Skip to navigation – Site map

The middle Pleistocene Muridae (Mammalia, Rodentia) from l’Igue des Rameaux (Tarn-et-Garonne, France)

Marcel Jeannet and Pierre Mein
p. 177-205
This article is a translation of:
Les Muridae (Mammalia, Rodentia) du Pléistocène moyen de l’Igue des Rameaux (Tarn-et-Garonne, France)

Abstract

The discovery of the karstic system of “l’Igue des Rameaux” at Saint-Antonin-Noble-Val (Tarn-et-Garonne, France) by the Speleologic Group of Caussade, allowed to bring out an interesting and important faunasboth of large and small mammals (Rouzaudet al. 1990). Unusually four species of Apodemus have been recognized and this paper has for purposeto define their biometrical and morphological particularities. This opportunity incites us to gather some small size members of this genus, from different sites of Middle Pleistocene and propose a tentatively phyletic classification in the genealogical tree of the prolific murids family, despite the weakness of the datamaking the operation statistically random.

Top of page

Full text

We wish to thank the Museums of Bonn and Basel for lending us the Apodemus agrarius molars which enabled us to confirm our determinations and to multiply our observations and comparisons.

Introduction

1The partial filling of a karst fissure on the southern fringes of the Causse de Quercy has yielded a rather scant but very diversified microfauna. A previous study (Jeannet 2005) brought to light the predominance of rodents, showing the progressive installation of a continental steppe environment from the base to the top, followed by an abrupt reforesting under a temperate humid climate.

2Field mice (genus Apodemus) are represented by four species (Apodemus flavicollis, A. sylvaticus, A. agrarius and A. microps or uralensis), which we will describe here. The morphological and biometric study enables us to discern the evolutionary stage and to position them biochronologically in a recent phase of the Middle Pleistocene, probably at the Holsteinian/Saalian limit. This is, in any case, what the biometric study of Arvicola cantiana showed (Jeannet 2005), using the SDQ method (Schmelzband-Differenzierunge Quotient: Heinrich 1990), which corroborates the conclusions of Rouzaud et al 1990.

3Igue des Rameaux is a karst fissure opening in the southern fringes of the Causse du Quercy, about 400 m above the Aveyron River (fig. 1). A narrow shaft leads to the upper part of a network of partially filled sinuous galleries of varying widths. The joint effects of the slope and the underlying void played a primordial role in the formation of the infilling and the distribution of the microfauna. The complex is divided into an upper part (B, C, D), and a lower part (E, F, G) (fig.2).

Figure 1 – Location of l’Igue des Rameaux at Saint-Antonin-Noble-Val (Tarn-et-Garonne, France) in Quercy (south-western France).

Figure 1 – Location of l’Igue des Rameaux at Saint-Antonin-Noble-Val (Tarn-et-Garonne, France) in Quercy (south-western France).

IGN map 2012 and Géoportail

Figure 2 – L’Igue des Rameaux. Topographic map of the site (2a) and longitudinal profile (2b) showing the excavated areas.

Figure 2 – L’Igue des Rameaux. Topographic map of the site (2a) and longitudinal profile (2b) showing the excavated areas.

From ROUZAUD and al., 1990.

4The first part contains a mass of sediments retained by collapsed blocks and capped by a flowstone. As suffosion removed part of the infilling, the lower part of the flowstone has been dismantled and the sediments have partially crept downwards. This part of the network is still active, judging by the accumulation of lagomorph remains on the surface, and the subsidence marks engraved on the clay against the wall. In an earlier note (Jeannet 2005), the distribution and the biometric study of Arvicola showed that these sediments are nested under deposits containing older remains. In this way, layer 30 remained suspended above more recent deposits but is older than the deposits in the lower sector. Beyond sector G, the gallery forms a right angle filled with scree that may come from a possible second, now obstructed entrance, through which carnivores could have entered the cavity. Three large predators used this sector of the cavity as their den: the lion, the wolf and the hyena. They left abundant bones, sometimes in loose connection, and left wide incisions on several large herbivore remains (equids in particular), indicating a certain degree of consumption (scavenging?). Carnivores of all ages are represented. The hyena in particular left a layer of coprolites of mixed sizes (Rouzaud et al. 1990). Humans also visited the cavern but only left rare choppers and quartz artefacts in the upper sector. No large herbivore remains in this sector bear butchery marks.

Sampling and preparation of the material

5The search for microfauna took place over the last three excavation campaigns (1988-1990) and samples were taken during the course of the excavation from the whole cavity and from all the main layers. As time was limited, we could not sieve all the sediments removed from the cavity, and it was up to the excavators, first of all, to carry out sampling. Subsequently, the initial results guided complementary sampling, corresponding in volume to the capacity of a bucket (about 8 to 10 kg). The central axis yielded the most material and gave a significant overall view (tab. 1 and 2).

Table 1 – L’Igue des Rameaux. Suammarized plan of the site showing sampling zones. Each square indicates the layer and the number of samples. Areas B, C and D constitute the upper zone; E, F and G, the lower zone.

Table 1 – L’Igue des Rameaux. Suammarized plan of the site showing sampling zones. Each square indicates the layer and the number of samples. Areas B, C and D constitute the upper zone; E, F and G, the lower zone.

Table 2 - L’Igue des Rameaux. Distribution of species by square and layers in the upper and lower zones, pointing out the similitude of assemblages between the two sections despite a lower abundance and the lack of Amphibians and the low representation of Reptiles in the upper zone.

Table 2 - L’Igue des Rameaux. Distribution of species by square and layers in the upper and lower zones, pointing out the similitude of assemblages between the two sections despite a lower abundance and the lack of Amphibians and the low representation of Reptiles in the upper zone.

6The sediments were sieved under a jet of water with adjustable pressure, on a column of five sieves piled onto a decantation vat. Meshes of 10, 5, 2, 0.8 and 0.5 mm were used to accelerate the dispersal of sediments and facilitate subsequent drying and sorting operations. The use of the decantation vat avoids transforming the sieving zone into a mud bath. The installation of wet sieving, which is often difficult on site, is of prime importance: it accelerates sieving, makes it less laborious and above all, prevents micro-remains from breaking, making it possible to retrieve and identify them. At Igue des Rameaux, without this procedure, considering the low number and fragility of the juvenile teeth, there would hardly have been any material to identify.

7At this site, the particularly sticky sediment required additional treatment. The blocks of water-saturated clay were laid out in the sun before being sieved. The action of the sun’s rays led to the disaggregation of the sediments during subsequent sieving, with no damage or risk of loss for the remains.

8The discharge from the two upper sieves (10 and 5 mm) was sorted on site with the naked eye and the other remains were laid out separately in the sun (on empty plastic fertiliser bags) and labelled. After drying, the remains were sorted using a hand-held magnifying glass, against a pale-coloured plate where the silhouette of each grain is clearly visible. This initial “lens-type” sorting resulted in the separation of microvertebrate bones and teeth, shards and small bones from large mammals that went unnoticed at the excavation, as well as those from the different families of rodents, insectivores and diverse classes of vertebrates and molluscs.

Taphonomic diversity

9The overall aspect of the remains is very diversified as bone mineralization is very variable among the samples. Some bones are translucent, others are brownish-red or even sometimes bright red; others are totally black or marbled. Most of them are white and opaque. Other very white bones are totally covered in calcite. Only the bones from sector B are consolidated by mineralization. Most of the remains are from young individuals and did not undergo corrosion. The alteration of the remains is predominantly mechanical, causing clean breaks and rounded angles which could be imputable to transport or possibly to soil compaction.

10The taxa are relatively abundantly represented in all layers, apart from four taxa. Yet, we observe that some well-represented taxa in one sample are very rare, or even absent in another sample, even it is copious and of the same origin. The remains from the same species (maybe even from the same individual) seem to have been deposited in bulk in a given place or as if the animal died in this narrow space. This is the case, in particular, for the vertebrae from reptiles and several rodents. It is also the case for the David’s hedgehog (Erinaceus davidii Jammot, 1973), which was found in four contiguous squares, in the same layer. This pattern points to death during hibernation. But how can we interpret the presence of a small snake lying on a lateral rocky bench, in perfect anatomic connection and covered with a film of calcite? Its stretched-out position is not generally the position adopted during hibernation. Chiroptera bones were petrified, probably in identical conditions. This group appears to have been the first occupant of the cavern.

11The lagomorphs are a case apart. They are overabundant in sector D, with an enormous quantity of juvenile bones and teeth (non-worn milk teeth), clearly showing that this taxon nested in this sector, where it must have considerably disturbed the levels.

12The distance between this accumulation from the entrance and the relatively good conservation of fragile non-mineralized bones appears to preclude the transport of these remains. In the other sectors, their presence is less marked and may correspond to their dispersal into deeper levels, and thus to the transport of bones rather than to a habitat.

13Taphonomic conditions play an important role in understanding the stratigraphy, the chronology and the environment, through faunal associations, but they are still difficult to elucidate. Generally, microvertebrate remains come from bird of prey pellets, who could easily have nested on the cliff or in a fissure. In this case, the remains accumulate on the ground in a narrow space and are constant throughout the infilling, although there are variations in density, due to run-off or modifications as a result of the wall effect. At Igue des Rameaux, the fairly uniform distribution of the remains in the axis of the gallery and throughout the levels, in spite of the fact that they are not very well represented, implies that overall, apart from the case of hibernating animals and lagomorphs, the animals fell into a natural lapiaz trap and their skeletal remains could have undergone creep action with the solifluction silts. The upper deposits (C and D) are scree, with a gradual lateral rarefaction of the remains. The absence of small predator bones (birds of prey and small mustelids) in the series, in keeping with the difficult access of the site, clearly accounts for the good surface condition of the micromammal bones, the absence of digestion marks, as well as the absence of the clustering of the remains around a possible nesting area (tabl.1). The rarity of breaks, in spite of fragility due to corrosion, is probably due to movement in muddy masses, which is characteristic of this type of transport. In addition, for different reasons, bioturbation is always possible on micro-remains. However, in sectors E, F and G, the successive habitats of large predators tend to show the stability of the stratigraphy, which could also be the case for the underlying small bones. The transverse layout of zone H brings to light the absence of a link between the levels of zones E, F, G and the scree sediments. In this case, there cannot be any mixing of the microvertebrates.

14In accordance with the behaviour and the topographic distribution of the taxa, we compared the associations from the upper and lower areas to assess the similarities or the differences between them (fig. 3). The circular diagrams underline the low representation of the reptiles and batrachians from the upper zone. This implies that the herpetological group entered the lower zone, probably to hibernate, through the entrance currently obstructed by scree and concretions. The other taxa are distributed relatively evenly in both parts of the cavern.

Figure 3 –L’Igue des Rameaux. Pie charts showing microfaunal distribution between upper and lower zones (in % per species).

Figure 3 –L’Igue des Rameaux. Pie charts showing microfaunal distribution between upper and lower zones (in % per species).

15In spite of the abundance of lagomorph remains, it is difficult to assess differences within the group and in relation to other groups, due to the restricted number of taxa (four species).

16Therefore, it seems that, in spite of stratigraphic disturbances, the remains in the lower area appear to mostly come from the upper area. This is also clear in the columns in table 2, where the same taxa coexist in both sectors.

Taxonomic diversity (tabl. 2)

17A first sorting operation brought to light the predominance of rodents from the five main European families (Arvicolids, Murids, Glirids, Cricetids and Sciuridae). Early forms such as Pliomys lenki, Arvicola cantiana and Allocricetus bursae point to a Middle Pleistocene chronology and more specifically to the Holsteinian / Saalian interface, or even to an early phase, due to the climatic conditions recorded in the diverse layers, where the oldest phase records a dry steppe-like period evolving towards a humid warming period favouring the development of forest cover. The second argument refers to the SDQ Heinrich values obtained regularly on all the Arvicola molars (Jeannet 2005), confirming its chronological position. Lastly, the taxonomy of certain taxa (lion, hyena, wolf, equids) confirms this age, according to the conclusions of Rouzaud et al. (1990). Among the other groups, lagomorphs are the most abundant and belong, at least in sector D, to the European rabbit (Oryctolagus cuniculus), which also lives in dry caverns where it shelters its abundant young. This behaviour contrasts with that of the brown hare (Lepus europaeus), who prefers to live outside in vast open spaces. Its remains are only found in cave contexts when it falls in accidentally (Jeannet 2000), or more frequently due to the intervention of a predator. Then come the batrachians, with a limited number of taxa (4: Rana dalmatina, Bufo bufo, Bufo calamita, Urodela cf. Triturus sp.), but plentiful bones. Again, hibernation may account for this abundance. The numerous bat species (11: Myotis myotis, M. blythii, M. bechsteinii, M. nattereri, Plecotus auritus, P. austriacus, Rhinolophus mehelyi, Rh. euryale, Rh. ferrumequinum, Miniopterus schreibersi, Barbastella batbastellus) are also a result of hibernation, but there are few remains per taxon. They tend to be clustered in the same sector (15 M) and in the same level 60. Although several bat species often hibernate in the same cave, it is at times impossible to find a common link between them attracting them to the same site, as their behaviour is very different. We can deduce that the atmosphere of the place varied throughout time, rather than the adaptive capacities of the bats, which can flee if they are disturbed. No guano accumulation has been identified, and thus it is plausible that their visits to the site were short rather than for nursery purposes. We can conclude that the site was only used occasionally during local, often individual movements, as is traditional for bats. David’s hedgehog (Erinaceus davidii) comes from the deep layers 60 and 62 (lower sector) and may also have hibernated,. But if the habits of this hedgehog are the same as those of the present-day hedgehog, this is not its preferred place: it prefers heaps of rotting leaves that give off heat. The hedgehog was also discovered at La Fage (Jammot 1973), in layers dated to the Middle Pleistocene and at Baume Moula at Soyons (Desclaux and Defleur 1997). Its presence in zone B of the upper sector, far from sectors E, F, and G, indicates that most of this infilling is from the Middle Pleistocene, as this species is not known at other periods, and is associated here with Arvicola cantiana. The other insectivores include the European mole (Talpa europaea), represented by a high number of remains that may belong to a small number of individuals. The small mole, Talpa minor is only represented by a single individual. The group of shrews is well represented with seven species indicating a very diversified ecosystem, including the rather rare dwarf shrew. The latter lives in continental climates and its rarity is due to its tiny size and fragility. The other species of the shrew family are Crocidura russula (the great white-toothed shrew), Crocidura suaveolens (the lesser white-toothed shrew), Suncus etruscus (the Etruscan shrew), Neomys fodiens (the Eurasian water shrew), Sorex araneus (the common shrew), Sorex minutus (the pygmy shrew). They are only represented by several remains but live in very diverse environments. There are also abundant reptile remains, such as the green whip snake (Coluber viridiflavus), the grass snake (Natrix natrix), the ladder snake (Rhinechis scalaris), the European aspic (Vipera aspis), Ursini’s viper (Vipera ursinii) and the common European adder (Vipera berus). These remains are mostly clustered together in two contiguous squares, and appear to come from hibernating animals, although they do not generally hibernate so deep down.

Environment

18In order to confirm this homogeneous distribution and biodiversity, we drew up histograms of the main environmental parameters (fig. 4 to fig. 8), where possible internal variations underline significant differences. In this aim, we used the “quantified ecology” method (Jeannet 2010; Jeannet et al . 2013; Jaubert et al. 2005; with the chronoclimatic base; Martinson et al. 1987).

19In spite of the uncertain stratigraphic sequence due to the considerable drop separating the upper and lower sectors and the presence of older layers (c30) suspended on flowstones (Rouzaud et al. 1990) affected by underlying suffosion (Jeannet 2005), we respected the previously established order.

20As regards the temperature (fig. 4), the seemingly low variations between the levels are interesting to observe given their stratigraphic position. The lowest value (8.4°C) is from what is considered to be the oldest layer. It is equivalent to the annual average values for Bourg-Saint-Maurice (Savoy), Bamberg (Bavaria), Cluj (Romania) or Rostov (C.E.I./Azov). The highest (11.4°) is the current value from the centre of France (Lyon, Bourges). It is slightly higher than the value recorded in 5e at Coudoulous II (10.7°C) (Jeannet 2010). The winter/summer difference is limited to 27.5°C and conveys a semi-continental aspect on the deeper levels. The climate warms progressively in the recent levels to become cool and temperate.

Figure 4 - L’Igue des Rameaux. Temperature diagram.

Figure 4 - L’Igue des Rameaux. Temperature diagram.

21Precipitation (fig. 5) confirms the climatic conditions inferred from temperatures. Precipitation is low: 60 cm (currently 90 cm in France) and spread out over a rather long period (5 months under a rigorous climate in c.30). In spite of climatic warming, precipitation remains low and increases very gradually. The parallel between the frost (N) and snow curves (I) are naturally oriented in the opposite direction to the TMA (annual average temperature).

Figure 5 - L’Igue des Rameaux. Precipitation diagram.

Figure 5 - L’Igue des Rameaux. Precipitation diagram.

22The climatic atmosphere (fig. 6) gathers secondary but palpable data regarding daily life (frost, sunshine, nebulosity).

Figure 6 - L’Igue des Rameaux. Atmosphere diagram.

Figure 6 - L’Igue des Rameaux. Atmosphere diagram.

23At Rameaux, frost (N) lasts for 4 months under cloud cover (M) every other day, masking the sun (R).

24The vegetation (fig. 7) under cool temperate thermal conditions consists of 20 to 30% of high forest (W). Coppice and scrub (U, V) occupy the outskirts of the forest judging by their fluctuation at the same time as trees. The meadow (T) is dominant on 37 to 46 % of the territory. The marked difference in level reveals relatively constant exposed soils (S). Paradoxically, their extreme reduction in the least favourable climatic phase (c. 30), which is compensated for by the high forest, could be explained by the development of the pine grove on dry and cool soil, announcing a sharp warming in a contrasted landscape where forest species (P) then decline sharply (c. 54), with the expansion of meadowland.

Figure 7 - L’Igue des Rameaux. Vegetation diagram.

Figure 7 - L’Igue des Rameaux. Vegetation diagram.

25Soil humidity (fig. 8) is particularly marked by drought over 50 % of the area. It is associated with cool soils and affects between 80 to 95 % of the territory. These two parameters coincide closely with the steppe-type cover.

Figure 8 - L’Igue des Rameaux. Ground hygrometry.

Figure 8 - L’Igue des Rameaux. Ground hygrometry.

26Hygrophilous species (Q) are poorly represented but also tend to be more frequent in open landscapes. Their reduced rate in c. 30 confirms the presence of pine forests.

27Table 2 contains two interesting points. The agile frog (Rana dalmatina) is more terrestrial. The other anurans are toads (Bufo bufo and Bufo calamita) with semi-aquatic lifestyles. Their presence is thus linked to hibernation. As mentioned above, their presence in the lower zone shows that they came into the cave by the now obstructed opening.

28The diverse environmental parameters show the complexity of the biostratigraphy. The temperatures denote a semi-continental climate evolving from a rigorous stage to a warm temperate interglacial-type phase before becoming cool and humid with frequent but not very abundant precipitation. These thermal and hygrometric conditions are experienced differently depending on the type of vegetation and ground humidity, and vary greatly in the different faunal associations. Woody vegetation is always dominant.

29Coppice and shrubs, on the forest edge, conserve constant values and evolve with high forest fluctuations. The high percentage of hygrophilous species (Q) underlines this. The cool soil favours forested and open environments.

European field mice (Corbet 1978)

30Before addressing the systematics, we wished to situate the field mice from Rameaux in a European context (Miller 1912), based on the biometry of the first molars from several sites from the literature (tabl. 3, tabl. 4 and tabl. 5), excluding the small forms (A microps or uralensis type) due to their limited number. Table 3 presents the data from the dispersion diagrams from figures 9 and 10. The species names are presented as initials (AA= Apodemus agrarius; AF= Apodemus flavicollis; AM= A. microps; AS= Apodemus sylvaticus). The determinations are those of authors cited in the bibliographic references (Bartolomei 1964; Brugal 1981; Dietrich and Maul 1984; Jeannet 1977; Koenigswald 1972; Kowalski 1956; Kretzoi 1956; Malez and Rabeder 1984; Mauk 1990; Meulen 1973; Kratochvil and Rosicky 1952; Popov 1988; Schaub 1938; Steiner 1978; Storch and Lutt 1989). The term “northern” is applied to populations in latitudes above 45° N. The term “southern” concerns the sites below this latitude. Measurements are given in mm/100.

31The diagram of the first upper molars (M1) (fig. 9) shows the clustering and the lateral position of the present-day or fossil A. agrarius M1. We also observe the “advanced” position of the A flavicollis molars, with lengths often exceeding 193, for fossil and present-day, “southern” and “northern” specimens. The biometry does not distinguish them from A. sylvaticus on the sole basis of this criterion as the fossil forms of this species cover the whole cloud for the yellow-necked mouse A. flavicollis. With one exception, we observe that the present-day or “northern” wood mouse (A. sylvaticus), does not exceed a length of 190. It is thus possible to differentiate A. sylvaticus and A. flavicollis populations on the basis of length.

32Like for the M1, it is possible to distinguish the two present-day species on the sole basis of the length of the lower first molars (M1) (fig. 10), as those of A. sylvaticus are limited to 176. The fossil or present-day A. flavicollis M1 constantly exceeds 174. The fossil forms may be the smallest for the species.

Systematics of murid rodents

33We will only retain the systematic study of the Murid rodents from the study of Arvicola published in an earlier note (Jeannet 2005). They are not very abundant but seem interesting for the systematics, paleogeography and biochronology.

34Biometry and morphoscopy are invaluable tools for the description of the different odontological forms characterizing species and sub-species. Note that morphometric geometry methods, in particular, have developed a lot over the past years; they have proved useful for accurately differentiating species, subspecies or populations, allowing for the quantification of variations in conformation and size. Many authors have focused on this method of determination and comparison, particularly for Murids, but most of the research concentrates on whole skulls. The teeth studied are upper first molars (the largest and the most complex). Comparative studies of Apodemus agrarius are thus still very limited, especially as those from Rameaux were the first discovered in France. Although the present study is not based on geometric morphometry, it cannot ignore these techniques which allow for the description of a new subspecies, attributing the smallest species to Apodemus maastrichtiensis and differentiating Apodemus sylvaticus and Apodemus flavicollis at a European scale (e.g. Janzekovic and Krystufek 2004, Renaud 2005, Barciova and Macholan 2006, Jangjoo 2010).

Table 3 – European Apodemus. Data table for dispersal diagrams (fig. 9 and 10).

Table 3 – European Apodemus. Data table for dispersal diagrams (fig. 9 and 10).

Table 4 - L’Igue des Rameaux. Measurements, type and localization of upper molars of Apodemus. The sketched elements are indicated in the last column.

Table 4 - L’Igue des Rameaux. Measurements, type and localization of upper molars of Apodemus. The sketched elements are indicated in the last column.

Tableau 5 - L’Igue des Rameaux. Measurements, type and localization of lower molars of Apodemus. The sketched elements are indicated in the last column.

Tableau 5 - L’Igue des Rameaux. Measurements, type and localization of lower molars of Apodemus. The sketched elements are indicated in the last column.

Figure 9 - European Apodemus. Dispersal diagram of M1 from some European populations, including all types, periods and locations (see tabl. 3)

Figure 9 - European Apodemus. Dispersal diagram of M1 from some European populations, including all types, periods and locations (see tabl. 3)

Figure 10 - Dispersal diagram of M1 from some European Apodemus populations, including all types, periods and locations (see tabl.3): LM1= length of the first lower molar; W/1 = width of the first lower molar. Except the small types, the diagram points out the difficulties, or even the impossibility, of discriminating species using only biometric data.

Figure 10 - Dispersal diagram of M1 from some European Apodemus populations, including all types, periods and locations (see tabl.3): LM1= length of the first lower molar; W/1 = width of the first lower molar. Except the small types, the diagram points out the difficulties, or even the impossibility, of discriminating species using only biometric data.

35Below, we recall the classification of the Murid family:

36Family: Muridae Gray, 1821
Sub-Family: Murinae Baird, 1857
Genus: Apodemus Kaup, 1829
Sub-Genus: Sylvaemus Ognev and Worobiew, 1923
Species: Apodemus flavicollis Melchior, 1834; Apodemus sylvaticus Linné, 1758 ; Apodemus uralensis Pallas, 1778
Syn. Apodemus microps Kratochvil and Rosicki, 1952; † Apodemus maastrichtiensis Kolschoten, 1985
Sub-species: Apodemus Kaup, 1829
Species: Apodemus agrarius Pallas, 1771
Sub-species: Apodemus agrarius iguensis n.ssp

The murids from igue des rameaux

37When we began the study of the murid rodents from Igue des Rameaux, we thought that, like for most Pleistocene sites, we would only encounter one or two species at the most. A more in-depth examination revealed characteristic forms of Apodemus flavicollis, A. sylvaticus, A. agrarius and the measurements brought to light A. uralensis (cf Apodemus microps). We thus tried to define the species for each molar. Biometry and statistic tests enabled us to isolate four forms of M1 and M1 but they were not sufficient to separate the M2 from M2. For the latter, it was necessary to combine two techniques: morphology and biometry. For the M2, in particular, uncertainty prevails concerning A. flavicollis and A. sylvaticus, as the latter is rare. The nomenclature of the dental structures is from Michaux (1971) (fig. 11), slightly modified by the addition of cingulate margins and the longitudinal median crest. Tables 4 and 5 present the measurements of the molars and their position at the site and figures 12 and 13 illustrate the drawings. The reference molars of present-day A. agrarius are presented in figure 14. The dispersion diagram is established with lengths and widths (fig. 15), separating the diverse species. The regression lines follow the main axis of the ellipses and are not illustrated here.

38The measurements used are the classical length (L) and width (W). However, the technique used may differ from that adopted by certain researchers, without necessarily being more effective. It aims to reduce the parasitic influence of size variations imputable to certain subsidiary cusps. These external protuberances are often used as a reference for the initial position of the reticule of the micrometer, but we prefer to align the reticule on the tangents at t8 and t5 for the upper molars, and on the posterior angles of the chevrons (tC-tB and tE–tF) for the lower molars.

Figure11 – Descriptive schema of the various structures of the Apodemus mastication table (after Michaux 1971)

Figure11 – Descriptive schema of the various structures of the Apodemus mastication table (after Michaux 1971)

Apodemus flavicollis Melchior and Apodemus sylvaticus Linné (fig. 12 and 13).

39Although present-day Apodemus flavicollis is, on average, larger in size than A. sylvaticus, these two species are morphologically and biometrically very similar and the description of one requires a comparison with the other. For this reason, we will consider them simultaneously and compare them here. The specimens and measurements are indicated in tables 4 and 5.

Figure 12 - Apodemus agrarius (n° 1, 2, 3, 4, 5, 6, 7, 8, 11, 12) ; Apodemus flavicollis n° 9, 13, 16) ; Apodemus sylvaticus (n° 10, 15, 17, 18) ; Apodemus uralensis (n° 14)

Figure 12 - Apodemus agrarius (n° 1, 2, 3, 4, 5, 6, 7, 8, 11, 12) ; Apodemus flavicollis n° 9, 13, 16) ; Apodemus sylvaticus (n° 10, 15, 17, 18) ; Apodemus uralensis (n° 14)

1 =M1G ; 2 =M1D ; 3 =M1D ; 4 =M1D ; 5 =M2G ; 6 =M1G ; 7 =M1G ; 8 =M1D ; 9 =M2G ; 10 =M2G ; 11 =M2G ; 12 =M2D ; 13 =M2G ; 14 = M1D ; 1M2G ; 16M2D ; 17-18 =M1D

40M1 – As we know from the exhaustive studies of L. Pasquier (1974), the morphological characteristics of the M1 are too unstable to distinguish A. flavicollis and A. sylvaticus on the basis of the molars. In the present case, after separating them with biometry, we observe the following differences:
- the silhouette is slightly thicker for A. sylvaticus than for A. flavicollis (L/W is respectively 1.52 and 1.59);
- t1 in relation to t2/t3: clearly receding in A. sylvaticus; very slightly receding in A. flavicollis. This situation results in a very open angle of the syncline separating t1 from t2 in A. sylvaticus;
- t4 in relation to t6: receding in A. sylvaticus, advanced in A. flavicollis;
- t7: isolated in A. flavicollis, linked to t8 or adjoining t4 in A. sylvaticus. These characteristics are not yet visible on n° 7 (fig. 13) due to its young age.

Figure 13 - Apodemus flavicollis (n° 1, 2, 3, 4, 9, 10, 11, 13, 14, 16, 18 ) ; Apodemus sylvaticus(n° 5, 6, 7, 8, 12, 15, 21) ; Apodemus uralensis (n° 17, 19, 20).

Figure 13 - Apodemus flavicollis (n° 1, 2, 3, 4, 9, 10, 11, 13, 14, 16, 18 ) ; Apodemus sylvaticus(n° 5, 6, 7, 8, 12, 15, 21) ; Apodemus uralensis (n° 17, 19, 20).

1 = M1G ; 2 = M1G ; 3 = M1D ; 4 = M1D ; 5 = M1D ; 6 =M1D ; 7 = M1D ; 8 = M2D ; 9 = M2G ; 10 = M2G ; 11 = M2D ; 12 = M2D ; 13 = M2D ; 14 = M1G ; 15 =M1G ; 16 = M1D ; 17 = M1-2G ; 18 = M3G ; 19 = M1G ; 20 = M1G ; 21 = M1G.

41M2 – This molar is the easiest to determine on account of the morphology of t9. But there are two intermediary types between the two typical forms of A. flavicollis (fig 12 n°13) and A. sylvaticus (fig.13 n° 8). The “reduced t9” form characteristic of A. flavicollis reaches a climax on specimens 9 and 10 (fig. 13), where t9 is absorbed by t6. The other extreme case concerning A. sylvaticus (fig. 13 n° 8) presents a t9 directly linked to a t8 by a brief isthmus of enamel.
- The posterior cingulum (cp) is reduced to a narrow rod placed on the side of t8 and dipping towards the front;
- t7 is low, not worn, isolated from t4 and t8 (type A. flavicollis).

42

Figure 14 - M1/-M2/ and M/1-M/2 of Apodemus agrarius Pallas, 1771.

Figure 14 - M1/-M2/ and M/1-M/2 of Apodemus agrarius Pallas, 1771.

N° 1 to 5: Bonn Museum collections.
N° 6: Basel Museum collections.
Origin: n°1: (n° 55/03 ♀) –Niedersachsen-Hörden a Hartz (Germany).
N°2: (N° 56/7 ♂) – Niedersachsen Willershauser a Harz (Germany).
N°3: (N° 55/18 ♂) – Niedersachsen Willershauser a Harz (Germany).
N°4: (N° 54/15 ♀)- Niedersachsen Willershauser a Harz (Germany).
N°5: (N° 56/3 ♂) - Niedersachsen Willershauser a Harz (Germany).
N° 6: (N° 97/70 ♀) – Siaolin-Kirin; Manchou du Kuo (Manchuria).
Note: on the M1/ cusp lengthening; on the M2/ the absence of t3 (except n° 6); on M/1 and M/2 the absence of the external cingulate margin

43The dispersion diagram (fig. 15) shows that biometry cannot be used to separate the two species based on the M2.

Figure 15 – L’Igue des Rameaux – Molar dispersal diagrams of the various Apodemus species collected. The M2 cannot be specifically separated by their dimensions but by their structures. LM1/=length of the first upper molar; WM1/=width of the first upper molar, etc.

Figure 15 – L’Igue des Rameaux – Molar dispersal diagrams of the various Apodemus species collected. The M2 cannot be specifically separated by their dimensions but by their structures. LM1/=length of the first upper molar; WM1/=width of the first upper molar, etc.

44The morphological comparison of M1 and M2 shows that the form of the posterior part is rather different for these molars. This is due to the variability of the shapes, as for a same individual, the “rosette” pattern is very similar.

  • the posterior cingulum, always present and often enormous on the M1, is absent or simply laminar on the M2;

  • strong to very strong t9 on the M1, it is still quite strong for A. sylvaticus and can reduce and disappear on the A. flavicollis M2.

45These disparities identified between the two main molars of these two species are not universal but are nonetheless very interesting on a local basis for their distinction.

46M1– No morphological criteria for the M1 enabled us to clearly differentiate these two field mice species. This was only possible through biometry and dispersion diagrams (fig. 15). Like for the M1, we observe that the A. sylvaticus molars are stocky, shorter and proportionally wider, with a Length/width ratio of 1.72 for A. flavicollis and 1.61 for A. sylvaticus, but this difference is not statistically significant (tabl. 6). For comparative purposes, we illustrate (fig. 9 and 10) the dispersion diagrams of the first molars of all types of European field mice, from all periods. This showed that the M1 from southern populations of A. sylvaticus are proportionally wider than in northern and eastern populations, for present-day and fossil specimens. As for the length, it is slightly shorter than for A. flavicollis during the Middle Pleistocene, but is still very marked in present-day southern populations; in particular for the Iberian forms that reach the size of present-day western A. flavicollis (Jeannet 2000b). The dimensions of the fossil and present-day molars of A. flavicollis seem to be stable (superior in length to A. sylvaticus and proportionally narrower). We also note that the populations from Eastern Europe (Austria and Berlin) are larger than western populations. This diagram classification is largely based on and confirms the observations of Pasquier (1974).

47Table 6 summarizes (the detailed calculations are presented in tables 9 to 12) the results of the comparisons between the length and width on one hand, through the correlation coefficient, and between the diverse species using the t test. No calculations were made for the scantest series. We observe in particular that the t test based on the L/W ration between A. flavicollis and A. sylvaticus shows a highly significant difference for the M1 and non-significant for the M1.

48On the M1: - the tma is of average size and never absent or separate;
- the external cingulate margin only once bears four conules and most often two, four times three and one (n = 19);
- the median longitudinal crest is very frequently absent or hardly outlined;
- c1 is rarely linked to tA but most often contiguous or isolated. Cases of total independence are not rare but are not predominant.

49M2- In the absence of sufficient material attributable to A. sylvaticus, we could not reliably differentiate the two species with statistical tests. However, the dispersion diagram (fig. 15) quite clearly shows a lateral position of three molars with a proportionally higher width than the others, like on the M1. Against all odds, the external cingulate margin is absent or slightly outlined (fig. 12 n° 10), unlike the others (fig. 12 n° 16). When these three molars are associated with each other in the calculations, the correlation coefficient is 0.53; individually, the coefficient reaches 0.95 for the forms with a significant external cingulate margin. This implies, with no formal statistic proof, that there is a clear difference between these two groups.

50All these elements place Apodemus sylvaticus and Apodemus flavicollis in a relatively early Middle Pleistocene, as many Lower Pleistocene characteristics (absence of median crest, reduction of the tma, position of c1, number of conules) are still frequently marked, while those of the present-day forms are rare. We also observe the frequency of an enamel loop separating tE and tF.

Table 6 - L’Igue des Rameaux. Statistical comparison between the molars of Apodemus.

Table 6 - L’Igue des Rameaux. Statistical comparison between the molars of Apodemus.

AA = Apodemus agrarius; AF = A. flavicollis; AM = A. microps; AS = A. sylvaticus.
In normal populations, the correlation coefficient (r) is close to 0.85. A negative correlation indicates that one parameter evolves in the opposite direction to the other (cf. comparison between the product and the ratio of length and width of m1).

Apodemus agrarius Pallas, 1771 (fig. 12 and 13)

51The striped field mouse is currently an unknown species in Western Europe but extends from Eastern Europe to the Far East (China and Korea). It lives in wooded steppes, on the fringes of birch forests in a wet atmosphere up to an altitude of 900 m (von Böhme 1978; B. Sala 1974).

Description and comparison : (4M1; 1 M2; 7 M1; 22M2; 1 M3)

52M1It is similar in size to that of Apodemus flavicollis, but with a very different morphology, not on account of the number or layout of the tubercles, but by their shape extending towards the front, giving the central cusps a recumbent attitude.

53The anterior root is clearly protrusive (fig. 12 n° 17 and 18).
- t1 is slightly remote in relation to t2-t3, but the anterior extension of t2 considerably widens the syncline angle separating t1 from t2;
- the posterior cingulum, in most cases, forms an angular and bulging protuberance on the side of t8;
- t7 is strong and linked to t8 by a strip of enamel. After a period of wear, it is linked to t4.

54The clearest difference is in the L/W ratio (1.71), which denotes the tapered morphology of this molar.

55M2- (fig. 12 n° 5). A single molar was found but its shape is typical: the total absence of t3 is characteristic of the form of the western species. A present-day series from Manchuria lent by the Museum of Basel (fig. 14 n° 6) presents a tiny t3. The dispersion diagram places it outside the overall cloud of points (fig. 15).

56M1 - (fig. 12 n° 6 to 8). Their morphological specificity is the absence of the external cingulate margin. Like for the present-day forms, the presence of c4 is inconsistent and can only be represented by a slight tuberosity contiguous to tE.
- c1 is isolated and is only linked to t4 after extensive wear;
- tA and tB converge but their union is never prolonged by even an outline of a longitudinal median crest;
- the tC – tD pair also converges;
- the posterior cingulum forms a transverse crest between tA and tB. Unlike for present-day forms, there is no protrusion at the back of the molar;
- the median anterior tubercle (tma) is constantly small in size.

57Comments: The difference in the silhouette between the upper and lower molars is surprising. The upper molars are slender and the lower molars are chunky. This observation is clear in the L/W ratios: 1.71 for the upper molars; 1.56 for the lower molars.

58M2- Two typical molars were identified (fig. 12 n° 11 and 12). Like the M1, they are differentiated by the absence of the external lateral margin completed by the presence of a well-defined c1.
- tE is reduced and does not extend along the external edge of the molar;
- like on the M1, the posterior cingulum develops into a transverse crest between tA and tB;
- like for the M1 again, the tubercles match up to form two chevrons with no trace of a longitudinal median crest;

59The position of these molars in the dispersion diagram (fig. 15) shows that we cannot distinguish them on the basis of their dimensions.

60Compared to the present-day forms (fig. 14), we observe that c1 is much more developed on the latter. It forms a slightly more reduced tubercle than the main cusps, almost as significant for the M1 c1. It remains separate from the neighbouring cA. The Manchuria type does not bear a c1; just a slight bulge.

Apodemus agrarius iguensis n.ssp.

61In spite of reduced numbers, the morphometric difference observed on the Apodemus agrarius molars incites us to create a subspecies with a biochronological vocation allowing us to position the population from Igue des Rameaux in an evolutionary framework to be broadened by future discoveries.

62Family: Muridae Gray, 1821
Genus: Apodemus Kaup, 1826
Sub-genus: Apodemus Kaup, 1829
Species: Apodemus agrarius Pallas, 1778
Sub-species: Apodemus agrarius iguensis n.ssp.

63HOLOTYPUS: First lower right molar n° SIR. 158-12 (fig. 12 n° 8) (Length x width = 172 x 110 mm/100).

64DERIVATIO NOMINIS: derived from the local term “igue” designating the form of the karstic shaft opening in the lapiaz (synonyms: sink hole, joint, fissure etc.).

65DIAGNOSIS: first Apodemus type lower molar presenting typical characteristics of the Apodemus agrarius species: absence of the external cingulate margin. Posterior cingulum reduced to a ridge of enamel set between the two posterior tubercles, but does not exceed the back of the molar.

66DIFFERENTIAL DIAGNOSIS: It differs from the type species by its well-developed width, giving the molar a chunky appearance with a length/width ratio of 1.56 (n= 4; max = 1.59; min = 1.55). For Apodemus agrarius, this is the lowest ratio for this type of molar. For present-day forms from Niedersachsen, the ratio is 1.70 for a size equivalent to 1.74. For A. agrarius iguensis n.ssp, this situation is more paradoxical as the silhouette of the upper molars is slender, with a L/W ratio of 1.71, which is similar to that of present-day forms (L/W = 1.74).

67The M2 adopt the same profile, as the cp is inserted between the two tubercles tA and tB. This layout “shortens” the molar. tE tends to lengthen narrowly against tC but the main trait is the absence of lateral cingulate margin, like for the M1.

68The M2 is very characteristic due to the absence or extreme reduction of t3, in a very low position which only highlights it after long wear (cf. the specimen from Siaolin-Kirin, fig. 14 n° 6).

69LOCUS TYPICUS: Site of Igue des Rameaux; commune of Saint-Antonin-Noble-Val (Tarn-et-Garonne, Quercy, France).

70STRATUM TYPICUM: Zone G2M, layer 60, attributed to the Middle Pleistocene. The associated fauna is illustrated in table 2 and denotes a wooded steppe environment under a cool, humid and temperate climate, similar to the prevailing conditions in present-day Eastern Europe (Jeannet 2005).

71PARATYPUS: The other molars likely to belong to this species contributed to the general description of the morphotypes and their biometric data are presented in tables 4 and 5. They can also be identified on the dispersion diagram (fig. 15).

72The comparative elements entrusted to us by the Museum of Bonn probably come from the same enriched Niedersachsen population studied by J.P. Aguilar et al. (2008) (tabl. 7). Here, we use the results adding the dimensions from the Siaolin-Kirin series lent by the Museum of Basel (n°9070) (fig. 14).

Table 7 - Apodemus agrarius.

Table 7 - Apodemus agrarius.

Origin: Bouziès (Lot), Tardiglacial period; Niedersachsen (Germany), sub-contemporaneous.

Data from Aguilar and al. (2008) for comparison.

73We regret not having more abundant series in order to bring to light with more reliability a certain evolutionary cline that seems to be outlined on the basis of a biometric gradient. In fact, through the examination of the simple elongation ratio, we observe the growth of the index, passing from 1.56 at Igue des Rameaux (Middle Pleistocene) to 1.65 for Bouziès (Lot) (Final Pleistocene; towards 17ka B.C.) and 1.69 for the present-day forms from Niedersachsen. On the other hand, this same ratio is relatively stable for the upper molars (1.71 for the population from Rameaux, 1.65 for Bouziès and 1.71 for Niedersachsen).

74Apodemus uralensis Pallas, 1771 or Apodemus microps Kratochvil and Rosicki, 1952.

75Apodemus maastrichtiensis Kolfschoten, 1985. (fig. 12 and 13; tabl. 8)

76The Ural field mouse is a very small rodent living in the steppes of Central and Eastern Europe and the south of the Ural. Its double identity comes from the fact that certain authors distinguish two different species. It seeks out high grasses, shrubs and coppice. It is not abundant but is nonetheless present in many Pleistocene sites but due to the small size of its molars, it often goes through the sieve mesh or is mixed up with the slightly smaller Eurasian harvest mouse (Micromys minutus). It is however easy to differentiate their first molars. The M1 of the Eurasian harvest mouse has five roots whereas the Ural field mouse only has three, of which the two lateral roots are germinated and their liaison is highlighted by a rather shallow longitudinal furrow. The M1 of the Eurasian harvest mouse bears three roots, while the Ural field mouse only has two.

77M1 – The only two M1 at our disposal have a very stocky aspect, due to the symmetric position of t1 and t3, which are not very far from each other.

78The rather sudden straightening of their posterior wall creates a wide space between the t1-t3 and t4-t6 pair. The posterior rosette is clearly serrated by very straightened and not very convergent peripheral cusps due to their constrained junction. The posterior cingulum is well defined and forms a posterior-external bulge.

79Morphologically, it seems that the specimens at our disposal are similar to the present-day forms from the Czech Republic described by L. Pasquier (1974 p.111), or at least more evolved than the ancestral forms. The only lateral root with an oval section bearing early molars tends to flatten, developing a bulge at the anterior and posterior ends, announcing a hypothetical bifid morphotype. The reduced and elongated t7 on the early forms has become a cusp, although it is still distant from t8, to which it is linked by a strip of enamel. The posterior cingulum, imperceptible on archaic types, is particularly well developed here and independent of t9. The t1 is in a slightly more distant position in relation to t2-t3 than on the previously described forms. Size increases slightly but remains lower than that of A. sylvaticus and within the variation domain of A. uralensis.

80The comparison with Apodemus maastrichtiensis is very limited as we only have eight molars, two of which belong to the same individual (fig. 13 n°17).

81On the M1, the straightening of the cusps is obvious on the anterior pair. The tma is generally small in size or absent. The isthmus joining the anterior set to the median pair forms a very tight narrowing and can be absent, or only appear after very pronounced wear. The length of the external cingulate margin is very variable and can be limited to a simple strip or be reinforced by two well-developed conules.

82The longitudinal median crest tending to link the posterior part to the median pair is slightly outlined as tB advances and is clearly delimited from tA, with which it remains slightly joined by its anterior-internal wall. The c1 is relatively well developed, but smaller than tA, to which it is increasingly attached during the course of wear. The posterior cingulum (cp) is generally rounded and goes over the tA-tB tangent towards the back.

83The only M2 at our disposal has no external cingulate margin and no c1. tE is just a small bulge at the base of tC with no posterior stretching. It is separated from tE by a short and narrow furrow. The form, volume, and position of the posterior cingulum are identical to those of the M1.

84All these specific characteristics are described in the detailed description by T. van Kolfschoten (1985) on the molars from Belvédère at Maastricht (Netherlands), which he used to create Apodemus maastrichtiensis. In spite of the small number of specimens at Igue des Rameaux, it is clear that our small field mouse belongs to this species.

Table 8 - Apodemus maastrichtiensis. Data from Kolfschoten (1987). Sites of Belvedere 3 and 4 (Maastricht, The Netherlands), Saalian period.

Table 8 - Apodemus maastrichtiensis. Data from Kolfschoten (1987). Sites of Belvedere 3 and 4 (Maastricht, The Netherlands), Saalian period.

Correlation and test (tabl. 6 and tabl. 9 to 12)

85Table 6 summarizes the results of the calculations of the correlation coefficient (r) and the test (t) on the averages when there is a sufficient number of elements. There are insufficient numbers of M2 in particular, to carry out these calculations and obtain reliable results.

86The correlation coefficient enables us to verify the homogeneity of a population. Experience shows that for field mice, the correlation between the length (L) and the width (W) of the (upper and lower) M1 is around 0.85 (reminder: a perfect correlation would give a coefficient of 1.00). When it is too low (below 50), it indicates the lack of homogeneity or insufficient data; when it is negative, it can translate an inverse evolution or a total contradiction. However, the association of members of two populations can present a high correlation without necessarily belonging to the same species. Therefore, it is essential to complete the examination by a “t” test, comparing the average lengths, widths and their association as a product or ratio.

87Without examining these results in detail, it is interesting to comment the main trends (as an example), that is, those of morphologically similar species, such as A. flavicollis (AF), A. sylvaticus (AS), and A. agrarius (AA) or A. microps/ maastrichtiensis/uralensis (AM). A parallel reading of table 6 and the dispersion diagram (fig. 9) facilitates this interpretation.

88The M1 of A. flavicollis (AF) and A. agrarius (AA) populations are very homogeneous (r= 0.93 and 0.98), and A. sylvaticus (AS) a little less so (0.79). When we associate populations (AF + AA + AS), the correlation falls to 0.53. On the M1, the differences increase and fall respectively to 0.83 and 0.85 to 0.15 collectively.

89If doubt persists as to the independence of the link, the “t” test confirms the integrity of each group, indicating a highly significant difference between AF and AS, although it is not significant for the widths, and conversely, a significant difference between the widths of AF and AA for a non-significant difference between the lengths.

90The M1 widths always present significant differences, even between Apodemus flavicollis and A. sylvaticus.

91For the combined L and W values, as products or ratios, only the elongation ratio L/W of the M1 enables us to separate the three species (AF, AS and AA). The results on the M1 remain random.

92Due to an insufficient quantity of material, we will simply present a list of several murid species and the data at our disposal.

Table 9- L’Igue des Rameaux. Biometric data of Apodemus M1 and comparative statistical interspecific analyses.

Table 9- L’Igue des Rameaux. Biometric data of Apodemus M1 and comparative statistical interspecific analyses.

Table 10 – L’Igue des Rameaux. Biometric data of Apodemus M1 and comparative statistical interspecific analyses.

Table 10 – L’Igue des Rameaux. Biometric data of Apodemus M1 and comparative statistical interspecific analyses.

Tableau 11 – L’Igue des Rameaux. Biometric data of Apodemus M2 and comparative statistical interspecific analyses.

Tableau 11 – L’Igue des Rameaux. Biometric data of Apodemus M2 and comparative statistical interspecific analyses.

Small murids of diverse origins

93It is not rare to find isolated molars from very smaIl murids The morphological or biometric variability of the members of the family often makes their determination uncertain. Those from other sites present similarities with Apodemus maastrichtiensis Kolfschoten, 1985 and A. agrarius (Pallas, 1778) from Igue des Rameaux, inciting us to compare them. As the unicity of each morphotype rules out attribution to a given species or the creation of a new taxon, we did not compare the morphotypes from Rameaux to discoveries with insufficient numbers.

94As we cannot add these sparse and unrepresentative elements to this study, we are adding a table (tabl. 13) for information, but we regret the fact that we cannot establish a phyletic and/or biochronological table for this group of murids.

Tableau 12– L’Igue des Rameaux. Biometric data of Apodemus M2 and comparative statistical interspecific analyses.

Tableau 12– L’Igue des Rameaux. Biometric data of Apodemus M2 and comparative statistical interspecific analyses.

Table 13 – Comparative data. Location and measurements of field mice from European sites of various periods. The numbers in the first column allow to connect each site to its own values.

Table 13 – Comparative data. Location and measurements of field mice from European sites of various periods. The numbers in the first column allow to connect each site to its own values.

Discussion and conclusion

95Based on rather sparse but well-characterized populations, it is thus possible to determine molars dispersed in time and space.

96These species evolve and are transformed over time in response to climatic and environmental demands. This influences metric and morphological parameters. Which corporal characteristics can be associated with specific thermal, hygrometric or orographic influences?

97It is tempting to seek evolutionary or biological criteria in each morphological or biometric variation. It is also tempting to create a new taxon to concretize these differences on statistical or formal bases without identifying the natural reasons for these characteristics. Adaptation to the environment is generally the cause of morphological and metric modifications.

98Which criteria allow for the creation of a sub-species? The number of individuals affected by a distinctive sign? The constant and repetitive morphological and biometric differences guarantee the reality of a morphotype, assimilated to a sub-species. These basic criteria are perfectly clear for the molars of Apodemus agrarius iguensis n. ssp, in spite of the limited number of samples.

99However, it seems appropriate, or at least of interest, to outline a lineage with sporadic or isolated elements (for want of better) likely to bear proportional modern and archaic features. In this regard, murids have huge potential considering the number of landmarks observed on their molars and the available variation resources. But statistics have to be taken into account and therefore from Parapodemus coronensis during the Lower Pliocene to present-day Apodemus microps, there is only one genealogical step: Apodemus maastrichtiensis.

100We do not wish to reform the principles of research and analysis, but it would be interesting, or even instructive, to be able to insert isolated morphotypes, deemed to be unclassifiable, into recognized populations and to point out their individuality. It is regrettable that even relative and random chronology must be backed up by other phyletic chains or other disciplines.

Top of page

Bibliography

Bibliographic references

AGUILAR J.P., PÉLISSIER T., SIGÉ B. et MICHAUX J. 2008 - Occurrence of the Strip Field Mouse lineage (Apodemusagrarius (Pallas, 1771) ; Rodentia, Mammalia) in the Late Pleistocene of Southwestern France. C. R. Palevol, 7: 217-225.

BARCIOVA L. et MACHOLAN M. 2006 – Morphometricc study of two species of wood mice Apodemussylvaticus andA. flavicollis ( Rodentia : Muridae) : traditional and geometric morphometric approach. Actatheriologica51 (1): 15-27.

BARTOLOMEI G. 1964 – Mammiferi di breccepleistoceniche dei ColliBerici (Vi,cenza). MemoriedelMuseoCivico di StoriaNaturale, Verona,XII: 221-290.

BÖHME von W. 1978 – Apodemusagrarius (Pallas,1771)- Brandmaus. InNiethammerJ. und Krapp F. Handbuch der SäugetiereEuropas, Band 1, p. 368-381. Wiesbaden.

BRUGAL J. Ph. 1981 – Balaruc VII (Sète, Hérault). Un nouveau remplissage de fissure de la fin du Pléistocène moyen. Quaternaria, Roma, XXIII : 99-141.

CORBET G.B. 1978 – The Mammals of the Palearctic region: a taxonomic review. British Museum (Natural History), Cornell University Press. London : 313 p.

DESCLAUX E., DEFLEUR A. 1997 – Étude préliminaire des micromammifères de la Baume Moula-Guercy à Soyons (Ardèche, France). Systématique, biostratigraphie et paléoécologie. Quaternaire, 8 (2-3): 213-223.

DIETRICH W. D. et MAUL L. 1984 - Skelettreste von nagetieren (Rodentia, Mammalia) aus dem fossilen Tierbautensystem von Pisede bei Malchin. Teil 1. Taxonomische und biometrische Kennzeichnung des Fungutes. Wissenscchaftliche Zeitschrift der Humboldt-Universität zu Berlin, Math.-Nat.,XXXII (6) :729-743.

HEINRICH W.D. 1990 – Some aspects of evolution and biostratigraphy of Arvicola (Mammalia, Rodentia) in the central europeanpleistocene. Int. Symp. Evol. Phyl. Biiostr. Arvicolids .O. Fejfar& W. D. Heinrich Edit. : 165-182, Praha 1990.

JAMMOT D. 1973 - Les insectivores (Mammalia, Rodentia) du gisement Pléistocène moyen des Abîmes de La Fage à Noailles (Corrèze).Nouv. Arch. Muséumd’Hist. Nat. Lyon, II: 41-51.

JANGJOO M. 2010 – Geometric morphometric analysis of the second upper molar of the genus Apodemus(Muridae :Rodentia) in Northern Iran.Iranian Journal of Animal Biosystemaics, 6(2): 33-44.

JANZEKOVIC F. et B. KRYSTUFEK 2004 - Geometric morphometry of the upper molars in European wood mice Apodemus. Folia Zool., 53 (1): 47-55.

JAUBERT J., KERVAZO B.,BRUGAL J.P., FALGUERES C., JEANNET M., LOUCHARD A., MARTIN H., MAKSUD F., MOURRE V., QUINIF Y. 2005 - La séquence Pléistocène moyen de Coudoulous I (Lot). Bilan pluridisciplinaire. In : Les premiers peuplements en Europe: Données récentes sur les modalités de peuplement et sur le cadre chronostratigraphique, géologique et paléogéographique des industries du Paléolithique ancien et moyen en Europe, N.Molines, M.H.Moncel, J.L.Monnier (éds), (Actes du Coll.intern.Rennes, 22-25 septembre 2003), Oxford. British Archaeological Reports, International Series 1364: 237-251

JEANNET M. 1977 - Recherche de microfaune à la grotte de l’Escale, Saint-Estève-Janson (Bouches du Rhône ; campagne 1977).Nouv. Arch. Mus. Hist. Nat Lyon (15 suppl.), 53.

JEANNET M. 2000 - Gruta da Figueira Brava : Les Rongeurs.Memorias da academiadasciencias de Lisboa, T. XXXVIII, 180-243.

JEANNET M. 2005 – La microfaune de l’Igue des Rameaux à Saint-Antonin-Noble-Val (Tarn-et-Garonne, France). Essai de Biostratigraphie. Préhistoire du Sud-Ouest, 2 : 109-125.

JEANNET M. 2010 – L’Écologie Quantifiée : essai de description de l’environnement continental à l’aide des microvertébrés. Préhistoires méditerranéennes. I varia :1-26, et annexes. (http ://pm.revues.org/index492.html).

JEANNET M., BRUGAL J.Ph., JARRY M. 2013 - Microfaune et paléoclimatologie dans le Pléistocène moyen et supérieur du Quercy. Essai d‘écologie quantifiée. In : Modalité d’occupation et exploitation des milieux au Paléolithique dans le Sud-Ouest de la France : l’exemple du Quercy, Jarry M., Brugal J.P., Ferrier C. (dir), Paleo, supplément n°4 : 107-143(Actes de la session C67, XVème Congrès mondial de l’UISPP, Lisbonne, sept. 2006).

KOENIGSWALD W. von 1972- Sudmer-Berg-2. Eine Fauna des frühen Mittelpleistozäns aus des Harz. N. Jb. Geol. Paläont. Abh. Stuttgart, 141-2 :194-221.

KOLFSSCHOTEN T. van 1985 – The Middle Pleistocene (Saalian) and Late Pleistocene (Weischelian) Mammal Faunas from Maastrischt-Belvedere (Southern Lim burg, The Netherlands), InMaastrischt-Belvedere : Stratigraphy, Paleoenvironment and Archaeology of the Middle and Late Pleistocene deposits.MededilingenRijksgeologischedienst.,Gravenhage, 39 (1) : 45-74.

KOWALSKI K. 1956 – Insectivores, Bats and Rodents from the Early Pleistocene bone breccia of Podselice near Kroczyce (Poland). ActaPalaontologicaPolonica, 1(4) : 331-394.

KRATOCHVIL J. et ROSICKY B. 1952 – K Bionomii a TaxonomiiMysiRoduApodemus, Zijicich v Ceskoslovensku. Zool. Listy,I: 57-70

KRETZOI M. 1956 – Die Altpleistozänen Wirbeltierfaunen des Villanyer Gebirges. Geologica Hungarica S. Paleontologica. Budapest, 27 : 1-264.

MALEZ M. et RABEDER G. 1984 - Neues Fundmaterial von Kleinsäugern aus der altpleistozänen Spaltenfüllung Podumci1. In : Norddalmatien (Kroatien, Jugoslawien).Beit. Paläont. Österr., Wien, 11 : 439-510.

MARTINSON D.C., PISIAS N.G., HAYS J.D., IMBRIE J., MOORE T.C. et SHAKLETON N.J. 1987 – Age dating and the orbital theory of a high resolution 0 to 300 000 years chronostratigraphy. Quaternary research, 27 : 1-29.

MAUL L. C. 1990 - Die Müridenreste (Mammalia, Rodentia) aus der unterppleistozänen Fundstellle Voigstedt (Bezirk Halle, DDR). Quartärpaläontologie, Berlin, 8: 193-204.

MEULEN A. J. 1973 – Middle Peistocene Smaller Mammals from the Monte Peglia, (Orvieto, Italy) with special reference to the phylogeny of Microtus (Arvicolidae, Rodentia). Quaternaria, Roma, XVII : 1-144.

MICHAUX J. 1971 – Muridae (Rodentia) néogènes d’Europe sud-occidentale. Évolution et rapports avec les formes actuelles. Paleobiologie continentale, Montpellier, II (1) :1-67.

MILLER G .S. 1912 – Catalogue of the mammals of Western Europe. British Museum (Natural Hitory), London, 1019 p.

PASQUIER L. 1974 – Dynamique évolutive d’un sous-Genre de Muridae, Apodemus (Sylvaemus). Étude biométrique des caractères dentaires de populations fossiles et actuelles d’Europe occidentales. Thèse de Doctorat de Spécialité (Paléontologie). Université de Montpellier. 183 p.

POPOV V. V. 1988 – Middle Pleistocene Small Mammals (Mammalia, Insectivora, Lagomorpha, Rodentia) from Varbeshnitsa (Bulgarie). Actazoologicacracov.,31(5) : 193-234.

RENAUD S. 2004 – First upper molar and mandible shape of wood mice (Apodemussylvaticus) from Northern Germany: ageing, habitat and insularity. MammalianBiology, 70 (2005) 3 : 157-170.

ROUZAUD, F., SOULIER M., BRUGAL J.-Ph., JAUBERT J. 1990 – L’Igue des Rameaux (Saint-Antonin-Noble-Val, Tarn et Garonne). Un nouveau gisement du Pléistocène moyen. Premiers résultats. Paleo, n°2 : 89-106.

SALA B. 1974 – Novidati su Apodemusagrarius (Pallas) delFriuli. Boll. Soc. Naturalisti « Silvia Zenari », Pordedone, Anno V-N. 1-2: 40-50.

SCHAUB S. 1938 – Tertiäre und QuartäreMurinae. Abhandlungen der Sweizerischer Paläontologischen Gesellschaft., Basel, LXI(1): 1-38.

STEINER H.M. von 1978 – Apodemus microps Kratochvil und Rosicki, 1952. In Niethammer J., Krapp F. (eds), Handbuch der Saugetiere Europas, Wiesbaden, Vol. 1 : 359-367.

STORCH G. 1984 – The Neogene mammalian faunas of Ertemte and Harr Obo in Inner Mongolia (Nei Mongol), China- 7 Muridae (Rodentia). Senckenbergiana lethaea, 67(5-6) : 401-431.

STORCH G. von et LUTT O. 1989 – Artstatus der Alpenwaldmaus, Apodemus alpicola Heinrich, 1952. Z. saugertierrkunde,54: 337-346.

Top of page

List of illustrations

Title Figure 1 – Location of l’Igue des Rameaux at Saint-Antonin-Noble-Val (Tarn-et-Garonne, France) in Quercy (south-western France).
Credits IGN map 2012 and Géoportail
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-1.jpg
File image/jpeg, 300k
Title Figure 2 – L’Igue des Rameaux. Topographic map of the site (2a) and longitudinal profile (2b) showing the excavated areas.
Credits From ROUZAUD and al., 1990.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-2.jpg
File image/jpeg, 232k
Title Table 1 – L’Igue des Rameaux. Suammarized plan of the site showing sampling zones. Each square indicates the layer and the number of samples. Areas B, C and D constitute the upper zone; E, F and G, the lower zone.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-3.png
File image/png, 51k
Title Table 2 - L’Igue des Rameaux. Distribution of species by square and layers in the upper and lower zones, pointing out the similitude of assemblages between the two sections despite a lower abundance and the lack of Amphibians and the low representation of Reptiles in the upper zone.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-4.png
File image/png, 110k
Title Figure 3 –L’Igue des Rameaux. Pie charts showing microfaunal distribution between upper and lower zones (in % per species).
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-5.jpg
File image/jpeg, 132k
Title Figure 4 - L’Igue des Rameaux. Temperature diagram.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-6.jpg
File image/jpeg, 208k
Title Figure 5 - L’Igue des Rameaux. Precipitation diagram.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-7.jpg
File image/jpeg, 236k
Title Figure 6 - L’Igue des Rameaux. Atmosphere diagram.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-8.jpg
File image/jpeg, 272k
Title Figure 7 - L’Igue des Rameaux. Vegetation diagram.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-9.jpg
File image/jpeg, 288k
Title Figure 8 - L’Igue des Rameaux. Ground hygrometry.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-10.jpg
File image/jpeg, 296k
Title Table 3 – European Apodemus. Data table for dispersal diagrams (fig. 9 and 10).
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-11.png
File image/png, 339k
Title Table 4 - L’Igue des Rameaux. Measurements, type and localization of upper molars of Apodemus. The sketched elements are indicated in the last column.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-12.png
File image/png, 174k
Title Tableau 5 - L’Igue des Rameaux. Measurements, type and localization of lower molars of Apodemus. The sketched elements are indicated in the last column.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-13.png
File image/png, 208k
Title Figure 9 - European Apodemus. Dispersal diagram of M1 from some European populations, including all types, periods and locations (see tabl. 3)
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-14.png
File image/png, 45k
Title Figure 10 - Dispersal diagram of M1 from some European Apodemus populations, including all types, periods and locations (see tabl.3): LM1= length of the first lower molar; W/1 = width of the first lower molar. Except the small types, the diagram points out the difficulties, or even the impossibility, of discriminating species using only biometric data.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-15.png
File image/png, 41k
Title Figure11 – Descriptive schema of the various structures of the Apodemus mastication table (after Michaux 1971)
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-16.jpg
File image/jpeg, 168k
Title Figure 12 - Apodemus agrarius (n° 1, 2, 3, 4, 5, 6, 7, 8, 11, 12) ; Apodemus flavicollis n° 9, 13, 16) ; Apodemus sylvaticus (n° 10, 15, 17, 18) ; Apodemus uralensis (n° 14)
Caption 1 =M1G ; 2 =M1D ; 3 =M1D ; 4 =M1D ; 5 =M2G ; 6 =M1G ; 7 =M1G ; 8 =M1D ; 9 =M2G ; 10 =M2G ; 11 =M2G ; 12 =M2D ; 13 =M2G ; 14 = M1D ; 1M2G ; 16M2D ; 17-18 =M1D
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-17.jpg
File image/jpeg, 472k
Title Figure 13 - Apodemus flavicollis (n° 1, 2, 3, 4, 9, 10, 11, 13, 14, 16, 18 ) ; Apodemus sylvaticus(n° 5, 6, 7, 8, 12, 15, 21) ; Apodemus uralensis (n° 17, 19, 20).
Caption 1 = M1G ; 2 = M1G ; 3 = M1D ; 4 = M1D ; 5 = M1D ; 6 =M1D ; 7 = M1D ; 8 = M2D ; 9 = M2G ; 10 = M2G ; 11 = M2D ; 12 = M2D ; 13 = M2D ; 14 = M1G ; 15 =M1G ; 16 = M1D ; 17 = M1-2G ; 18 = M3G ; 19 = M1G ; 20 = M1G ; 21 = M1G.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-18.jpg
File image/jpeg, 276k
Title Figure 14 - M1/-M2/ and M/1-M/2 of Apodemus agrarius Pallas, 1771.
Caption N° 1 to 5: Bonn Museum collections. N° 6: Basel Museum collections.Origin: n°1: (n° 55/03 ♀) –Niedersachsen-Hörden a Hartz (Germany).N°2: (N° 56/7 ♂) – Niedersachsen Willershauser a Harz (Germany).N°3: (N° 55/18 ♂) – Niedersachsen Willershauser a Harz (Germany). N°4: (N° 54/15 ♀)- Niedersachsen Willershauser a Harz (Germany).N°5: (N° 56/3 ♂) - Niedersachsen Willershauser a Harz (Germany).N° 6: (N° 97/70 ♀) – Siaolin-Kirin; Manchou du Kuo (Manchuria).Note: on the M1/ cusp lengthening; on the M2/ the absence of t3 (except n° 6); on M/1 and M/2 the absence of the external cingulate margin
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-19.jpg
File image/jpeg, 224k
Title Figure 15 – L’Igue des Rameaux – Molar dispersal diagrams of the various Apodemus species collected. The M2 cannot be specifically separated by their dimensions but by their structures. LM1/=length of the first upper molar; WM1/=width of the first upper molar, etc.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-20.jpg
File image/jpeg, 188k
Title Table 6 - L’Igue des Rameaux. Statistical comparison between the molars of Apodemus.
Caption AA = Apodemus agrarius; AF = A. flavicollis; AM = A. microps; AS = A. sylvaticus.In normal populations, the correlation coefficient (r) is close to 0.85. A negative correlation indicates that one parameter evolves in the opposite direction to the other (cf. comparison between the product and the ratio of length and width of m1).
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-21.png
File image/png, 57k
Title Table 7 - Apodemus agrarius.
Caption Origin: Bouziès (Lot), Tardiglacial period; Niedersachsen (Germany), sub-contemporaneous.
Credits Data from Aguilar and al. (2008) for comparison.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-22.png
File image/png, 29k
Title Table 8 - Apodemus maastrichtiensis. Data from Kolfschoten (1987). Sites of Belvedere 3 and 4 (Maastricht, The Netherlands), Saalian period.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-23.png
File image/png, 23k
Title Table 9- L’Igue des Rameaux. Biometric data of Apodemus M1 and comparative statistical interspecific analyses.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-24.png
File image/png, 265k
Title Table 10 – L’Igue des Rameaux. Biometric data of Apodemus M1 and comparative statistical interspecific analyses.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-25.png
File image/png, 86k
Title Tableau 11 – L’Igue des Rameaux. Biometric data of Apodemus M2 and comparative statistical interspecific analyses.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-26.png
File image/png, 230k
Title Tableau 12– L’Igue des Rameaux. Biometric data of Apodemus M2 and comparative statistical interspecific analyses.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-27.png
File image/png, 293k
Title Table 13 – Comparative data. Location and measurements of field mice from European sites of various periods. The numbers in the first column allow to connect each site to its own values.
URL http://journals.openedition.org/paleo/docannexe/image/3264/img-28.png
File image/png, 166k
Top of page

References

Bibliographical reference

Marcel Jeannet and Pierre Mein, « The middle Pleistocene Muridae (Mammalia, Rodentia) from l’Igue des Rameaux (Tarn-et-Garonne, France) », PALEO, 27 | 2016, 177-205.

Electronic reference

Marcel Jeannet and Pierre Mein, « The middle Pleistocene Muridae (Mammalia, Rodentia) from l’Igue des Rameaux (Tarn-et-Garonne, France) », PALEO [Online], 27 | 2016, Online since 01 June 2018, connection on 19 November 2018. URL : http://journals.openedition.org/paleo/3264

Top of page