Pliocene mammals and climatic reconstruction in the Western Mediterranean area more

The Pliocene: Time of Change Including Papers From The Symposium Entitled "The Palynology, Climate And Sequence Stratigraphy Of The Pliocene" (Held during the 26th Annual Meeting of the American Association of Stratigraphic Palynologists, Inc., Baton Rouge, Louisiana ,1993) Edited by Iof,n H. Centre for Palaeoecology Center For Excellence in Palynology Centre de Pal6ontologie Department of Geology & Geophysics stratigraphique et Pal6o6cologie Queen's University of Belfast (UMR 5565 CNRS) Belfast BT7 1NN Louisiana State University Northem Ireland Baton Rouge, LA 70803 Universitd Claude Bemard-Lyon 1 Wrenn Jean-Pierre Suc Suzanne A.G. Leroy US.A. Z7-43,boulevard du ll Novembre 622 Y rLleurbalne C6dex France 69 American Association of Stratigraphic Palynologists Foundation 1999 PLIOCENE MAMMALS AND CLIMATIC RECONSTRUCTION IN THE WESTERN MEDITERRANEAN AREA ]ean-Pierre Aguilar, Serge Legendre, jacques Michaux and Sophie Montuire Abstract The specffic diversity of extant mammalian faunas, the bodyweight distribution of component species (as expressed by cenograms), and the specific richness of some taxonomic groups change with the climate. Using these parameters, we attempt to estimate the climatic conditions of earlier times and to estimate past temperatures and precipitation. Cenogramanalysis shows that there is a contrastbetween rather closed and warm environments at the beginning of chute de la diversit6 des Murinae survient en France conune en Espagne, au m€me moment. Elle est interpr6t6e comme le signe d'une d6t6rioration du climat vers le froid et l'aridit6 entre1,2et2,0 Ma. Vers -1,0 Ma environ,leclimat 6taitplus sec, avec des estimations de l'ordre de L3"C et 700 mm. Ainsi, les premidres phases glaciaires sont caract6ris6es par des temp6ratures assez 6lev6es. INTRODUCTION Paleontologists have always used manunalian remains to reconstruct the climates of the past. At the end of the 18th century, Buffon (in Roger, 1988) used elephantbones found in Siberia, to infer that the past climate of the Earth was warmer than today. Later, the remains of Pliocene rhinoceroses and tapirs, excavated in the vicinity of Montpellier (France), were likewise taken to show that the climate of the past was warmer (Dep6ret, 1890). Not only large mammals, but small mammals also have been used for paleoclimatic reconstruction, especially by Central European workers (e.g., Kowalski, 1971.; Kretzoi, 1962; 1969). These findings have been backed by Western European studies (Aguilar and Michaux, 1984; 1990; Chaline , t972; Daams et al., 1988; Daams and van der Pliocene times, and more open and cooler environments during the Pleistocene, and also that the Iberian Peninsula, as compared to southern France, was characterized by more open environments. The specific diversity of Murinae in Western Europe shows that the climate was warm with high precipitation (mean annual temperature = 21oC and mean annual rainfall = ca. 1 100 mm) at the end of the Lower Pliocene (-4.0 to -3.5 My). The drop in diversity in Murinae that can be seen in Spain and in France is probably related to a deterioration of the climatetowards coolness and drlmessbetween-3.2 My and -2.0 My. About 1 My ago, the climate became dryer, with estimates of about 13'C and 700 mm. The first glacial phases are thus characterized by rather mild temperatures. R6sum6 Dans la nature actuelle, la diversit6 sp6cifique des faunes, la distribution des espdces en fonction de leur poids (exprim6e par des courbes appeldes c6nogrammes), et la richesse spdcifique de certains groupes taxonomiques, varient avec le climat. Dans le cas de faunes de l'ancien monde, le groupe utilis6 sera la sous-famille des Murinae. Sa diversit6qui se modifie nettement avec les tempdratures et les pr6cipitations, sera utilisde pour un essai de quantification de ces paramEtres. Les cdnogrammes montrent l'opposition entre les environnements plut6tferm6s et chauds du d6but du Pliocbne, et ceu; plus ouverts et moins chauds du d6but du Pl6istocdne. De plus, la p6ninsule ib6rique se caract6rise par des environnements toujours plus ouverts que ceux du Sud de la France. Des changements affectent la diversit6 des Murinae en_Europe occidentale. Ils indiquent qu'd la fin du Pliocdne les pr6cipitations 6lev6es (temp6raturemoyenne annuelle = 2LoC et moyenne annuelle des pr6cipitations = 1100 mm). Une Aguilar, J.-P., Legendre, S., Michaux, J. and Montuire, S. 1999 Pliocene mammals and climactic reconstruction in the Western Mediterranean area; In: Wrenn, f.H., Suc, J.-p. and Leroy, S.A.G. (eds), The Pliocene: Time of Change; American Association of Stratigraphic Palynologists @ 1999 inf6rieur (-4,0 a -3,5 Ma) le climat 6tait chaud et Foundation, p. 109-120. AASP Foundation ISBN 9-931871-04-2 Meulen, 1984; Mein, 19U; van der Meulen and de Bruijn, 1982;van der Meulen and Daams, 1992). The most common remains of extinctmammalian species are the teeth, although in some few cases we have also the limb bones. Both teeth and bones may show ecomorphological features, which can then be used to assess life traits. For example, in the case of fossil species of Myoxidae, their diets can be inferred from their teeth because their morphology is similar to those of extant species, whose diets are known (van der Meulen and de Bruijn, 1982). Extrapolating from what we know of an extant species to its extinct relatives may raise difficulties, because the present habitat of a species is a function not only of its adaptive requirements but also of its biogeographical history. These factors - adaptation and biogeography - may be very different today from those of former times, which means that one cannot simply extrapolate from the present climate of a species to those of its older relatives. For instance, the relative abundance of individuals in fossil assemblages, a phenomenon which has long being used for climatic reconstruction (e.9., Kretzoi, L962), may reflect fossilization processes and accumulation conditions rather than the species abundance in past communities. It is because of such difficulties, that workers have tried to avoid the taxonomic assessment of individuals and the trouble due to taphonomic biases, and why they have instead relied upon the specific richness of the faunas as well as upon the specific diversity of component groups. It is for this reason that in studying Pliocene and Pleistocene climates in Europe, people have analyzed community structures and species THE PLIOCENE: TIME OF CHANGE Montuire, 1995). past climates. diversities (e.g., Chaline and Brochet, L989; Horiicek, 1990; It is these two approaches, community analysis and specific diversity, that we will use to determine We have obtained the cenograms of more than 200 rnod- ern faunas from all continents (Africa, Eurasia, the New World, and Australia) (Legendre, 1989, Legendre, Badgley, and Gingerich, in prep.; Text-Figure 2). Comparison of these cenograms shows that they are broadly similar for faunas from similar habitats and differ markedly for faunas from different habitats. In fact, cenograrns can be distributed into METHODS USED FOR CLIMATIC ANALYSIS Mammalian Paleocommunities Because size is correlated to physiology, life history, and ecology (Calder, 1984; Eisenberg, 198L; Peters, 1983; fthmidtNielsen, 1984), the analysis of the body mass distribution of the species in mammalian faunas has proven a fruitful ap- a small number of categories (Text-Figure 3), leading to several pertinent observations. A gap in medium size interval (between 500 g and 8 kg) occurs in faunas from more open environments, for example savannahs, whereas, the cenograrns for medium-size species from more closed habitats, for example forests, are more continuous. The slope of large mammals varies in relation to the aridity of the environment, with fewer species, that is a steeper slope, in more arid proach for paleobiological studies (e.9., Damuth and MacFaddery 1990). The mean body weight of species can be determined from tooth size, with extant species of mammals from the different taxonomic groups being used as references for calculation (see, for example: Damuth and MacFadden, 1990; Gingerich and Smith, 1985; Gingerich et al., 1982; Legendre, 1989; Legendre and Roth, L988). The body weight distribution of component species of a environments compared with many large species, that is a shallower slope, in more humid environments. The slope for modern community is represented by a curve called "cenogram" (Text-Figure 1; see: a Legendre, 1986; 1989). Other than for bats and carnivores, species are plotted according to the logarithm of their body weight on the Y-axis and to their rank in order of decreasing body weight on the X-axis. The slope of the cuwe reflects both the number of species in the size-interval considered, as well as the spacing of species by size. A gap represents a portion of the weight spectrum that is unoccupied by any species (Legendre, 1986;1989). small species varies in relation to temperature, with fewer small species separated by larger size steps, that is a steePer slope, in temperate habitats, compared with many small species closely spaced in size, that is a shallower slope, in tropical habitats. One can therefore see that cenograrts give a taxon-free method of interpreting changes in mammalian communities in relation to qualitative environmental characteristics. Specific Diversity of Murinae defined by Musser and Carleton (in Wilson and Reeder, 1993), with the commensal species being excluded from the analysis. The murines have been an important component of the small mammal faunas of Western Europe during late Neogene times. Todap as shown in Text-Figure 4, the high diversity of non-commensal Murinae is characteristic of tropical climate regions of the Old World and of Australia (Misonne, 1969). The number of species of Murinae in any local fauna shows significant differences, which are related to geographic Our second approach is based on the analysis of the specific diversity of the rodents of the sub-family Murinae, as Ln (Body weight, in g) 15 and climatic gradients. In tropical climates the number of 10 species is much higher than five, and indeed, this number seems to be a clear threshold value (Misonne,1969). In earlier publications, using such obseryations, we have already drawn inferences about the Pliocene climate in southwestern Europe (Aguilar and Michaur; 1984; Michaux, 1971). Now, however, onthebasisof the availability of morenumerous and more precise data, we intend to take our analysis one step further, using the Murinae to quantify our estimations of climatic conditions and changes. These precipitation. will be based on correlations which can be observed between the number of species and climatic parameters, such as temperafure and THE MAMMALIAN FOSSIL RECORD Text-Figure Cenogram of the modern fauna from the Rwindi-Rutshuru Plain (Zaue). The mean body weight of species from a local fauna is plotted on the ordinate, ranked in decreasing size order on the abscissa. Strait lines (in gray) schematize a cenogram. The slope reflects the richness: a shallow slope reveals a great number of specieswithin a size interval(right part of the cenogram), whereas a steeper slope attests to theirpaucity (left part). The length of the abscissa corresponds to the total number of species in the community. 1. In the area to be studied, extending from Maghreb to Central Europe (Text-Figure 5), fossil mammal-bearing localities are nurnerous. Southern France and southeastern Spain are particularly rich in localities, especially in those bearing small mammals. Localities with a rather complete fauna, including both small and large species, and thus useful for cenogram analysis, are less Nunerous. Fossil mammal-bearing localities fall into two classes (Table 1). First, there are karstic fissure fillings. The sediment is usually richly fossiliferous, the fossils here being mostly 110 PLIOCENE MAMMALS AND CLIMATIC RECONSTRUCTION IN THE WESTERN MEDITERRANEAN AREA Text-Figure 2. Geographical distribution of the extant faunas used for the cenogram analysis. those of small mammals. These have been accumulated by means of owl pellets. The diet of the nocturnal birds of prey gives a good picture of the local fauna of small mammals (Andrews, 1990; j6nossy and Schmidt, 1970; Mikkola ,1983). The locality of layna, in Central Spain, is interestingbecause one finds there not only the fossil remains of small animals broughttherebyowls,butalsotheremains of largemammals HUMID F a llJ which havebeen either brought there by carnivores or trapped in the cavity (Crusafont et a1.,1969;Perezand Soria, i989-90). Second, there are fluvio-lacustrine localities, which yield either the remains of large mammals (Cornillet Gu6rin et al., 1970), or those of small mammals (Terrats: Michaux, 1976; Celleneuve: Aguilar et al., 1989; Michaux,t969). Rarely, as in the cases of the fauna of Perpignan, a locality also known as "Serrat d'en Vacquer", and that of Villeneuve de la Raho (Aymar, 1992), one has a more complete assemblage. In fluviatile deposits, remains usually result from animals being trapped near watering holes or drowned in floods o t! c( (Behrensmeyer and Hook, 1992). Small vertebrate remains may be enriched by owl pellets. These assemblages yield Text-Figure 3. The four categories of cenogram shape. The body weight distribution of species in a mammalian community is linked to its environment. ln an open habitat (like a savannah or a deset), medium-sized species between 500 g and 8 kg - are rare or absent (bottom), whereas they are abundant (top) in a closed environment (as a forest). ln humid conditions, large species -above 8 kg - are abundant (shallow slope, from the left part of left cenograms), whereas they are rare in arid conditions (steeper slope, on the left part of right cenograms). mixed samples of the local and regional faunas. We have established faunal lists fromthe following sources: Coiffait-Martin (1991) and Jaeger (1975) for the northern part of Africa, Fejfar and Heinrich (1981.;1985;1990) and Jiinossy (1986) for Central Europe. For southern France, data aretaken from Aguilar, Bachelet er al. (1993), Aguilar and Michaux (1984), Aguilar, Michaux et al. (1993), Aguilar et al. (1989; 1991), Aymar (1992), Clot et al. (1976), Gu6rin and Mein (1971), Mein and Aymar (19U), and Michaux et al. (1990). Data for the Spanish Pliocene are compiled from Aguilar, Michaux et al. (1993), Castillo Ruiz (1990), Mein et al. (1983; 1989-90), and Perez and Soria (1989-90). All local faunas correspond to a single locality, with the exception of the fauna of Montpellier in southern France. This is a composite assemblage, with the large mammals belonging to the well-known Montpellier fauna collected in marine littoral sands (Gu6rin and Mein, 1971; Michaux, \969), and small mammals coming from the locality of Celleneuve, where one finds fluvio-marine deposits contemporaneous with the marine sands (Aguilar et al., 1989; Michaux and Suc, 1e80). 111 THE PLIOCENE: TIME OF CHANGE *Jt F,TT i11 '-r2 77771 6-20species Text-Figure 4. Species diversity of modern Murinae in the Old World (after Misonne, 1969). Commensal species are excluded. Notethe high numberof species occurring in the Equatorial regions of Africa and Southeast Asia, comparedtothe low number of species found in the Palearctic region. CHRONOLOGY CORRELATTON AND CALIBRATION In each area studied, chronological successions of localities were established by two main criteria: faunal composition and the recognition of evolutionary lineages. Immigration events were also used (Aguilar et al., 1991 ; Aguilar and Michaux, 1984; 1"987; Aguilar, Michaux et al., 1993; Bachelet, 199Q; Castillo Rutz, 1990; Mein et al., 1983; 1989-90). Correlatioru between Spain and southem France (Table 1) were based on the occurrence of cofiunon species in the faunas, or on similar faunal compositions (Aguilar, Bachelet et al. ,1993; Agutlar, Michaux et al., L993;Bachelet, 1990; Castillo Ruiz, 1990). The chronological assessment of the Alcoy mammal-bearing locality of Spain is uncertain. Its lowermost Pliocene age has still not been clearly demonsfrated (Aguilar et al., 1990). The correlations between the faunal successions in central Europe and southwestern Europe follow Fejfar and Heinrich (1981,;1985;1990). These were based on the same criteria. However, because the two areas are rather distant, the corre- lations are accordingly weaker, especially for the lower Pliocene faunas. Even worse, because their composition is di.fferent from those of southwestem Europe, is the situation with faunas from North Africa. The biostratigraphy there follows Jaeger (1975),Jaeger et al. (1977) and Coiffait-Martin (1ee1). Text-Figure 5. Geographicaldistributionofthefossilmammal-bearing localities used for the present study. The task of time calibration is based on correlation. The calibration established for southern France is used as a reference for the three areas under discussion. Some of the French localities in the lower Pliocene are correlated geometrically with marine deposits, and others, in the upper Pliocene, are in sections which have been calibrated with the paleomagnetic scale. The time interval between fossil mammal localities is rather narrow. 1t2 PLIOCENE MAMMALS AND CLIMATIC RECONSTRUCTION IN THE WESTERN MEDITERRANEAN AREA Magnetostratigraphic data have been used to determine absolute ages. In this study we have followed the chart of Cande and Kent (1992), using also information given by Hilgen (1991). The marine stratigraphical scale is that of Cita et al. (1993). In southern France, information on magnetostratigraphywas obtained fromlindsay (1985, modified in Aguilar and Michaux,1984), Biquand (198n, Biquand et al. (1990, modfied in Clauzon et a1.,1990), Clauzon et al. (1987;1990), and information on general stratigraphy from Aguilar et al. (1989) and Clauzon et al. (1987;1990). To calibrate upper Pliocene mammal localities, most of which are karstic deposits, they have been correlated with localities ftom the Valensole Basin (southem France), for which magnetostratigraphic data are available. In addition, two imrnigration events in Europehavebeenused, thatof the horce Equus (Lindsay et al., 1980) and th at of the v ole Miuotus (Allophaiomy s) (Chaline, 1986). nearly a sub-desert environment at the Miocene-Pliocene boundary. La Calera and Layna faunas (respectively at ca. 4.0 My and -3.0 My) indicate a still open environment later in the Pliocene, although less open than in Alcoy. At ca. -2.5 My, the Rincon L fauna shows a very arid environment. In southern France (Text-Figwe 7), lhe studied faunal succession began slightly later than in Spain. At ca. -5.0 My, the fauna of Montpellier/Celleneuve indicates a closed environment in a humid climate. One million years later, the fauna of Perpignan indicates an environment slightly war(ner, less humid but still rather closed . At the time of the Montouss6 5 fauna, the environment was dryer than before. Finally around 1..0 My, the environment of Le Vallonet was still dryer and more open than at Montouss6 5. ln Central Europe (Text-Figure 8), between -4.0 and -2.0 My, the faunas indicate respectively, for Csarn6ta, a rather warmclimate and a relatively openvegetation; for theHajnacka fauna (ca. -3.0 My), a dryer and colder environment; and for Vill6ny3, amorehurnid and a somewhatwarmerenvironment than for Hajnacka but less than in Csarn6ta time. At about -1.2 My, Nagyharsanyhegy 4 shows a more open vegetation under colder and more arid conditions. RESULTS Cenograms Faunas of three geographic areas have been used for cenogram analysis : central Europe, Southern France, and Spain. Cenograms are given in Text-Figures 6-8. For Spain (Text-Figure 6), at ca. -5.0 My, the fauna of Alcoy indicates an open environment and a dry climate, SOUTHERN FRANCE dl SPAIN I o)l <l ol ,-] Rincon 1 l oll c')l I G'l Le Vallonnet Montouss6 5 :l ]' :i I 4) I I La Calera ls lo u -l' ]F l5 Rank order ol sp€cies l<g 4-l<E ti I ]F lo lo lt '+' 0510t52025S Bank order of sp€cies Text-Figure Sequence of cenograms from Spain between €.0 My and -2.0 My. Cenograms are arranged according to their chronological position 6. Sequence of cenograms from Southern Text-Figure France between -5.0 My and -1 .0 My. Cenograms are arranged according to their chronological position. 113 7. THE PLIOCENE: TIME OF CHANGE CENTRAL EUROPE -0.5 -1.0 Nagyharsanyhegy 4 Murine species diversity -1.5 -2.O I | t"tt r"' I -. France Spain Maghfeb Villdny 3 ._.qq_q.i. -2.5 -3.0 (E o E .9) o D 6 ! E o -3.5 -4.O tf) ' -4.5 Csarn6ta 2 10 t5 20 25 30 5 Rank order ot species -5.0 -5.5 -6.0 -6.5 Text-Figure Sequence of cenograms from Central Europe between -3.5 My and -1 .0 My. Cenograms are arranged according to their chronological position. 8. The overall conclusion of our cenogram analysis is that Europe was climatically differentiated during the fust half of the Pliocene: a more open vegetation with a relative aridity characterized the Iberian Peninsula, incontrast to France and central Europe. In the second half of the Pliocene, a trend 0510 Number of species Text-Figure Evolution of species diversity of Murinae from France, Spain, and Maghreb, between -6.0 My and appeared everywhere towards aridity and lower temperafures. 9. -1.0 My. If these results confirm previous observations, the strong Specific Diversity of the Murinae The specific diversity of the Murinae during the Pliocene is givenby Tables 2-4 and Text-Figure 9 for Spain, France and the Maghreb. During the lower Pliocene and until the beginning of the upper Pliocene (ca. -2.4 My), the number of species generally surpass the threshold value of 5 species in southern France (continuous line), and in Spain (broken line), contrary to the Maghreb (dotted line) where it is less than 5. The maximum diversity is reached between -4.0 and 1.3 My, whereas in southern France and Spain a minimum appears to occur around -4.5 My. Changes in specific diveisity during the Pliocene are almost parallel in southem France and in Spain. The small difference noticed between southern France and fluctuation noticed at ca. -4.5 My must be explained. It may result from a bias due to fossilization: the fossil faunas involved have an exclusively fluviatile origin. An alternative explanation is that its origin lies in a change of climate. The fact that this fluctuation was observed in France, Spain and North Africa favours the second hypothesis. A preliminary analysis using more than 70 extant faunas ftom the Old World shows a high correlation between diversity of the Murinae and climatic parameters (Text-Figure 10). This observation is the basis of a method developed for quantitative estimation of late Neogene climate (Legendre, Montuire et al., in prep.). The method will be explained in more detail in a further paper. Preliminary results are given for the Pliocene of southern France and of Spain, the estimates being calculated using regression of the number of murine species on the mean annual temperature and rainfall (Tables 2 and 3; TextFigure 11). Warm conditions prevail during the first half of the Pliocene, highest temperatures being reached between ca. -4.0 and -3.5 My, with values of 2l'C for annual mean temperature. The presence of the bat Syndesmotis independently confirms these warm temperature estimates for the faunas of SEte and Mont-H6lbne (Bessedik et al., 1984; Spain may be due to the low accuracy of the correlation between the two sequences of localities. This situation markedly differs from that in the Maghreb, where specific diversity remains more or less stable and remains below 5 species. Nevertheless, a minimum occurs at around -4.0 My and a maximum at ca. -3.0 My. In Southern France as well as in Spain, a decrease in the murine diversity has begun ca. -3.3 My. Since -2.5 My, the number of species in southwestern Europe has never been more than 6 or 5. 114 PLIOCENE MAMMALS AND CLIMATIC RECONSTRUCTION IN THE WESTERN MEDITERRANEAN AREA Murine species and temperature 2000 g'u gzo E E a E E (E 1500 8. ts o I I 'd ;10 g c(u t (! 1000 (E5 T=1.15xSp+9.15 N=114 Error = 4.8 r = 0.7O4 Eo -5 (6 o) 500 = 0 051015 Number ol species 0 5 10 Number of species 15 Text-Figure Diagram displaying the relation between the number of extant murine species and temperature (a) and rainfall (b). The linear regression equation used to predict temperature (T) and rainfall (R) from the number of species (Sp) is given, with the number of faunas (N) used for calculation, the error of the prediction (Error), and the correlation coefficient (r). 10. Legendre, 1982). Thereafter, the mean annual temperatures fell to 13oC and precipitation decreased from 1100 mm at the mid-Pliocene to 700 mm at the end of the Pliocene. Low mean annual temperatures and precipitations also charac- terize the early Pliocene, with values more or less similar to those of the late Pliocene. CONCLUSION Both of our approaches, community analysis and species diversity, agree in their qualitative estimate of the climate. Moreover, the latter approach supplies quantitative estimates of the climatic parameters. However, we stand in need of additional sources of information in order to cross-check the validity of these results. For the lower Pliocene, cenograms show a differentiation of the climate in Europe, between the Iberian Peninsula and the rest of Europe: the climate being more arid and the environment more open in the former than in the latter. France and central Europe have lower temperatures and more arid conditions during the second half of the Pliocene. The murines point to a somewhat tropical climate during the lower Pliocene. We interpret the diminishing of diversity during the upper Pliocene as a shift in the climate towards more arid and colder conditions. However, the temperature was probablynotas cold as during a glacialperiod. Mammalian paleofaunas show that the mean temperatures in the northwestern Mediterranean areas during the Pliocene were never low: from 17'C at the beginning of the Pliocene, they reached the high value of 21oC at ca. -4.0 and -3.5 My, falling then to 13'C at the end of the Pliocene, when glacial cycles started. Mean annual rainfall (mm) 500 -1.0 -1.5 -2.O 1000 France Spain -2.5 G' = 6 CD -3.0 -g.s -4.5 -5.0 -5.5 -6.0 -6.5 ACKNOWLEDGEMENTS 10 Text-Figure 15 20 25 The French National Committee on Stratigraphy provided financial support. J. Morales (Madrid) provided information about the Layna fauna. P. Luckett (Puerto Rico), B. Marandat (Montpellier), and M. Ruse (Guelph) improved the English text. ences de Mean annual temperature ("C) Estimation of temperature and rainfall between €.5 My and -1 .0 My, in France and Spain, based on murine species diversity. 11. This is contribution n'98-007 from the Institut des Scil'Evolution de Montpellier, France. 115 THE PLIOCENE:TIME OF CHANGE References cited AGUILAR, J..P., BACHELET, B,, BONNET, A., LESAGE, }.-L. ANd MICHAUX,J. L993 Le gisement karstique du Grand Serre (commune BIQUAND, D., DUBAR, M. and SEMAH, F. 1990 Paleomagnetic correlation of the Mediterranean Upper Neogene biochronology and Villafranchian vertebrates sites of theMassif Central, France. Quaternary Research,33: 241,-252. de Montclus, Gard). Donn6es nouvelles sur les faunes de rongeurs du Pliocdne terminal dans le Sud de la France. G e ob io s, Ly on, 26 : 633-&0. AGUILAR, J.-P., CALVET, M. and MICHAUX, I. 199I Pr6sence de Progonomys (Mrndae, Rodentia, Mammalia) dans une association de rongeurs de la fin du Miocdne moyen (Castelnou 1B; Pyr6n6es-Orientales, France). Ly on, 24; 503-508. AGUILAR, J.-P., CLAUZON, G. and MICHAUX. f. 1989 La lirnite Mio-Pliocdne dans le Sud de la France d'aprEs les faunes de rongeurs;6tat de la questionetremarques sur les datations l'aide des rongeurs. Bollettino della Societi P aleontologica ltaliana, Modena, 28:. 137 -1.45. Geob ios, CALDE& W.A., Itr function and life history. Haward University Press, Cambridge,431 p. CANDE, S.C. and KENT, D.V. 1992 A new geomagnetic polarity time scale for the late Cretaceous and Cenozoic. lournal of geophysical Research,9T Size, 1984 (8L0): L3917-t3951.. CASTILLO RUIZ, C. t990 Paleocomunilades demicromanriferos de los yacinientoskirsticos del Ne6geno superior de Andalucia oriental. Thesis Doctoral, i Uruv. Granada, 255 p. [Unpublished]. CHALTNE,J. AGUILA& 1984 J.-P. and MICHAUX I. 1972 Les rongeurs du Pl6istocdne moyen et sup6rieur de France. LegisementimicromammifdtesduMont-H6ldne(Pyr6neesOrientales): apporlsii laconnaissance de l'histoiredes faunes 1986 Continental et des environnements continentaux, implications 7987 stratigraphiques pour le PliocEne du Sud de Ia France. Pal4obialogie continentale, Montpellier, 1.4 (2): 19-37. Essai d'estimation du pouvoir #parateur de la m6thode biostratigraphique des ligndes chez les rongeurs n6ogEnes. B Cahizrs de Pal€ontologie, Paris: 1-410. faunal units of the Plio-Pleistocene of France. Mentorie della Societd Geologica ltalinna, Roma, 31: 175183. t990 ulletin de Ia S ociitd g€olo gique de France,P aris, (8), 3: IIl3-1124. A paleoenvironmental and paleoclimatic interpretation of a Miocene faunal succession in southern France. Critical evaluation of the use of rodents in paleoecology. Pal1obiologie mntinentale , Monpellier, t6: 311-327 . ACUILAR, J.-p., MICHAUX, 1., BACHELET, B., CALVET, M. and FAILLAT,J.-P. CHALINE, f. and BROCHET, G. 1989 Les rongeurs. Leurs significations pal6o6cologiques et paleoclimatiques; k: Campp M., Chaline,f . and Vuillemey, M. (eds), I-a Baume de Gigny (lura). Calha Pr6histoire, Edition du C.N.R.S., Paris, 17" suppl:97-70. CITA,M.B., RIO, D. and SPROVIERi, R. 1993 The Pliocene series: chronology of the tyPe Mediterranean record and standard chronostratigraphy; In: Wrenn, J.H. (ed), 26th annual meeting, the American Association of Stratigruphic Palynologists, Program nnd abstracts, Baton Rouge: 13. [Abstract]. CLAUZON, G., AGUILAR, I.-P. and MICHAUX, I. 1987 Mise en 6vidence d'un diachronisme de 5 M.a. au mur de la molassemiocbne de Valensole (Alpes de Haute Provence, France). R6visions chronostratigraphiques et implications g6odynamiques. Comptes rendus des Sciances de I'Acadtmie des Sciences, Paris, (2),305: 133-136. CLAUZON, G.,SUC, J.-P., AGUILAR, J..P., AMBERT, P., CAPPETTA, H., CRAVATTE,I., DRTVALIARI, A., DOMENECH, R., DUBAR, M., LEROY, S., MARTINELL, J., MICHAUX, J., ROIRON, P., RUBINO, J.L., SAVOYE, B. and VERNET, J.-L. 1990 Pliocene geodynamic and climatic evolutions in the French Mediterranean region. Paleontologia i Eaoluci6, Sabadell, Mem. Especial 2: 1,33-186. CLOT, A., CHALINE, ]., HEINTZ, E., JAMMOT, D., MOURE& C. 1991 l-es nouvelles faunes de rongeurs de la limite Mio-Pliocdne en Roussillon. lmplications biostratigraphiques et biog6ogr aphique. P alaecnxrt ebr a ta, Montp elher, 20: 1 4917 2. AGUILAR, J.-P., MICHAUX 1., DELANNOY, J.-J. and GUENDON, I.-L. 7993 AlatePliocenerodentfatrnafromAlozaina (Malaga,Spain). Scipta geolo gica, Leiden, 1.03: 7-22. ANDREWS, P. 1990 Owls, caaes andfossils. Predntion,preseroation and accumulation of small mammal bones in caaes, with an analysis of the Pleistocene cnae faunas from Westbury-sub- Mendip, Somerset, UI( University of Chicago Press, Chicago, 232 p. AYMAR,J. d6couvertes de vertdbrds dans le Plioc€ne (Ruscinien) du Roussillon. Naturalia Ruscinonensia, Sect. Pal6ontologie, Perpignan, 1992 (2): 748. BACHELET, B. 1990 Muridae et Aruicolidae (Rodmtia, Mammnlia) du Pliocine du Sud de h France: systtmatique, nolution, biochronologle. Thbse deDoctorat,Universit6Montpellierll,2ll p. [Unpublished]. BEHRENSMEYE& A.K. and HOOK, R.W. (Rap.) 1992 Paleoenvironmental contexts and taphonomic modes; In: Behrensmeyer, A.K., Damuth, J.D., DiMichele, W.A., Potb, R., Sues, H.-D. and Wing, S.L. (eds), Terrestrial ecosystems through time. Eztolutionary paleoecology of terrestrial plants and animals. University of Chicago Press, Chicago, London,15-136. BESSEDIK, M., ACLIILAR, l.-P., CAPPETTA, H. and MICHAUX, l. Le climat du N6ogdne dans le Sud de la France (Provence, Languedoc, Roussillon), d'aprEs l'analyse des faunes (rongeurs, s6laciens) et des flores polliniques . Pal4obiologie 1992 Nouvelles and RAGE, f..C. 1976 Montouss6 5 (Hautes-Pyr6n6es), un nouveau remplissage de fissures ir faune de Vert6br6s du Pl6istocdne inf6rieur. Gtobias, Lyon, 9: 512-514. COIFFAIT-MARTIN, B. L991, Contribution des rongeurs du N4ogine d'Alg€rie b i Ia io ch r o no lo gie mammal i enne d' Afr i qu e n o r d-o c cide ntal e. Thdse d'Etat, Universit€ Nancy I,389 p. [Unpublished] CRUSAFONT, M., AGUIRRE, E. de and MICHAUX, J. 1969 Unnouveau gisementdemammifdresd'Agevillafranchien inf6rieur (Pliocdne terminal) d6couvert A layna (Soria, Espagne). Comptes rcndus des Sc4ances de L'Acad4mie des S ciences, Paris, (D), 268: 217 4-217 6. DAAMS, R., FREUDENTHAL, M. and MEULEN A.J. van der 1988 Ecostratigraphy of micromammal faunas from the Neogene of Spain. Scipta geologica, [,eiden, Spec. issue 1: 287302. DAAMS, R. and MEULEN, A.f. van der I9U continentnle, Monpellier, 14 (2): L81-190. BIQUAND D. 1987 Les formations villafranchiennes du bassin du Puy-en- 1984 Paleoenvironmental and paleoclirnatic intelpretation of micromammal faunal successions in the upper Oligocene Velay et du plateau de Perrier, Massif Central, France. Cons6quences chronostratigraphiques des 6tudes pald,omagn6tiques. L' Anthropologie, P and Miocene of North Central Spain. con t in en ta PalEobiologie le, aris, 91: 5-28. Monpellier, 1 4 (2) : 247 -257 . 716 PLIOCENE MAMMALS AND CLIMATIC RECONSTRUCTION IN THE WESTERN MEDITERRANEAN AREA DAMUTH, 1990 J. and MaCFADDEN, B.J. (eds) Body size in mammalianpaleobiology: estimation and biological imp lications. Canrbridge University Press, Camb ridge, 397 KOWALSKI,K. 1971 The biostratigraphy and paleoecology of late Cenozoic mammals of Europe and Asia; In: Turekian, K.K. (ed),The late Cenozoic glacial ages. Yale University Press, New Haven, London, p.465477. p. DEPERET, C. 1890 Les animauxplioc€nes du Roussilon.Mdmoires de Ia Soci€ti KRETZOI,M. g1ologique de France, Pais, 3: 1-194. 1962 EISENBERG, J.F. 1981 The mammalian radiations. University of Chicago Press, Fauna und Faunenhorizont von Csarn6ta. Magyar Allnmi Fdldtany Intdzet. Eai lelent1sei, Budapest, 1959:297-395. [In Hungarian with German summaryl. Chicago,610 p. FEIFA& O. and HEINRICH, W.-D. 1981" Zur biostratigraphischen Untergliederung des 1,969 Sketch of the Late Cenozoic (Pliocene and Quatemary) terrestrial stratigraphy of Hungary. Fdkltani Kdzldny, Budapest, 3: 179-204. [In Hungarian with English summaryl' LEGENDRE, S. kontinentalen Quartiirs in Europa anhand von Arvicoliden (Rodentia, Mammalia). Eclogae geolo gicae Helaetiae, Basel, 74:997-L006. 1982 1985 Zur Bedeutung der Wirbeltierfundstdtten von Ivanovce und Hajnacka ftir die Saugetierpaliiontologie im Plioziin und friihen Pleistozlin in Europa: Kenntnisstand und 213-224. Probleme. Vesf nik Ustredniho ustaou geologickeho,P raha, 60: 1986 1989 Hipposideridae (Mammalia, Chiroptera) from the Mediterranean middle and late Neogene, and evolution of the genera Hipposideros and Asellia. lournal of Vertebrate Paleontology, Tulsa, Oklahoma, 2: 372-385. Analysisof mammalian communities fromthe late Eocene Muroid rodentbiochronology of the Neogene and Quaternary in Europe; In: Lindsay, E.H., Fahlbush, V. and Mein, P. (eds), European Neogene mammal chronology. Plenum Press, New York, p. 9L-717. GINGERICH, P.D. and SMITH, B.H. 1985 Allometric scaling in the dentition of primates and insectivores; In: fungers, W.L. (eds), Size and scaling in primate bfulogy. Plenum Press, New York, p. 257-272. GINGERICH, P.D., SMITH, B.H. and ROSENBERG, K. 7982 Allometric scaling on the dentition of primates and prediction of body weight from tooth size in fossils. American GUERIN, C. and MEIN, P. 1971, Les principaux gisements de mammifdres miocdnes et pliocdnes du domaine rhodanien. Doutments du laboratoir e de Gdologie de Lyon,H.S.: 131-170. . GUERIN, C., MEIN, P. and TRUC, G. 7970 Nouveaux mammilbres et mollusques continentaux d'Age pliocdne terminal au toit du plateau de Valensole. Comptes rendus des Scdances de l'Acaddmie des Sciences,Pans,(D),271,: 7990 and Oligocene of southern France. Palaeoaertebrata, Montpellier, L6: 191 -212. Les communaut6s de mammifdres du Pal6ogdne (Eocdne sup6rieur et Oligocdne) d'Europe occidentale: structures, geowissenschaftliche Abhandlungen, (A), 16: 1-110. LEGENDRE, S., BADGLEY, C. and GINGERICH, P.D. milieux et 6volution. Miinchner . lournal ofPhysical Anthropolo3y, 58: 81-100. 2094-2097. HILGEN, F. J. 1991. Extension of the astronomically calibrated (polarity) time scale to the Miocene-Pliocene boundary. Earth Planetary Scicnce Letters, 1,07 : 349 - 368. On thecontextof Quatemary arvicolid evolution: changes in prep. Cenograms and environmental parameters for extant mammalian faunas of the world. LEGENDRE, S., MONruIRE, S., AGUILAR, J.-P. and MICHAUX, J. in prep. Muridae (Mammalia, Rodentia) as paleoclimatic indicators: a method for quantitative evaluation of past climate applied to late Tertiary European faunas. LEGENDRE, S. and ROTH, C. 1988 Correlation of the lower camassial size and the body weight in recentcamivores (Mammalia). Hisf orical Biology, London, l: 85-98. LINDSAY, E. 1985 European Late Cenozoic biochronology and the magrretic polarity time scale. National geographic Society, Research Reports,'1,979 Prqects, p. 449456. LINDSAY, E.H., OPDYKE, N.D. andIOHNSON, N.M. 1980 Pliocene dispersal of the horse Equus and, late Cenozoic mammalian dispersal events. Naturc, London, 287: 735138. MEIN,P. HORACEK,I. 1990 1984 Miocbne moyen sup6rieur de la r6gion lyonnaise. Paldobiologie continentale, Composition quantitative des faunes de mammifbres du in community developmenq ln:Feifar, O. and Heinrich, W.D. (eds), International Symposium Eaolution, Phylogeny and Biostratigraphy of An:icolids (Rodentin, Mammalin), Rohanoz.t (Czechosloaakia), Monpellier, 1,4 (2): 239-346. MEIN, P. and AYMAR, p. 201-222. JAEGER,J.-], 1975 Les Muridae (Mammalia, Rodentia) du Plincine et du P 16 istocine du Maghreb. Origine, 1ttolution, donnies biogdographiques et pal1oclimatiques. ThEse d'Etat, Universit6 de Montpellier, L25 p. [Unpublished]. IAEGER, l.-J., LOPEZ MARTINEZ, N., MICHAUX, j. and THALER, L. 7977 Les faunes de micromarnmifEres du N6ogbne supdrieur de Ia Mdditerran6e occidentale. Biochronologie corr6lations avec les formatioru marines et 6changes intercontinentaux. Bulletin de la Sociitd glologique de France,Pais,7,1.9: 501-506. lANossY, D. 1986 Pbistocene uertebrate faunas of Hungary. Developments in Paleontology and Stratigraphy, 8. Elsevier, Amsterdam, D6couvertes r6centes de nwunifdres dans le Pliocdne du Roussillon. Nouoelles Archittes du Muste d'Histoire nsturelle de Lyon, suppl.,22: 69-71. MEIN, P., MOISSENET, E. and ADROVE& R. 1983 L'extension et l'6ge des formations continentales pliocEnes du foss6 de Teruel (Espagne). Comptes rendus des Sctances de l'Acaddmie des Sciences, Paris, (2), 296 1503-1.610. 1989-90 BiostratigraphieduN6ogdnesup6rieurdu BassindeTeruel. Paleontologia i Eaoluci6,Sabadell, 23: 121-139. MEULEN, A.f. van der and BRUIJN, H. de 1982 The mammals from the lower Miocene of Aliveri (Island of 1984 J. Evia, Greece). Part 2. The Gliridae. Proceedings of the Koninklijke Nederlandse Akademie aan Wetenschappen, Amsterdam, Proc. B, 85:485-524. MEULEN A.J. van der and DAAMS, R. IANOSSY, D. and SCHMIDT, E. 21:?5-51,. . 208p. 1992 Evolution of Early-Middle Miocene rodent faunas in relation to long-term paleoenvironmental changes. 1970 Die Nahrung des Uhus (Bubo bubo). Palaeogeography, Palaeoclimatology, Palaeoecology, Regionale und Amsterdam, 93l. 227 -253. erdzeitliche Anderungen. Bonr er zoologische Beitriige,Bortn, t17 THE PLIOCENE: TIME OF CHANGE MICHAUX]. WILSON D.E. and REEDER, D.M. (eds) L969 r6gionmontpelli6raine. 3. Gisements pliocdnes. Bulletin du Bureau de Recherches gAologiques et miniires, Orl6ars, (2),1969: 31.-37. 197'1. Evolution et signfication des peuplements de murid6s (Rodentia) en Europe sud-occidentale au N6ogdne sup6rieur. Comptes rendus dcs Scdnnces de I'Acad4mie des Les gisements devert6br6s de la S 1993 Mammal species of the World. Smithsonian Institution Press & The American Society of Mammalogisb, Washingtory 1206 p. [2nd edition]. (Manuscript accepted for publication |une, 1996) ciences, Paris, (D), 273: 3L4-317. 1976 D6couverte d'une faune de petits mammifDres dans le Pliocdne continental de la vall6e de la Canterrane (Roussillon); ses cons6quences stratigraphiques. Bulletin de la Sociiti gdologique de France,Pans,1,8: 165-170. J., Authors' addresses ]EAN-PIERRE AGUILAR lnstitut des Sciences de l'Evolution UMR 5554 du CNRS Universit6 Montpellier II Place EugEne-Bataillon 34095 Monpellier C6dex 5 MICHAUX, La faune des rongeurs du sud de la France et sa dynamique au N6og€ne sup6rieur. Vie et Milieu, Paris,40: 130-135. MICHAUX, J. and SUC, J.-P. 1990 AGUILAR, J.-P. and BACHELET, B. 1.980 Le Pliocdne de Montpellier: analyses palynologiques 21.4. France SERGE LEGENDRE nouvelles et corr6lations stratigraphiques. Bulletin du Burmu de Recherches g1ologiques et miniires, Orl6ans, (L):207- MIKKOLA,H, 1983 Oznls of Europe. T & A D Poyser, Calton, England 397 p. MISONNE, X. 1969 Africanandlndo-AustsalianMuridae. Evolutionarytrends. Annales du Musee roy al d' Afrique centr ale,S ciences zoolo giques, Tervuren, 172: l-217. Centre de Pal6ontologie stratigraphique et Pal6odcologie UMR5565 du CNRS Universit6 Claude Bemard - Lyon 1 69622 Villeurbanne C6dex France MONruIRE,S, 1995 Evolution climatique et diversit6 chez les mammiGres en Europe centrale depuis le Pliocdne. G6obios, Lyory M6m. JACQUESMICHAUX Institut des Sciences de l'Evolution UMR 5554 du CNRS & Laboratoire de Pal6ontologie des Vert6br6s, EPHE Universit6 Monlpellier tI Place EugEne-Bataillon 34095 Montpellier C6dex 5 Spdcial 18: 313-327. PEREZ, B. and SORIA, D. 1989-90 Analisis de las comunidades de mamiferos del Plioceno de Layna (Soria) y La Calera (Teru el) . Paleontologia i Eooluci1 , Sabadell, 23:231-238. PETERS, R.H. France 1983 ROGE&1. The ecological impliution ofbody size. Press, Cambrid ge, 329 p. CanrbidgeUniversity SOPHIEMONTUIRE Institut des Sciences de l'Evolution URA 327 du CNRS Universit6 Montpellier II Place Eugdne-Bataillon 34095 BuIfon. Les Epoques de la Nature. Edition critiqu e. Mdmoires du Musdum national d'Histoire naturelle, Paris, 1.0: 1-343. SCHMIDT-NIELSEN K. 1984 Scaling. Why is animal size so importantT Cambridge Univer- 1988 Montpellier C6dex 5 France sity Press, Cambridge, 241 p. APPENDIX: List of abbreviations and localities. A3 A.B ALC ALO A.K L A.K 2 AR 3 BA 2 BAC L BAC 2 BAC 6 BE B.R CAS CEL 118 Amama3 Arn Brimba Alcoy Alozaina Argoub KemeIIaI 1 Argoub Kemellal 2 Arquillo 3 Bagtr 2 Balaruc L Balaruc 2 Balaruc 6 Belmez Bulla Regia Casablanca Celleneuve COR Cornillet CR Caravaca CSA 2 Csam6ta 2 CTN3 Castetrou3 FOU4 LoFoumas4 FOU 13 Lo Foumas L3 GEN Mas G6n6gals GS Le Grand Serre HAJ Hajnacka HAU Hautimagne LO Irhoud Ocre LA lry." L.C la Calera L.G 4 l^a Gloria 4 LTO laTour M-H MO a MOb MON MR 1 MR 2 MTS 5 NAG 4 NIM O.S PE PER PDV PUI RN 1 Mont-H6ldne Moreda 1A Moreda 18 Montpellier Mas Rambault 1 Mas Rambault 2 Montouss6 5 Nagyharsanyhegy 4 Nimes Oued Smendou Peralejos Perpignan Pli de la Ville Puimoisson Rincon 1 .{4 AB SEG SET SEY TE TER TRA V.A VAG 7 VAL VEN VIL 3 VIV 2 V.RA R.S R.S Rambla Seca ,{4 Rambla Seca AB SegriEs S€te Seynes Ternifine Terrats Trassanel Villalba Alta Valdeganga 7 Le Vallonet Vendargues Villdny 3 Vivds 2 Villeneuve de la Raho PLIOCENE MAMMALS AND CLIMATIC RECONSTRUCTION IN THE WESTERN MEDITERRANEAN AREA TABLE 1 Correlation chart of the fossiliferous localities used for the present study (abbreviations are explained in appendix). Localities used for cenogram analysis are underlined. . CORRELATION (6 CHA RT Eut L o o o o) ul o Foraminif. olg a (u9 and o (') Y! (d CD Fossiliferous localities Magnetostratigraphy Southern France (u E" (L3 (1) 6E oh o Nanno. Cita & al. 1993 >o E E (u tro L Fossil mammalian localities spain lff$1 o NAG4 lMashreb NN 1.0 Lrj oTE z ut 20 o oMRl VAL O I J o F UJ 1.5 (L s E o J o f F N o BA2 22 tilNN X 19b oBACI $r o o CAS o TRA \rpl 6 a LO 2.O IX 19a ilNN o MI95 .GS oMR2 VAGT ALO o VlL3 r4Pt 5b 2.5 vilt 18 z s N uJ .FOU4 '4NN 17 [-1"o" vlr UPI U) . o PDV . ;t o BAC6 SEY o Bltl o R.S AB o o A.K2 B.R () 3.0 (L 16b 5a VI c) f o GEN l,=" tl ll ll r MOb r oBE oLA HAJ o o O.S IJJ 3.5 z lrJ o 9 (L o 16a \,lPl 4 IV oBAC2 . o FOU 13 o SET J o NIM t PU, o o R.S A4 A3 MOa o CSA2 4.0 4-1 I z uJ \,lPl 3 ilt J () 4.5 INN 13 I* OV.RA .o M-H o AR3 o V.A L'C o A.B o o z N tr UJ F L.G4 ct j MPI 2 il 12 lrcn [f vEN .9 oPE (, (J 5.0 (9 MPI I 1 c, o o q, F t - f] VIV 2 cEL/ MON Fl 3 o HAU o ALC Tg 5.5 z UJ z l! 6.0 : z a o UJ O 9 o CTNa 6.5 oLTO oCR o A.K1 o Karstic cavities o Fluviatile and lacustrine localities t19 THE PLIOCENE: TIME OF CHANGE TABLE ("C)and rainfall(mm). 2. List of Spanish localities, with their ages, the number of murine species, and the estimated values for temperature MURINAE and SPAIN Locality Caravaca Peraleios La Gloria 4 La Calera Arquillo 3 Moreda 1A Rambla Seca A4 Belmez Moreda 1B Rambla Seca AB Alozaina Valdeoanoa 7 Casablanca Baqur 2 -6.2 N 5 7 14.9 17.2 20.6 14.9 Rainfall 824 934 1 -4.9 -4.5 -4.3 -4.O 10 5 7 11 100 824 934 1 17.2 21.8 18.3 -3.5 -3.3 -3.1 155 8 10 989 1 20.6 18.3 16.0 16.0 13.7 12.6 13.7 100 -2.8 -2.6 -2.2 -2.O I 6 6 989 879 879 769 7'14 4 3 4 1.7 1.2 769 TABLE List of Southern French localities, with their ages, the number of murine species, and the estimated values for temperature ('C) and rainfall (mm). MURINAE and SOUTHERN FRANCE N Castelnou 3 Hautimagne Vendarques Terrats Villeneuve de la Raho Perpignan Mont-H6ldne Lo Fournas 13 Sdte Balaruc 2 Mas G6n6qals Seynes Pla de la Ville Balaruc 6 Lo Fournas 4 Mas Rambault 2 Le Grand Serre Trassanel Mas Rambault 1 3. T Rainfall 18.3 16.0 14.9 -6.2 I 6 5 -4.9 -4.7 -4.5 -4.3 -4.0 -3.8 -3.5 -3.5 -3.2 989 879 824 7',14 3 4 9 10 8 10 12.6 13.7 769 1 1 19.5 20.6 18.3 20.6 16.0 16.0 14.9 16.0 14.9 14.9 13.7 14.9 045 100 989 1 100 6 6 5 -3.0 -2.8 -2.8 -2.6 -2.4 -2.2 -2.0 1.8 1.2 6 5 5 4 5 3 3 12.6 't2.6 879 879 824 879 824 824 769 824 714 714 TABLE 4. List of North African localities, with their ages and the number of murine species. MURINAE and MAGHREB Locality Arooub Kemellal 1 Age -6.0 -4.0 -3.3 -3.0 -2.4 N species 4 2 3 5 Ain Brimba Amama 3 Oued Smendou Bulla Regia Argoub Kemellal 2 lrhoud ocre Ternifine 3 3 3 J -2.2 1.8 -0.7 L20