venta Micena, Spain), arid sabana-like zones (La large populations data to find 'atavic' cases sho Puebla de Valverde, Spain) and steppes (as sug wing older morphologies. gested by loess deposits from Saint Vallier (France). Becacuse of its high morphological variability German authors wrote about a «deningeri For- menkreis» and the French ones about «groupe Ursus deningeri VON REICHENAU 1906 deningeri-. Some subspecies have also been defined, here they are considered as synony Synonymous: Ursus savini ANDREWS mous because they need a revision: U. denin 1922, Ursus arctos spelaeus ERDBRINK geri savini ANDREWS (Bacton Forest, England), 1953, Ursus deningeri hundsheimensis ZAP- U.d. suevicus KOBY (Jagsthausen, Germany), U. PE 1940, Ursus suszenbornensis SOERGEL d. hundsheimensis ZAPFE (Hundsheim, Germany), 1912, Ursus gombaszogensis KRETZOI 1941, U. d. suszenbornensis SOERGEL (Siissenborn, Ursus etruscus gombaszogensis KRETZOI Germany). U.d. romeviensis PRAT (La Romieu, 1945, Ursus deningeri suevicus KOBY 1951, France) appears as the only recently described Ursus arctos (Hundsheim) FREUDENBERG, U. deningeri subspecies and represents a late Ursus taubachensis RODE. ral branch of the normal species. This species This species marks a dramatic change when appeared along the Cromer. KURTEN (1959), di compared with its ancestor U. etruscus: the sappearing at the end of the Riss, BONIFAY skeleton became more heavily built, mainly at (1971). In PRAT and THIBAULT's (1975) opinion the bone epiphysis, while the diaphysis are still they disappeared in La Romieu Cave at the Riss slender. There is a net length decrease in first I while in the Riss II sediments there are re metacarpals and metatarsals and an enlarge mains of a spe/aeus-like bear. ment of the area of articular facets of tarsal Although there are no doubts about the role pla and carpal bones articular facetsts, which in so yed by this species as the cave bear ancestor, me cases result merged. it seems probable that there was a development The dental formula is simpler than in U. etrus of lateral groups not directly linked with U. spe cus: laeus, this is KOBY's (1951, p. 403) opinion about U. deningeri suevicus: «Cette sous spece ne 1,2,3 1 (0,(3),4 1,2 parait pas etre I'ancetre inmediat de I'ours des I C — P M cavernes mais plutot se trover sur un rameau 1,2.3 1 (0,(3),4 1,2,3 lateral...».

Thats is: the second premolars are always ab Although usually this species remains have been sent, but the first and/or the third can randomly found in caves (that is: related to karst on cal- appear. careus rocks) it is interesting to note that in There is an important change in the dental mor the oldest sites they are linked to open air envi phology too and the old sectorial structure is ronments, fluvial related in some cases, as Bac changed into a very marked bunodontism. The ton Forest Bed (England). Cromer Forest Bed first upper molar talus is wider than the trigon (England), Hundsheim (Germany), Mauer (Ger and there is a relative elongation on the second many) and Siissenborn (Germany). upper molar heel, where narrow forms dominate. In the paraconid-protoconid region of the fourth Ursus spelaeus ROSENMULLER-HElNROTH 1794 lower premolar there appear many cusplets and in the first lower molar the paraconid shows a Synonymous: Ursus arctoideus BLUMEN- morphology far enough from those of Ursavus, BACH 1799, Ursus pitorri DE SERRES 1930,

16 THE EUROPEAN DESCENDANTS OF URSUS ETRUSCUS C. CUVIER (MAMMALIA, CARNIVORA. URSIDAE) 4-637

Ursus metoposcainus DE SERRES 1330, Ur bear, so they only have a certain systematic sig- sus neschersis CROIZET 1839, Ursus spe nifiance U. spelaeus var. hercynica RODE., also laeus major SCHMERLING 1833, Ursus gi- known as, «Alpinenkleinenformen», the steppa ganteus SCHMERLING 1833, Ursus fornica- bear from Krasnodar (URSS) and a strange vas- tus maior SCHMERLING 1833, Ursus for- que subspecies U. s. parvilatipedis TORRES 1991, nicatus minor SCHMERLING 1833, Ursus characterized by extremelly short and broad leodiensis SCHMERLING 1833, Ursus pla paws. It is necessary to take into account the nus OKEN, Ursus spelaeus odessanus VON enormous intersexual variability of this especies, NORDMANN 1858, Ursus spelaeus raza mi which appeared in a more or less marked way nor GAUDRY y BOULE 1892, Ursus ligus- in the skeleton and dentition. There is a very ticus ISSEL 1885, Ursus spelaeus var. li- high intraspecific variability too, of which a ty gustica ISSEL 1890, Ursus spelaeus var. si- pical example was composed by a cave bear byllina FRAAS 1899, Ursus dentifricius ME skull from Gaylenreunth employed by GOLDFUSS YER, Ursus ferreo-jurassicus JAGER, Ursus (1823) to describe a new species: U. arctoi- prespeleus OSBORN, Ursus Gaudryi FIL- deus, and is only a dolicocephalous extreme of HOL, Ursus spelaeus var hercynica RODE, a population with more normal brachicephalous Ursus spelaeus forma nonata EHEREMBERG representings, CORDY (1972). During Ontogeny 1935, Spelaerctos spelaeus BORSSIAK 1931. there are also a lot of astonishing changes in skull morphology, TORRES (op. cit.). There is a very important reduction in the num ber of premolars in this species, its dental for The cave bear appeared at the end of Riss times, mula being as follows: and vanished without further descendents at the end of the Wurm III or a bit later, ALTUNA (1984). 1, 2,3 1 4 1,2 I C — P — M In spite of the fact that this species was appa 1,2,3 1 4 1,2,3 rently linked to karstic zones, it also colonized fluviatile — swampy zones, like the Brown Ridge This reduction in the number of teeth was ac soal in the North Sea where fishermen trap (in companied by an enormous augmentation in the their nets) cave bear bones. It has been found size of the premolars and molars, whose cusps also in steppe zones and in high mountain re are now more complicated and vertical. There gions over 2.500 m. is noticeable metrical growth in the total length of the fourth upper premolar, the first upper mo lar heel width, the second upper molar talus, Ursus prearctos BOULE 1919 length, the width of the fourth lower premolar, Synonymous: Ursus priscus (pro parte), the talonid width in the first and second lower Ursus arctos (pro parte) Ursus praearctos molars and a general size augmentation of the MIR y SALAS 1976. third lower molar. At the same time there is a This species has many affinities with its imme shortening in the paraconid- protoconid distance diate ancestor U. etruscus. It has a noticeably (it marks the cutting portion of the teeth) in the marked dental conservatism in the oldest mate firts lower molar. rial where the second premolar still remains. There are important changes in the skeleton too: This is usually absent in more modern material, an augmentation of the skull volume (in spite where some of the other two remaining first of intense pneumatization), the mandible ramus premolars (first or third) could also be absent: ascendentis became vertical and a general 1,2,3 1 1,(2),3,4 1,2 growth in the relative transversal measurements I C — P M of postcraneal bones (both diaphisys and epiphy 1,2,3 1 1,(2),3,4 1,2,3 sis). In carpus and tarsus bones there is some growth and a frequent fusion of articular facets. There is a certain, not profond, change in molar The first metacarpal and metatarsal became re and premolar morphology, when compared with lativelly shorter than in U. deningeri. that of U. etruscus: the second upper molar heel became longer and was full of cusplets, in U. This species seemed to have more morphologi etruscus it was ridulae draped. At the lower cal uniformity than its ancestor the Deninger molars there is a limited duplication of some

17 2 4-638 T. DE TORRES PEREZHIDALGO cusps that become slightly verticalized, in rela Valle (Madrid), cited by ALFEREZ et. al. (1982), tion with the convergency towards the antero first described by TORRES (1984), later ALFE posterior axis of the teeth that appeared in U. REZ et. al. (1985) published a skull from this site, etruscus, but are still convergent when compared of Uppermost Riss or Eem age, as Ursus sp. with U. arctos ones. In the lower carnassial it Outside of Spain U. prearctos has been found is possible, in some cases, to observe Ursavus in Mont Maurin (France), De SAINT PERIER —like morphologies, which are absent in the (1922), it is similar in age to the one from Gran majority of the other cases. Dolina; in the Grotte du Prince, BOULE (1906), In the postcraneal skeleton theere is a visible BONIFAY (op. cit.) a bit older than Pinilla del augmentation of robustness. Valle. PRAT and THIBAULT (op. cit.)) found in old levels an archaic brown bear p. 39: «...nous This species creates some problems, in ERD- sommes en presence d'une forme arctoi'de pri BRINK (op. cit.): «!n fact BOULES's theorv that mitive quil faut peut etre rapprocher d'Ursus pre this small bear, which he described as U. prearc- arctos M. BOULE... un Ours brun vivant deja en tos, forms the link between the real U. etruscus Gers durant I'avant dernier interglaciar...». and the Pleistocene Brown bear, U. arctos (with many sinonimia) does not seem improbable to The time span covered by these finds allow us me..... This species was first described by to doubt BONIFAY's (1971) suggestion of a post- BOULE (1906) from the Grotte du Prince (Italian Villafranchian stratigraphical hiatus lasting until Liguria) material from bones and teeth as having the Riss glacial when prearctoid bears appeared an etruscus-Wke aspect, and being small sized. in southern Europa that would be substituted by The author also found some remains of a bigger the true brown bear (U. arctos LIN) in the last brown bear that was thought to be from U. pris glacial. cus. Later BONIFAY (1965) ratified this species but also thought that there weer two coeval This species has only been found in karstic fil brown bears of different sized species, pp. 69-70: lings. «ll semble qu'a la Grotte du Prince nous soyons en presence de deux 'speces': I'une Tours brun typique qui est la mieux representee; I'autre un Ursus maritimus PHIPPS 1774 petit urside que ne parait pas etre la forme de Synonymous: Ursus marinus PALLAS 1776, transition erninenment variable decrite par M. Ursus albus MULLER 1776, Thalassarctos Boule». TORRES (1984) found that all this ma maritimus ERXLEBEN, Thalassarctos eogro- terial, both big and small sized, were from a enlandensis KONOTTNERUS-MEYER 1903, unique bear population. Size differences can be T. labradorensis K.-M., T. spitzbergensis explained as sexual dimorphism reflect, with me K.-M., T. jenaensis K.-M., 7". maritimus var. trical relationships quite different from those of ungavensis K.-M. Thalassarctos maritimus the recent brown bear population of the Iberian groenlandicus BIRULA 1932, T.m. marinus Peninsula. RENAULT-MISKOVSKY (1986) esti BIRULA 1932. mates an age of 85000-80000 years for the bony breccia underlying to tyrrenian age marine de The dental formula of this species is: posits. 1,2,3 1 1,3,4 1,2 The oldest iberian material came from Gran Do- I C — P M lina, Atapuerca (Burgos). This paleontological lo 1,2,3 1 1, 4 1,2,3 cality was placed by the author in the Giinz, TO The first premolar is absent in most of the cases. RRES (op. cit.). More recent datings placed it in the lower Cromer or a bit older, GIL et. al. As is normal in a species with carnivorous ha (1987). bits an important amount of «primitive» charac teristics remain in its dentition: small sized There are some remains of Mindel age from Mo- teeth, cusps with cutting edges, slightly conver llet Raco (Banyoles, Girona) first described by gent towards the anteroposterior axis of molars, MIR and SALAS (1976). and an uncomplicated morphology. The small The most recent material comes from Pinilla del development of the second upper molar heel is

18 THE EUROPEAN DESCENDANTS OF URSUS ETRUSCUS C CUVIER (MAMMALIA, CARNIVORA, U RSI DAE) 4-639

noticeable and, consequently, that of the third silis (GOLDFUSS 1821) LYDEKKER 1897, lower molar, subcircular shaped and with scarce Ursus libycus POMEL 1897, Ursus syriacus inner ridulae. The skeleton is quite similar to (HEMPR & EHEREMBERG 1828), ZUMOFEN that of the brown bear, but more heavily built. 1900 (apud BATE 1927), Ursus arctos anti- In spite of its marine habits, its scapula does qui POHLIG 1909, Ursus procerus HAY not differ from that of the brown bear, KOBY 1911 (1912?) aut. MILLER 1899 (teste HAY (1955). 1923, FREUDENBERG 1918), Ursus arctos var. priscus (GOLDFUSS 1822) FREUDEN In spite of KURTEN's (1964, p. 23) opinion about BERG 1914, Ursus arctos var. priscus (CU the ancient origin of some characteristics of VIER 1823) BOULE 1919, Ursus anglicus this species, maybe as long ago as during the GUNTHER 1923, Ursus arctos subs, nucifra- Cromer-Mindel: «...thus a maritimus like pattern gus LONNBERG 1923, Ursus arctos raza li seems to date back again to approximately the bycus (POMEL (1897) DEPERET 1928, Ursus Cromer-Mindel stage-, there is not any fos arctos mut. faidherbi BOURGUIGNAT (1867) sil remain until Ursus maritimus tyrannus ARAMBOURG 1932, Ursus arctos mut. lar- KURTEN 1964, from the Lower Wurm and some teti (BOURGUIGNAT 1868) ARAMBOURG of those maritimus —like characteristics, main 1932, Ursus arctos nemoralis DEGERBOL ly those of the dental morphology, can be re 1933, Ursus arctus priscus (GOLDFUSS cognized, in U. prearctos material, as in 1822), DEGERBOL 1933, Ursus arctos fos some second upper molar (unfortunatelly bro silis (GOLDFUSS 1821) KOBY 1944. ken in part) from Mollet Raco, which have short and narrow heels. This list does not include the enormous amount of synonymia defined from living material, be This species is restricted today to the Arctic cause this would make it too long. An example circle neighborhood. But at the maximum extent can be found in ERDBRINK (op. cit.) who gave of the ice cap during the vViirm, it probably could a list of two hundred and thirty two «species» have reached the northern coast of the Iberian created by MERRIAM (1918 non vidi) from the Peninsula, but its special habitat could make the present North American brown bear material. preservation of bones and teeth difficult. An implicit recognition of the unusfullness of these species proliferation was in COUTURIER (1953, p. 326): «cette partie du squelette (the Ursus arctos LINNEO 1758 skull) revet une importance particuliere, car elle Synonymous of fossil and sub-fossil species seule souvent constitue I'unique materiel pour from Europa and circunmediterranean border. decrire une espece ou une sous - spece. Je pense que MERRIAM aurait cree 17 especes Ursus fossilis GOLDFUSS 1821, Ursus pris avec les 17 tetes osseuses d'Ours des Pyrenees cus GOLDFUSS 1822 (1810?), Ursus arctoide ma collection, car toutes sont differentes...». deus DE SERRES, DUBREUIL & JEANJEAN 1829, Ursus arctos subfossilis von MIDDEN- The dental formula of the brown bear is as fo DORFF 1851, Ursus planifrons DENNY 1864, llows, TORRES (1984): Ursus faidherbianus BOURGUIGNAT 1867, 1,2,3 1 (0,(3),4 1,2 Ursus lartetianus BOURGUIGNAT 1868, Ur I C — P M sus letourneuxianus BOURGUIGNAT 1868, 1,2,3 1 (1), 4 1,2,3 Ursus bourguignati LARTET 1867, Ursus rouvieri BOURGUIGNAT 1868, Ursus pome- In spite of certain similarities with U. etruscus lianus BOURGUIGNAT 1868, Ursus tarandi and U. prearctos morphologies, it shows its own FRAAS 1872 (teste PORTIS 1878), Ursus evolved characteristics: cusp are usally dupli ferox fossilis (GOLDFUSS 1821) BUSK 1873, cated, having lost their cutting appeareance. The Ursus horribilis fossilis (GOLDFUSS 1821) enamel was covered by smail cusplets, while in LYDEKKER 1885, Ursus arctos raza priscus the other two species it was ridulae draped. (CUVIER 1823) GAUDRY & BOULE 1892, The cusp are almost vertical, having lost a net Ursus arctos var. isabellinus (HORSFIEL obliquity, mainly in their lingual walls, which was 1827) von FRITSCH 1893, Ursus arctus fos tipycal in the older species. The relative length

19 4.640 T- DE TORRES PEREZHIDALGO of the second upper molar grows significativelly, these formulae or indices were intended for use and the paraconid in the first lower molar shows as determining factors...". It is evident that it a «modern- habit. The «Ursavoid» morpholo is an extreme opinion which can be discarded gies which were present in all U. etruscus car- by regression analysis. nassials and in a certain number of specimens of U. prearctos are absent. There is a rather important augmentation of the third lower molar length. 4. ISSUES As in the speloid evolutive lineage, there is a — U. ruscinensis must be considered as the similar evolution in the postcraneal skeleton of arctoid lineage representings. There is a growth common ancestor of all European Pleistocene of transversal measurements of long bones epi bears. From it, two evolutive lineages are physis, and in the extent of the arcticular facets derivated: one, more conservative, represen of carpal and tarsal bones. There is not a en- ted by U. mediterraneus. had small teeth and tepicondilar foramen in the humerus, but a small a small-sized skeleton and had limited ecolo apophysis still remains, a vestige of a bony gical success. The other one, more evolved, bridge that defined the outer limit of the fora was represented by U. etruscus and showed men. an enormous size augmentation and ecologi cal success. It seems quite possible that in the Wurm begin ning a massive migration took place, due to an — U. minimus seems to be linked to U. rusci enormous enlargement of ice covered areas. U. nensis but not to U. etruscus, because of its arctos covered the whole of Europe, North Ame more evolved characteristics. It could be ta rica (Rancho La Brea, California) and the circun- ken as a first attempt of the general hipocar- mediterranean border of Africa and Asia. In nivorous tendency of all Pleistocene beers. North America giant forms appeared in favoura ble environments: U.a. qyas and U.a. midden- •— U. etruscus appears as the ancestor of two dorffi, that show a certain morphological con big sized evolutive lineages of Pleistocene vergence with the cave bear in its skul! morpho bears: speloid and arctoid. logy. — The speloid lineage was composed by U. de ningeri and U. spelaeus; both species have The name of U. priscus has frequently been used well defined metrical and morphological dif to describe fossil and subfossil brown bear re ferential characteristics and net geographical mains. In fact MUSIL (1985) still employes it, and stratigraphical distributions. Both had but today there are not too many reasons to con an enormous ecological success. tinue its use, because the Gaylenreuth skull, on which Goldfuss based it, is from a normal brown — The whole of the arctoid lineage is less bear, DE BLAINVILLE (1839) and GAUDRY (1867) known because of the lesser frequency of were already of this opinion and GAUDRY (in finds. It is assumed a massive arrival of KOBY 1944) wrote: «j'ai constatee que Tours U. arctos at the Wurm beginning being co gris differe plus que I'ours brun d'U. priscus, car lonized North America, the whole of Europa notre squelette d'ours gris est plus massif que and the Mediterranean Border of Asia and celui de I'ours brun et son humerus se distingue Africa. Continuous finds of etruscoid-arctoid par une plus forte saillie de I'epicondyle... U. like bears, covering the whole of middle Pleis- priscus paraitre etre simplement un U. arctos toce times, support the hypothesis of a con de grande taille...». This fact becomes more tinuous presence of an arctoid— like bear evident when the enormous intraspecific variabi all through the Pleistocene: U. preactos. A lity of this species is taken into account, of highly adaptative potential in different eco which ERDBRINK (op. cit.) stated (p. 384): «ln logical systems can be assumed for it, view of the asthonishing variability in size of teeth of U. arctos it may be clear that the value — A continuous (in time), european etruscoid- of formulae or indices based on combinations of arctoid bear population permits the U. mari measuremens of teeth is almost negligible if timus origin to be placed in the European

20 THE EUROPEAN DESCENDANTS OF URSUS ETRUSCUS C. CUVIER (MAMMALIA. CARNIVORA. URSIDAE) 4-641

arctoid stock, without an asiatic ancestor MAMMALIA) en Cuevas de! Pais Vasco. MUNIBE, XXV, being necessary. 2-4:121-170. BERZI, A. (1966): L'orsi de Gaville nei Valdarno superiore. — It is evident that a continuity of an etruscoid- Pal. Ital., LX [n. ser. XXX):1-54. arctoid stock could cause taxonomic pro BLAINVILLE DE M. (1839-1844): Osteographie ou descrip blems, because of evolutive convergence, tion iconographique comparee du squelette et du systeme there could appear remains of a «relatively dentaire des maminiferes recents et fossiles pour servir modern- bear representing in fact indistin- de base a la zoologie et a la geologie, t. II.

gishable from that of an ancient one of the BONIFAY. M. F. (1965): Sur la valeur specifique de I'Ursus speloid lineage, and still very closely, related prearctos BOULE de la Grotte du Prince. Bull. Mus. Anth.- to the common ancestor U. etruscus. Prehist. Monaco, V-9:65-72.

BONIFAY, M. F. (1971): Carnivores Ouaternaires du SE — The cronological interval covered by these de la France. Mem. Mus. Hist. Nat. Nouv. ser. C, XX!:44- etrusco-arctoid bears does not seem too big 377. when compared with that of the U. etruscus BOULE, M. (1906): Les grottes de Grimaldi (Baousse-Rous- (from Villarroya to Venta Micena), and taking se), t. I, fasc. IV. into account its high morphological and me CORDY, J. M. (1972): Etude cle la variability des cranes cle trical conservadurism and its very probable la collection Schmerling. Ann. Pal., LVIII, fasc. 11:151-207. ecological opportunism. COUTURIER, J. (1953): i'Ours brun. Grenoble. — It remains a certain nomenclatural uncertain ERDBRINK, D. P. (1953): A review of fossil and recent ty: U. priscus must be taken as nomen cle- bears of the Old World. 597 pp., Jan. de Lange Eds.

lenda. because the original material was from FICCARELLI, G. (1979): Osservazione sull' evoluzione del an animal indistingishible from the modern genere Ursus. Boll. Soc. Pal. Ital., v. 18, n. 2: 166-172. U. arctos. U. prearctos could be an adecuate FRIANT, M. (1975): L'ours des cavernes Ursus spelaeus denomination since it was established from ROS. Principaux characteres anatomiques de sa mandibule. true fossil material with more modern cha Sond. Mitt. Nat. Ges. in Wien. N. folge. 18Bd:27-36. racteristics than U. etruscus but more ar FRICK, Childs (1926-1929): The Hemicyoninae ana an Ame chaic features than U. arctos. rican Tertiary Bear. Bull. Am. Mus. Nat. Hist., LVI: 1 -119. GAUDRY, A. (1867): Le petit Ursus spelaeus de Gargas. C. R. Sean. Acad. Sci., CIV:740-744. GIL, E.; AGUIRRE, E., and HOYOS. M. (1987): Con'exto es- AKNOWLEDGEMENTS tratigrafico. En: AGUIRRE, E.: CARBONELL, E., y BERMU- DEZ DE CASTRO, J. M.: El hombre fosil de Ibeas y el The author would like to acnowledge to Dr. J. Pleistoceno cle la Sierra de Atapuerca, pp. 47-55. Conseje- Morales from Museo Nacional de Ciencias Na- ria de Cultura y Bienestar Social. Junta de Comunidades de Castilla-Ledn. turales (CSIC) who made a number of valuable suggestions, to Dr. F. Ayala from the Instituto GOLDFUSS, A. (1823): Osteologische Beitrage Kenntniss verschiedener Saugetiere der Vorwelt (Fotsetzung). Nov. Tecnologico y Geominero de Espaha who suppor Act. Akad. Ces. Leopoldino-Carolinae, Nat. Curiosorum, I, ted this paper printing and to Mrs. M. Walsh P-ll-451-490 (non vidi). who reviewed the author's english manuscript. KOBY, F. (Ed.) (1944): Un squelette d'ours brun du pleis tocene italien. Verhand. Nat. Gesell. in Basel, Bd. L.VI:58-85.

KOBY, F. (Ed.) (1951): Un nouveau gisement a Ursus de ningeri von REICH. Ecclog. Geol. Helvet., v. 44, n. 2:398-403. REFERENCES KRETZOI, M. (1970-1971): Kritische b.imerkugen zur Abstam- ALFEREZ F.: MOLERO, G.; MALDOMADO, E.: BUSTOS. mung der Ursiden. Vert. Hung., 12:123-131. V.; BREA, P., and B'JITRAGO, A. M. (1982): Descubrimien- to del primer yacimiento cuaternario (Riss-WOrm) cle ver- KURTEN, B. (1959): On the bears of the Holstenian inter- tebrados con restos humanos en la provincia de Madrid. glaciar. Acta Univ. Stockholm Cont. Geol., 11:73-102. COL-PA. 37:15 32. KURTEN, B. (1964): The evolution of the Polar Bear (Ur ALFEREZ. F.; MOLERO. G. and MALDONADO, E. (1985): fs- sus maritimus PHIPPS). Acta Zool. Fenn.. 108:1-30. tudio preliminar del Ursido del yacimiento del Cuaterna rio Medio de Pinilla del Valle (Madrid). COL-PA, 40:59-65. KURTEN, B. (1969): Transberingian relationships of Ursus arctos LINNE (Brown and grizzy bars]. Soc. Csi. Fenn. ALTUNA, J. (1984): Hallazgos cle oso pardo (Ursus arctos Comm. Biol., 65, 7:1-10.

21 2-1 4-642 T. DE TORRES PEREZHIDALGO

KURTEN, B. (1975): The cave bear story. Life and deadth STEHLIN. H. G. (1933): In: DUBOIS & STEHLIN: La Grotte of a vanished animal, 163 pp. de Cotencher station musterienne. Mem. Soc. Pal. Suisse, vol. LII-Llll:39-64. KURTEN, B. & POULIANOS, A. (1977): New stratigraphical and faunal material from Petralona Cave with special re THENIUS, E. (1959): Ursiclenphylogenese und Biostratigra- ference to the Carnivora. Anthropos, 4:47-130. phie. Sond. Zits. Saugetierkunde, Bd. 24:78-84.

MERRIAM, CH. (1918): A review of the Grizzly and Brown TEILHARD DE CHARDIN, P. (1938): The fossils from lo Bears of N. America (genus Ursus) with description of a cality 12 of Choukoutien. Pal. Sin. New ser. C, n. 5, Ursi- new genus Vetularctos. U. S. Agr. Dep. Bur. Biol. Surv. dae:169-177. North Am. Fauna, n. 41:136 pp., non vidi. TORRES, T.; COBO, R., y SALAZAR, A. (1991): La poblacion MIR, A., and SALAS, R. (1976): Tres nuevos carnivoros del de oso de las cavernas (Ursus spelaeus parvilatipedis n. yacimiento cuaternario de la Cova d'en Mollet I Serinya ssp.) de Trosicaetalico - Icosea (Ataun - Guipuzcoa) (Cam- (prov. Gerona). Inst. Inv. Geol. Univ. Barcelona, XXXL97- pahas de excavacion de 1987 y 1988). MUNEBE, 43:3-85. 124. TORRES, T. (1978): Estudio comparativo de las mandibulas MOTTL (1951): Die arctoide und speloiden Merkmale der de Ursus spelaeus ROSENMULLER-HEINROTH, Ursus de Baren. Foldtany Kolzlony, 63, Budapest, in ERDBRINK (1953). ningeri Von REICHENAU y Ursus arctos LINNEO. Bol. Geol. y Min„ LXXXIX-lll:203-222. MUSIL, R. (1986): Paleohiography of terrestrial communi ties in Europa during the last Glacial. Acta Mus. Nat. Pra- TORRES, T. (1984): Ursidos del Pleistoceno-Holoceno de gae, XCI, 1-2:80pp. la Peninsula Iberica, T. D. Escuela Tecnica Superior de In- genieros de Minas de Madrid, 653 pp. PEI, W. C. (1934): On the carnivora from locality 1 of Choukoutien. Pal. Sin., ser. C, v. VIII, fasc. 1, Ursidae: 501- TORRES, T. (1986): Spanish karslic fillings. The key for 505. Pleistocene Ursids knowledge. IX Int. Congr. Spel. Barce lona (25:215-217. PEI, W. C. (1939): An attempted correlation of Quater nary Geology, Paleontology and Prehistory in Europe and TORRES, T. (1988): Osos (Mammalia, Carnivora, Ursidae) China. Occ. pap., n. 2, Univ. of London, Itt. of Arch., 16 pp. del Pleistoceno Iberico. I. Filogenia, distribucidn estratigra- (non vidi), in ERDBRINK (1953). fica, distribucidn geografica, estudio anatornico y metrico del craneo. Bol. Geol. y Min., t. XCIX, fasc. 1:3-46. PRAT, F. & THIBAULT, CI. (1975): Le gisement de La Nau- terie a La Romieu (Gers.). Fouilles de 1967 a 1973. Mem. TORRES, T. (1988): Evolucion de la carnicera inferior en Mus. Nat. Hist. Nat. Paris, t. XXXV, 82 pp. los generos Ursavus y Ursus. Paleontologfa y Evolucion (en prensa). RENAULT-MISKOVSKY, J. (1986): Relations entre les spec tres archeopollinigues clu sud-est de la France et les os ZAPFE, H. (1946): Die teufels-oder Fuschenlucken bei Eg- cillations climatigues entre 125000 ans et le maximum gla- genburg NO II die Holenbarenreste. Ost. Akad. Wiss. Math.- ciaire. Bull. Ass. Fr. Et. Quat. 2eme. ser., 25-26 (1-2):151- Nat. Klasse. Bd. 114:14-23. 157, ZDANSKY, O. (1920): Die saugetiere fauna cler Quarta- SAINT PERIER, R. DE (1922): Nouvelles recherches dans la fauna von Chou-k' ou-Tien. Pal. Sinica, Ser. C, vol. 5, caverne de Mont Maurin. L'Anthropologie, XXXil:193-202. fasc. 4:3-146.

Original recibido: Enero de 1992. Original aceptado: Febrero de 1992.