Temporal evolution of Arvicola molar morphology

2D geometric morphometrics reveals temporal patterns in the evolution of modern and Pleistocene Arvicola (Rodentia, Cricetidae) in Europe

Emmanuel Desclaux, Amanda Lindahl, Daniela C. Kalthoff, Oleksandr Kovalchuk, Alejandro Centeno-Cuadros, Jacinto Román, Mikel Arlegi, Andion Arteaga-Brieba, Zoltán Barkaszi, Jessica Cohen, Alban Defleur, Mónica Fernández-García, Damien Flas, Karl Frafjord, Monika Vlasta Knul, Arturo de Lombera-Hermida, Ahmad Mahmoudi, Boyan Milchev, Adam Nadachowski, Piroska Pazonyi, Iván Rey-Rodríguez, Xosé Pedro Rodríguez-Álvarez, Vincent Bonhomme, Lové Dalén, and John R. Stewart

Article number: 29.3.a35
https://doi.org/10.26879/1683
Copyright Paleontological Society, September 2026

Author biographies
Plain-language and multi-lingual abstracts
PDF version
Appendices

Submission: 2 March 2026. Acceptance: 17 August 2026.

ABSTRACT

This study applies a comprehensive two-dimensional (2D) geometric morphometric framework to investigate dental morphological variation in modern and Pleistocene populations of the genus Arvicola. A reference dataset of 390 first lower molars (m/1) representing four extant species (A. amphibius, A. sapidus, A. italicus, and A. persicus) was analysed using Elliptic Fourier Analysis combined with multivariate statistics (PCA, MANOVA, LDA, CDA, Random Forest, and Mahalanobis distances). All modern species of Arvicola exhibit highly significant morphological differences, demonstrating that m/1 outlines shape reliably discriminates species-specific variation. This validated classification framework was applied to fossil assemblages from five well-dated localities spanning the late Middle Pleistocene to Upper Pleistocene (MIS 6 to MIS 3). The MIS 6 assemblage from Lazaret Cave displays a heterogenous and intermediate morphospace distribution, reflecting the presence of archaic lineages attributed to Arvicola ex gr. cantianus/amphibius and Arvicola ex gr. cantianus/sapidus. By MIS 5.5 (Baume Moula-Guercy), Arvicola amphibius -like morphologies dominate, although several specimens retain plesiomorphic traits. In contrast, MIS 3 assemblages (Abauntz Cave, Cova Eirós, and Trou Al'Wesse) exhibit predominantly modern morphologies, with high classification agreement (up to 100%) and clear species-level assignment. A temporal trend emerges in which discrimination accuracy and morphological distinctiveness increase from the late Middle Pleistocene to MIS 3, documenting the evolutionary transition from archaic Arvicola forms to fully differentiated modern species. The results further suggest a complex biogeographic history of Arvicola sapidus, involving multiple Middle and early Late Pleistocene dispersal events into Western Europe. Overall, this study demonstrates that 2D outline-based geometric morphometrics of the m/1 is a robust and informative tool for resolving Arvicola systematics and for reconstructing evolutionary and palaeoenvironmental patterns in the Pleistocene of Europe.

Emmanuel Desclaux. Laboratoire départemental de Préhistoire du Lazaret, 33 bis boulevard Franck Pilatte, 06300 Nice, France and CEPAM - UMR 7264 CNRS - Université Côte d’Azur, avenue des Diables Bleus, 06300 Nice, France.(Corresponding author) [email protected]
Amanda Lindahl. Centre for Palaeogenetics, Stockholm, Frescativägen 8, 106 91
Stockholm, Sweden and Department of Zoology, Stockholm University, 106 91 Stockholm, Sweden. [email protected]
Daniela C. Kalthoff. Department of Zoology, Swedish Museum of Natural History, Frescativägen 40 and 48, 104 05 Stockholm, Sweden. [email protected]
Oleksandr Kovalchuk. National Museum of Natural History of the National Academy of Sciences of Ukraine, Bohdana Khmelnytskoho 15, Kyiv, 01054 Ukraine and Department of Palaeozoology, Faculty of Biological Sciences, University of Wrocław, Sienkiewicza 21, 50-335 Wrocław, Poland. [email protected]
Alejandro Centeno-Cuadros. Instituto de Investigación Marina (INMAR). Faculty of Marine and Environmental Sciences. University of Cadiz, 11519 Cadiz, Spain. [email protected]
Jacinto Román. Biología de la Conservación. Estación Biológica de Doñana-CSIC, Avda Americo Vespucio n°26, 41092 Sevilla, Spain. [email protected]
Mikel Arlegi. HNHP - UMR 7194 CNRS, Museum National d’Histoire Naturelle, Paris, 75005, France and University of Cambridge, CB2 1TN Cambridge, United Kingdom. [email protected]
Andion Arteaga-Brieba. Centro Nacional de Investigación en Evolución Humana (CENIEH), P.° Sierra de Atapuerca, Burgos, 09002, Spain and Institut Català de Paleoecologia Humana i Evolució Social (IPHES-CERCA), Zona Educacional, 430007 Tarragona, Spain. [email protected]
Zoltán Barkaszi. National Museum of Natural History of the National Academy of Sciences of Ukraine, Bohdana Khmelnytskoho 15, 01054 Kyiv, Ukraine and Department of Agricultural Sciences, John von Neumann University, Izsáki út 10, 6000 Kecskemét, Hungary. [email protected]
Jessica Cohen. Laboratoire départemental de Préhistoire du Lazaret, 33 bis boulevard Franck Pilatte, 06300 Nice, France. [email protected]
Alban Defleur. Institut Català de Paleoecologia Humana i Evolució Social (IPHES-CERCA), Zona Educacional, 430007 Tarragona, Spain. [email protected]
Mónica Fernández-García. Institut Català de Paleoecologia Humana i Evolució Social (IPHES-CERCA), Zona Educacional, 430007Tarragona, Spain and University of Lancashire, Fylde Rd., PR1 2HE Preston, United Kingdom. [email protected]
Damien Flas. Service de Préhistoire, UR "Art, Archéologie et Patrimoine", Université de Liège, Allée du Six-Août, 10, 4000 Liège, Belgium. [email protected]
Karl Frafjord. Tromsø Museum, UiT the Arctic University of Norway, Lars Thørings veg 10, 9006Tromsø, Norway. [email protected]
Monika Vlasta Knul. University of Winchester, Sparkford Rd, SO22 4NR Winchester, United Kingdom. [email protected]
Arturo de Lombera-Hermida. Departamento de Historia, Facultad de Filosofía y Letras, Universidad de Oviedo, C/ del Teniente Alfonso Martínez, s/n, 33011 Oviedo, Spain. [email protected]
Ahmad Mahmoudi. Department of Biology, Faculty of Science, Urmia University, University blvd Urmia, Iran. [email protected]
Boyan Milchev. University of Forestry, Department of Wildlife Management, g.k. Darvenitsa, bul. "Kliment Ohridski" 10, 1756 Sofia, Bulgaria. [email protected]
Adam Nadachowski. Institute of Systematics and Evolution of Animals, Polish Academy of Sciences, Sławkowska 17, 31-016, 31-016 Kraków, Poland. [email protected]
Piroska Pazonyi. HUN-REN-MTM-ELTE Research Group for Paleontology, Ludovika tér 2, 1083 Budapest, Hungary. [email protected]
Iván Rey-Rodríguez. MAPAS Lab, Centro de Investigación Mariña, Departamento de Ecoloxía e Bioloxía Animal, Universidade de Vigo, Campus Lagoas-Marcosende, 36310 Vigo, Spain. [email protected]
Xosé Pedro Rodríguez-Álvarez. Universitat Rovira I Virgili, Departament d’Història i Història de l’Art, 430007 Tarragona, Spain and Institut Català de Paleoecologia Humana i Evolució Social (IPHES-CERCA), Zona Educacional, 430007 Tarragona, Spain. [email protected]
Vincent Bonhomme. ISEM - Institut des Sciences de l'Evolution de Montpellier, Université de Montpellier, place Eugène Bataillo, 34090 Montpellier, France. [email protected]
Lové Dalén. Centre for Palaeogenetics, Stockholm, Frescativägen 8, 106 91 Stockholm, Sweden; Department of Bioinformatics and Genetics, Swedish Museum of Natural History, SE-104 05 Stockholm, Sweden and Department of Zoology, Stockholm University, 106 91 Stockholm, Sweden. [email protected]
John R. Stewart. Bournemouth University. School of Life & Environmental Sciences, Faculty of Health, Environment and Medical Sciences, Poole, BH12 5BB, United Kingdom. [email protected]

Keywords: Arvicola; systematics; diversity; 2D morphometrics; morphological divergence, discrimination rates; modern species; Pleistocene; Europe

Final citation: Desclaux, Emmanuel, Lindahl, Amanda, Kalthoff, Daniela C., Kovalchuk, Oleksandr, Centeno-Cuadros, Alejandro, Román, Jacinto, Arlegi, Mikel, Arteaga-Brieba, Andion, Barkaszi, Zoltán, Cohen, Jessica, Defleur, Alban, Fernández-García, Mónica, Flas, Damien, Frafjord, Karl, Knul, Monika Vlasta, de Lombera-Hermida, Arturo, Mahmoudi, Ahmad, Milchev, Boyan, Nadachowski, Adam, Pazonyi, Piroska, Rey-Rodríguez, Iván, Rodríguez-Álvarez, Xosé Pedro, Bonhomme, Vincent, Dalén, Lové, and Stewart, John R. 2026. 2D geometric morphometrics reveals temporal patterns in the evolution of modern and Pleistocene Arvicola (Rodentia, Cricetidae) in Europe. Palaeontologia Electronica, 29(3):a35.
https://doi.org/10.26879/1683
palaeo-electronica.org/content/2026/5941-temporal-evolution-of-arvicola-molar-morphology

Copyright: September 2026 Paleontological Society
This is an open access article distributed under the terms of Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0), which permits users to copy and redistribute the material in any medium or format, provided it is not used for commercial purposes and the original author and source are credited, with indications if any changes are made.
creativecommons.org/licenses/by-nc-sa/4.0/

INTRODUCTION

Voles of the genus Arvicola (Rodentia, Cricetidae) are among the most common and geographically widespread small mammals in Quaternary deposits across Europe. Their abundance in Middle and Upper Pleistocene archaeological and palaeontological contexts, combined with rapid evolutionary rates and ecological sensitivity, has long made them valuable indicators for biostratigraphic correlation and palaeoenvironmental reconstruction (Chaline, 1972; Koenigswald and Van Kolfschoten, 1996; Markova, 2007; Markova and Puzachenko, 2017, 2018; Mahmoudi et al., 2020; Masini et al., 2020; Krokhmal’ et al., 2021, 2023; Popova et al., 2025). Traditional morphometric approaches have revealed clear temporal trends in Arvicola dental evolution (Hinton, 1926; Heinrich, 1978, 1982, 1987, 1990; Fejfar and Heinrich, 1983; Agadzhanyan, 1983; Rekovets, 1990, 1994; Kolfschoten, 1992; Marquet, 1993; von Koenigswald and Van Kolfschoten, 1996; Abbassi et al., 1998; Desclaux et al., 2000; Maul et al., 2000, 2020; Kalthoff et al., 2007; Escudé et al., 2008a, 2008b; Ruddy, 2011) particularly in enamel thickness differentiation, named SDQ (Schmelzband-Differenzierungs-Quotient) by Heinrich (1978), which decreases through time and reflects broader functional and ecological shifts (Koenigswald et al., 1994).

Despite this well-established biostratigraphic significance, geometric morphometric approaches have been applied only sporadically to the genus Arvicola. In contrast, two-dimensional geometric morphometrics (2D GMM) have proven highly effective for quantifying continuous shape variation across a wide range of biological systems (Martinón-Torres et al., 2006; Escudé et al., 2008b; Kayser et al., 2012; Evin et al., 2013; Sagonas et al., 2014; Seetah et al., 2014; Delgado et al., 2015; Gómez-Robles et al., 2015; Dykes, 2016; Cucchi et al., 2020; Matos et al., 2020; Noerwidi et al., 2020; Galli et al., 2021; Arlegi et al., 2022; Natarajan et al., 2024; Gomez Rodriguez et al., 2025; Pelletier, 2025), including various rodent taxa (Marcolini, 2007; Escudé et al., 2013; Souto-Lima and Millien, 2014; Stoetzel et al., 2017; Montuire et al., 2019; Fox et al., 2020; Khidas et al., 2020; Moclán et al., 2023; Rey-Rodriguez et al., 2021, 2024; Arbez et al., 2026). However, landmark-based configurations may not fully capture the continuous outline variation characteristic of rootless arvicoline molars, which exhibit intricate enamel patterns and lack clear homologous landmark positions. Among outline-based methods, Elliptic Fourier Analysis (EFA) has been successfully applied to various species (Rufino et al., 2006; Bonhomme et al., 2021; Bouby et al., 2024, Machado et al., 2025) and in particular to quantify molar shape variation in rodent teeth (Renaud et al., 1996; Cucchi et al., 2002, 2005; Hurth et al., 2003; Michaux et al., 2007; Escudé et al., 2008b; Valenzuela-Lamas et al., 2011; Kimura et al., 2013; Stoetzel et al., 2013, Gomez Cano et al., 2017; Carro-Rodríguez et al., 2020; Menéndez et al., 2023). EFA is especially well-suited for rodent taxa with continuously growing teeth, such as arvicolines including the genus Arvicola, as it captures global shape properties without requiring predefined anatomical landmarks. We can note that earlier morphometric studies, primarily based on outlines, demonstrated substantial intra- and interspecific variability in first lower molar (m/1) morphology, suggesting complex evolutionary dynamics within Arvicola (Escudé et al., 2008b).

The present study applies a comprehensive 2D outline-based geometric morphometric protocol to investigate morphological variation in modern and fossil populations of Western European Arvicola. We focus on the first lower molar (m/1), a tooth widely used in taxonomic and palaeobiological studies due to its diagnostic features and frequent preservation in the fossil record. A key question addressed here is whether the two modern Western European species - Arvicola amphibius (predominantly fossorial) and Arvicola sapidus (mainly aquatic) - can be reliably distinguished based solely on m/1 outline morphology. Because these species occupy distinct ecological niches, differentiating them in fossil assemblages offers important insights into past environments and habitat structure. In addition, fossil populations of the late Middle Pleistocene age often show intermediate or archaic morphologies referred to Arvicola cantianus Hinton, 1910 or Arvicola mosbachensis Schmidtgen, 1911, raising questions about evolutionary continuity and phenotypic transitions leading to modern species.

The aims of this study are threefold: (1) to test the efficacy of outline-based 2D GMM for discriminating modern Arvicola species based on m/1 morphology; (2) to evaluate the taxonomic composition of late Middle and Late Pleistocene fossil assemblages, including the identification of archaic lineages; and (3) to reconstruct temporal trends in morphological divergence and assess the evolutionary transition from archaic Middle Pleistocene forms to fully differentiated modern species. By integrating modern comparative datasets with fossil material from diverse chronological and geographic contexts, this study provides a robust methodological and evolutionary framework for understanding Arvicola systematics and for refining their use as palaeoenvironmental and biostratigraphic indicators.

MATERIAL AND METHODS

Material

To address these issues, we built a modern reference dataset of 390 m/1 specimens representing the four extant Arvicola species - Arvicola amphibius (AA, N = 231), subdivided into Western European (AAwest, N = 194) and Eastern European (AAeast, N = 36) populations, Arvicola sapidus (AS, N = 145), Arvicola italicus (AI, N = 5), and Arvicola persicus (AP, N = 9) - originate from nine countries and spanning a broad geographical range from Norway to Iran, thus capturing broad geographic and ecological variation within the genus (Table 1, Figure 1). All material derives from natural history museum collections and researchers from various institutions and includes both wild-caught individuals and remains from owl pellets with reliable taxonomic attribution. Only well-preserved m/1s with intact occlusal surfaces and clearly defined enamel outlines were selected. Full details of specimen provenance, including institution names and inventory numbers, are provided in Appendix 1.

Using this dataset, we developed a multivariate statistical framework combining Principal Component Analysis (PCA), Multivariate Analysis of Variance (MANOVA), Linear Discriminant Analysis (LDA), Canonical Discriminant Analysis (CDA), Random Forest classification (RF), and Mahalanobis distance analysis to evaluate species-level discrimination based on molar outline shape.

This validated framework was then applied to fossil assemblages from five well-dated localities in Europe spanning the late Middle and Late Pleistocene: Lazaret Cave (France, MIS 6, ~150 ka), Moula-Guercy (France, MIS 5.5, ~125 ka), Abauntz Cave (Spain, MIS 3, ~40-50 ka), Cova Eirós (Spain, MIS 3, ~30-40 ka), and Trou Al'Wesse (Belgium, MIS 3, ~30-40 ka). This chronological framework allows us to examine temporal patterns of morphological variation and reassess the taxonomic composition of key fossil populations.

The fossil dataset includes 87 m/1 specimens from five well-stratified and well-dated archaeological sites spanning the late Middle to Late Pleistocene:

• • Lazaret Cave (Nice, France): MIS 6 (~150 ka), N=24
  • Moula-Guercy (Ardèche, France): MIS 5.5 (~125 ka), N=17
  • Abauntz Cave (Navarre, Spain): MIS 3 (~40-50 ka), N=29
  • Cova Eirós (Galicia, Spain): MIS 3 (~30-40 ka), N=12
  • Trou Al'Wesse (Modave, Belgium): MIS 3 (~30-40 ka), N=5

All fossil molars are well-preserved, free of digestion traces or taphonomic damage, and exhibit complete occlusal surfaces suitable for outline capture. Data concerning Pleistocene sites are provided in Appendix 2.

Methods

All first lower molars were photographed in standardized occlusal view using a Zeiss Stemi 508 stereomicroscope equipped with a 9-megapixel digital camera at 25× magnification. A ring-light illumination system ensured uniform lighting, and a calibrated scale bar was included in each image for size standardization. Tooth outlines were digitized manually in Fresco software using a digital pen. Tracing followed the external enamel contour of the crown, from the anterior to the posterior loop, capturing all triangles (T1-T5) and producing a complete closed outline curve (Figure 2).

All outlines were aligned using the algorithm developed by Bonhomme (2014), which is specifically designed for complex biological shapes lacking homologous landmarks. This procedure standardizes outline position, size, and orientation while preserving biologically relevant variation. The algorithm uses iterative optimization to minimize misalignment across the specimens, ensuring that subsequent analyses capture genuine morphological variation rather than artifacts of specimen positioning or digitization (Figure 3).

Aligned outlines were decomposed into harmonic components using Elliptic Fourier Transform (EFT; Kuhl and Giardina, 1982). Standard EFA was preferred over the Centered Discrete Fourier Transform (CDFT; Escudé et al., 2013), which requires identifiable biological reference points on the outline. Given that arvicoline m/1s lack clear homologous landmark positions due to their continuous, rootless growth pattern, standard EFA represents an appropriate method for capturing the full range of outline variation in these taxa. Twelve harmonics were retained, capturing 99.9% of total shape variation, yielding 48 coefficients (A, B, C, D per harmonic). However, only 13 principal components (PCs) were used in subsequent analyses (LDA, CDA), explaining 89.2% of total variance. This dimensionality reduction substantially mitigates overfitting risks, particularly for small sample sizes (Kovarovic et al., 2011). Only shape variables were analysed; size-related variables were deliberately excluded to avoid confounding effects of ecogeographic scaling (Cope’s and Bergmann’s rules; Bergmann, 1847; Mayr, 1956; Stanley, 1973; Meiri, 2011). All geometric morphometric analyses were conducted in R v. 4.5.1 (R Core Team, 2025) using the Momocs 1.5.0 package (Bonhomme et al., 2014).

A multivariate analytical framework was constructed to quantify morphological variation, assess species-level discrimination, and develop predictive models for fossil probabilistic assignment.

Principal Component Analysis (PCA) was performed on the EFT coefficient matrix to visualise morphospace structure. Multivariate Analysis of Variance (MANOVA) assessed overall morphological differentiation among species. Post-hoc univariate ANOVAs and Tukey HSD tests identified the principal components contributing most strongly to interspecific differentiation. Complementary MANOVA statistics (Pillai’s trace, Wilks’ lambda, Hotelling-Lawley trace, Roy’s greatest root) were used to ensure robust inference.

Three probabilistic assignment approaches were applied:

1. Linear Discriminant Analysis (LDA) identified linear combinations of shape variables that best separate species. Classification accuracy assessed using leave-one-out cross-validation (LOOCV). Confusion matrices were used to quantify interspecific misclassifications.

2. Canonical Discriminant Analysis (CDA) provided complementary results by incorporating within-group covariance. Mahalanobis distances between species centroids quantified morphometric divergence (Appendix 3).

3. Random Forest (RF) implemented via the caret package with 5-fold cross-validation in R (Kuhn, 2008). RF is a non-parametric ensembled method capturing nonlinear shape relationships. Variable importance scores identified the most influential principal component.

Pairwise Mahalanobis distances between specimens and modern species centroids were used to evaluate proximity and identify potentially ambiguous or intermediate morphologies. Hierarchical clustering (Ward’s method) using Euclidean and Mahalanobis distances was used to visualise interspecific relationships; cophenetic correlation coefficients assessed clustering robustness. Heatmaps of Mahalanobis distances were generated into patterns of morphological similarity and divergence. The results and interpretations are in Appendix 3.

The classification protocol developed for modern species was adapted for fossil assemblages, with two key modifications to accommodate both the palaeontological context and potential morphological intermediacy characteristic of late Middle Pleistocene and Late Pleistocene populations. Because Arvicola italicus and A. persicus occur in peripheral geographic ranges and have small sample sizes, fossil probabilistic assignment used a two-species reference framework restricted to modern Arvicola amphibius (AA) and A. sapidus (AS). Western and Eastern European AA populations were pooled, consistent with their demonstrated conspecific status. For each fossil assemblage, the following analytical sequence was applied:

1. Fossil specimens were projected onto PCA axes defined by modern specimens. MANOVA assessed morphological affinities among fossils and modern groups.

2. LDA, RF, and Mahalanobis distance approaches were applied to each fossil specimen. Posterior probabilities and distance values quantified classification confidence.

3. Agreement among the three classification methods (LDA, RF, Mahalanobis) was assessed for each fossil specimen. Consistent results indicated clear affinities; discordant classifications flagged potentially intermediate or archaic morphologies.

4. Canonical Variate Analysis (CVA) was conducted for Lazaret Cave as a collective group to evaluate its morphometric position relative to modern species (Appendix 3).

RESULTS

Modern Specimens

Because the taxonomic status of some modern Arvicola forms, especially concerning Arvicola scherman, is not consensual (Mahmoudi, 2020; Chevret et al., 2021; Kryštufek and Schenbrot, 2022; Balmori-de la Puente et al., 2022; Somoano, 2024; Roman et al., 2025), this study follows the nomenclature established in the Mammal Diversity Database (ASM), retaining four widely accepted species: A. amphibius, A. sapidus, A. italicus, and A. persicus (Figure 4).

For this study, A. amphibius has been separated in two groups (AAeast = Eastern European Arvicola amphibius and AAwest= Western European Arvicola amphibius). Indeed, the conspecific status of Western and Eastern European A. amphibius populations is supported by ancient and modern DNA analyses, which confirm their belonging to the same species while revealing a clear genetic structuring between Eastern and Western European lineages derived from distinct glacial refugia (Brace et al., 2016).

Principal Component Analysis (PCA) of Elliptic Fourier coefficients shows clear species-level structuring. Although PC1 and PC2 together explain 53.3% of variance, MANOVA indicates that higher-order components contribute substantially to discrimination, particularly PC3 (7.3% of variance), which yields the highest F-statistic (F = 45.29). This highlights that interspecific differences are distributed across multiple axes and not captured solely by the first two PCs. Species occupy distinct regions of morphospace (Figure 5, Table 2): Arvicola amphibius from Western Europe (AAwest) forms a compact homogeneous cluster; Arvicola sapidus (AS) plots on the opposite side of PC1 with limited overlap; Arvicola amphibius from Eastern Europe (AAeast) occupies an intermediate position between AAwest and other species, suggesting geographical structuring; Arvicola persicus (AP) displays higher variation along PC2; Arvicola italicus (AI) forms a small, discrete cluster with intermediate PC1 values.

MANOVA confirms highly significant differences among all species (p < 2.2e-16; Wilks’ λ = 0.1445). Tukey’s post-hoc tests show that Arvicola sapidus and A. persicus are consistently well differentiated, whereas AAwest and AI show expected similarity (AI, N = 5). A multifactor MANOVA indicates significant species × geography interactions, supporting geographic differentiation within modern species belonging to the genus Arvicola.

Overall, 2D outline-based GMM provides strong species-level discrimination in modern Arvicola, including subtle east-west differentiation within A. amphibius.

Linear Discriminant Analysis (LDA) shows strong separation among species (Figure 6), with LD1 explaining 81.91% of between-group variance. Leave-one-out cross-validation (Figure 7) reveals high classification accuracy: AAwest - 98% (192/194 specimens correctly classified) AS - 93% (135/144), AAeast - 63% (24/36, with misclassifications occurring with AAwest and AS), AP - 56% (5/9), and AI - 40% (2/5). Misclassifications occur within A. amphibius, reflecting east-west clinal variation. All other species remain well separated.

Visualisation of LDA scores (Figure 8) confirms that (i) AS is rather homogeneous, compact and distinct but we also can note that AS shows a visible contact with AAeast, (ii) AP and AI, despite small sample sizes, form identifiable groups, and (iii) there is no overlap between AAwest and AAeast and they remain distinguishable from other species.

Canonical Discriminant Analysis (CDA), see Appendix 3, strongly supports these results (prediction accuracy >99% for species with large sample sizes), and Mahalanobis distances confirm the close affinity of AAwest and AAeast, reinforcing their conspecific status. A detailed discussion on that topic is available in Appendix 3. Random Forest (RF) achieves overall accuracy of 91.4%; variable importance identifies PC1 and PC3 as the strongest contributors. KNN performs well (86.45% at k=1), whereas K-medoids show limited discrimination (ARI = 0.195). Collectively, these methods demonstrate that modern Arvicola species can be reliably discriminated based on m/1 outline morphology.

Fossil Specimens

Among the four modern species included in this study, only A. amphibius and A. sapidus were retained as reference taxa for the classification of fossil specimens, for two complementary reasons. First, the sample sizes of A. italicus (N=5) and A. persicus (N=9) are too limited to ensure reliable individual-level classification, as explicitly acknowledged throughout the modern specimen analyses. Second, and independently, neither A. italicus nor A. persicus has been reported, to our knowledge, from any Pleistocene or Holocene fossil site within the geographic area covered by the five prehistoric localities studied here - spanning southern France, northern Spain, and Belgium. The exclusion of these two peripheral species from the fossil classification framework is therefore both statistically and biogeographically justified, and the two-species AA/AS reference framework should be understood as an assessment of morphological affinities rather than a strict taxonomic classification.

Lazaret Cave (Nice, France): MIS 6 (~150 ka). Specimens from this site were previously assigned to A. cantianus morphotype amphibius and A. cantianus morphotype sapidus (Desclaux et al., 2000). The 24 Lazaret m/1 specimens occupy an intermediate position between modern AA and AS in PCA space, showing broad dispersion (Figure 9A). MANOVA confirms strong differentiation among three groups. All statistical tests (Pillai = 0.818, Wilks λ = 0.267, Hotelling-Lawley = 2.429, Roy = 2.291) show extremely significant results (p < 2.2e-16), indicating clear morphological differentiation. Tukey tests indicate that Lazaret specimens align more closely with AA on PC1 and PC3, but with AS on PC2, and show unique characteristics on PC5. LDA probabilistic assignment yields 15 AA-like specimens (62.5%) and nine AS-like specimens (37.5%). Three specimens show probabilities <80%, indicating intermediate morphologies. Morphospace range along LD1 nearly spans the full AA-AS separation observed in modern specimens. Random Forest assigns 17 specimens (70.8%) to AA and seven (29.2%) to AS. Mahalanobis distances match LDA exactly (62.5% AA; 37.5% AS). The three probabilistic assignment methods show different levels of concordance (Table 3): LDA vs Mahalanobis - complete agreement (24/24 specimens; 100%); LDA vs RF - partial agreement (14/24; 58.3%); RF vs Mahalanobis - partial agreement (14/24; 58.3%). Discordant cases likely reflect archaic, transitional morphologies typical of MIS 6 populations. The results indicate the presence of two co-occurring Arvicola lineages, with AA-like morphotype dominant (ratio ~ 1.7:1). These results strongly support the interpretation that Lazaret specimens represent archaic lineages belonging to Arvicola ex gr. cantianus/amphibius and Arvicola ex gr. cantianus/sapidus. This pattern is consistent with expected evolutionary complexity of MIS 6 refugial contexts and with previous findings of transitional morphologies in late Middle Pleistocene Arvicola. In conclusion, the MIS 6 assemblage at Lazaret Cave already shows a degree of morphological differentiation allowing partial discrimination between A. amphibius -like and A. sapidus -like morphotypes, despite their archaic character.

Baume Moula-Guercy (Ardèche, France): MIS 5.5 (~125 ka). Seventeen specimens were analysed. Specimens from Baume Moula-Guercy were previously assigned to A. amphibius and A. sapidus (Desclaux et al., 2000). PCA indicates strong morphometric differentiation among groups, with the first three components explaining 60.6% of total variance (PC1: 40.0%, PC2: 13.3%, PC3: 7.3%). Visual inspection shows clear AA-AS separation, with Moula-Guercy specimens plotting mostly within the AA morphospace (Figure 10A). MANOVA confirms highly significant group differences (Pillai’s trace = 1.016, p < 2.2e-16). Individual ANOVAs identify significant separation across multiple PCs, particularly PC1 (F = 90.23, p < 2e-16), PC3 (F = 71.02, p < 2e-16), and PC6 (F = 25.65, p = 3.47e-11). Post-hoc Tukey tests consistently show significant differences between AA and AS across most axes. LDA probabilistic assignment of the Moula-Guercy specimens (Figure 10B) yields 13 specimens assigned to Arvicola amphibius, and four specimens assigned to A. sapidus. Random Forest probabilistic assignment indicates a more strongly AA-dominated assemblage: 16 AA specimens, and a single AS specimen. LDA and Mahalanobis distance show perfect concordance (100%), and the agreement between LDA and RF is moderate (70.6%; Table 4).

Most Moula-Guercy specimens can be confidently assigned to A. amphibius (13/17 by LDA and Mahalanobis; 16/17 by RF), indicating that this species dominated the MIS 5.5 assemblage. A small set of specimens (1071, 1072, 1081, 1082, 1085) shows ambiguous assignments, likely due to retention of plesiomorphic dental traits, similar to MIS 6 patterns but less pronounced. The larger Lazaret sample size allowed LDA and RF to discriminate more effectively between morphotypes, suggesting that the limited Moula-Guercy sample (n=17) may constrain probabilistic assignment confidence for borderline specimens. Such configuration supports the interpretation that MIS 5.5 watervoles are still close to archaic lineages (A. cantianus sensu lato, syn. A. mosbachensis) that have been observed, for example, at Lazaret Cave (MIS 6).

Abauntz Cave (Navarre, Spain): MIS 3 (~40-50 ka). The Abauntz Cave assemblage (MIS 3, ~40-50 ka, N=29) presents a markedly different pattern from the Middle Pleistocene Lazaret sample. Specimens from Abauntz cave were previously assigned to A. amphibius (Arteaga-Brieba et al., 2024). PCA shows that Abauntz specimens occupy an intermediate position in morphospace (Figure 11A), overlapping substantially with modern A. amphibius but shifted leftward along PC1 (mean PC1 = -2.59 relative to modern AA), indicating subtle morphological differences. MANOVA confirms significant differences from both reference species (Wilks λ = 0.241, p < 2.2e-16), with strong discrimination on PC1 (F = 101.6, p < 2e-16) and PC3 (F = 74.04, p < 2e-16). Post-hoc Tukey tests show that Abauntz differs significantly from both AA (p = 0.0008) and AS (p < 0.0001) on PC1, confirming its distinct morphometric signature. Classification results demonstrate exceptional convergence across methods. LDA assigns 28/29 specimens (96.6%) to A. amphibius and only 1/29 (3.4%) to A. sapidus (Figure 11B) with most specimens showing very high posterior probabilities (>99%). Random Forest probabilistic assignment is nearly identical: 27/29 (93.1%) as AA and 2/29 (6.9%) as AS. LDA-RF agreement reaches 96.6% (28/29 specimens), much higher than the 58.3% observed at Lazaret, and LDA-Mahalanobis concordance is perfect at 100% (Table 5). This high methodological agreement indicates well-defined, unambiguous morphologies characteristic of fully evolved Late Pleistocene Arvicola, contrasting sharply with the transitional forms observed in MIS 6. Mean Mahalanobis distances (AA: 8.477, AS: 8.649, difference: 0.172) confirm overall affinity with A. amphibius, while maintaining a slightly intermediate position consistent with MIS 3 chronology. The single specimen (1054_abauntz_espagne) identified as A. sapidus is consistently classified across methods: LDA assigns it to AS with 81.37% probability, RF with 77.8%. Mahalanobis distances place it marginally closer to AS. Although probabilities are lower than those of typical AA specimens (95%), the probabilistic assignment is robust and likely reflects genuine taxonomic presence rather than ambiguity. Its slightly intermediate position may indicate retention of plesiomorphic traits.

Cova Eirós (Galicia, Spain): MIS 3 (~30-40 ka). Twelve specimens were analyzed. Specimens from Cova Eirós were previously assigned to A. amphibius and A. sapidus (Rey-Rodriguez et al., 2011, 2016; Bal-Garcia et al., 2026). PCA revealed strong morphometric differentiation among modern groups, with the first three components explaining 60.2% of total variance (PC1: 39.4%, PC2: 13.6%, PC3: 7.3%). Cova Eiros specimens plot across both AA and AS morphospaces, indicating a mixed assemblage (Figure 12A). MANOVA indicates strong group differentiation (Pillai’s trace = 0.897, p < 2.2e-16). ANOVAs show significant separation across nine PCs, especially PC1 (F = 77.37, p < 2e-16), PC3 (F = 68.88, p < 2e-16), and PC9 (F = 18.43, p = 2.26e-08). This robust discrimination across multiple axes provides a solid foundation for fossil classification. LDA probabilistic assignment (Figure 12B) identifies eight specimens as A. amphibius and four specimens as A. sapidus. RF probabilistic assignment yields nine AA specimens and three AS specimens. Thus, Cova Eiros shows a relatively balanced co-occurrence of both species, contrasting with the predominantly monospecific assemblages at Moula-Guercy (MIS 5.5) and Abauntz (MIS 3). Concordance statistics are as follows: LDA-RF agreement: 9/12 (75%), LDA-Mahalanobis agreement: 12/12 (100.0%). Three specimens (1121_eiros_spain, 1131_eiros_spain and 1133_eiros_spain) show disagreement between LDA and RF, although Mahalanobis distances consistently support the LDA assignments. The Cova Eirós assemblage therefore exhibits a clear bimodal taxonomic structure, with both A. amphibius and A. sapidus well represented (Table 6). Mean Mahalanobis distances to modern centroids are nearly identical (AA: 10.388 ± 3.45; AS: 10.427 ± 3.58), indicating that the fossils occupy intermediate positions consistent with their MIS 3 age.

Trou Al'Wesse (Modave, Belgium): MIS 3 (~30-40 ka). This assemblage represents a small but highly informative Late Pleistocene sample. All specimens from Trou Al’Wesse were previously assigned to A. amphibius (Miller et al., 2007). Despite the small number of specimens, results are exceptionally clear, and strong convergence across analytical methods. PCA (Figure 13A) places the Trou Al'Wesse specimens in an intermediate position between modern A. amphibius and A. sapidus along PC1, with a pronounced leftward shift (mean PC1 = -3.61 relative to modern AA). MANOVA confirms highly significant group differentiation (Wilks λ = 0.267, p < 2.2e-16), with the strongest discrimination on PC1 (F = 89.48, p < 2e-16) and PC3 (F = 61.82, p < 2e-16). Notably, post-hoc Tukey tests indicate that Trou Al'Wesse differs significantly from AS (p < 0.0001), whereas the difference from AA approaches but does not reach significance (p = 0.070), suggesting closer morphological affinity with A. amphibius. Probabilistic assignment results show perfect methodological agreement (Figure 13B; Table 7).

All five specimens are consistently assigned to A. amphibius by LDA (100%), RF (100%), and Mahalanobis distances (100%). LDA posterior probabilities are extremely high (≥99.99%), and RF probabilities range from 84.6% to 99.2%, confirming unambiguous morphological signatures. This complete concordance reflects the well-defined Late Pleistocene morphology characteristic of fully evolved A. amphibius. Mean Mahalanobis distances (AA: 9.319; AS: 9.561; difference: 0.242) further support the close affinity of the Trou Al’Wesse specimens to modern Arvicola amphibius, while maintaining a slightly intermediate position consistent with MIS 3 chronology.

DISCUSSION

This study demonstrates that two-dimensional geometric morphometric study of the first lower molar (m/1) outlines provides a robust and highly informative framework for assessing taxonomic boundaries, evolutionary trends, and temporal transitions within the genus Arvicola.

By integrating a large modern reference dataset with fossil specimens spanning MIS 6 to MIS 3, our results reveal clear morphological differentiation among modern species, substantial structuring within Arvicola amphibius, and a progressive evolutionary trajectory from archaic Middle Pleistocene forms to fully modern and differentiated modern species A. amphibius and A. sapidus.

Analyses of modern specimens confirm that m/1 morphology provides strong discriminatory power across Arvicola. Multivariate exploratory analyses (PCA, MANOVA) show significant interspecific differences, enabling reliable species-level discrimination.

Among probabilistic assignment approaches, LDA achieved an overall accuracy of 91%, driven primarily by the robust performance on the well-sampled groups A. amphibius and A. sapidus; classification accuracy for A. italicus and A. persicus was more limited, consistent with their small sample sizes and the associated statistical constraints discussed above. The strong separation of A. amphibius and A. sapidus validates long-standing ecological and morphological distinctions between fossorial and aquatic forms. The heterogeneity observed within A. amphibius - especially east-west differentiation - illustrates ecogeographic structuring likely influenced by climatic gradients, habitat variation, or historical population dynamics. Despite small sample sizes, A. italicus and A. persicus remain distinct, highlighting the sensitivity of the method even for peripheral species. While consistent morphometric signals are detectable for A. italicus and A. persicus across multiple methods - particularly the strongly negative LD4 scores for A. italicus and the Mahalanobis distances placing both species as morphometrically divergent - the degree of morphospace overlap observed and the limited classification accuracy precludes strong claims of distinctiveness. The sensitivity of the method for these peripheral species should be understood as preliminary and in need of confirmation with larger samples.

These morphometric results provide a robust basis for comparison with ongoing genetic analyses and forthcoming phylogenetic models. Multivariate distance analysis further refines species affinities, strengthening in applying this framework to fossil material. These results confirm that 2D outline-based GMM is methodologically reliable and biologically meaningful, capturing both species-level differences and fine-scale geographic variability.

The Lazaret Cave assemblage displays the most complex morphological pattern among the fossil samples. PCA, LDA, and Mahalanobis distances all reveal a pronounced intermixed distribution, with fossils spanning nearly the full morphospace between modern A. amphibius and A. sapidus. High dispersion and discordance between LDA and RF probabilistic assignments indicate the presence of intermediate or transitional morphotypes rather than discrete modern-like species. These results strongly support the interpretation that Lazaret specimens represent archaic lineages belonging to Arvicola ex gr. cantianus/amphibius (60-65%) and Arvicola ex gr. cantianus/sapidus (35-40%). This pattern is consistent with expected evolutionary complexity of MIS 6 refugial contexts in southeastern France (Hanquet et al., 2010; Lopez-Garcia et al., 2021; Barakat et al., 2025). The mosaic nature of these specimens underscores the importance of multivariate classification approaches: parametric methods (LDA, Mahalanobis) capture broad shape affinities, whereas RF is more sensitive to localised or nonlinear variation, which may explain its greater tendency to identify ambiguous morphotypes. The MIS 6 assemblage at Lazaret Cave already shows a degree of morphological differentiation allowing partial discrimination between A. amphibius -like and A. sapidus -like morphotypes, despite their archaic character - demonstrating that EFA can detect lineage differentiation even in pre-modern forms.

The MIS 5.5 assemblage at Baume Moula-Guercy shows reduced variability compared to Lazaret and a predominantly A. amphibius -like signal, consistent with the persistence of archaic lineages close to A. cantianus sensu lato at this time.

Molecular clock analyses place the split between A. amphibius and A. sapidus at ~1.5 Ma (Mahmoudi et al., 2020), with palaeontological evidence of sapidus -type morphology in Arvicola jacobaeus from the Early Pleistocene of Atapuerca (~1.2-1.5 Ma; Cuenca-Bescós et al., 2010). The presence of archaic sapidus -type morphology at Lazaret Cave (MIS 6) and Baume Moula-Guercy (MIS 5.5) suggests that A. sapidus reached southern France well before the 62-ka colonization date estimated from modern genomes (Centeno-Cuadros et al., 2009). While ancient DNA analysis could theoretically reveal morphological convergence rather than true phylogenetic affinity, our morphometric results suggest early A. sapidus presence in southern France, at least during MIS 6 and MIS 5.5. Our morphometric data indicates a more complex and dynamic biogeographic history involving multiple immigration waves from Iberia during the Middle and early Late Pleistocene, with subsequent local extinction(s) likely during MIS 4. The 62-ka estimate represents the most recent successful colonisation, not the first, highlighting the complementary value of morphological and genetic approaches in reconstructing evolutionary history.

Furthermore, the results obtained for Lazaret Cave and Baume Moula-Guercy are in accordance with previous studies, based on morphology and SDQ (Schmelzband-Differenzierungs-Quotient) values, which have highlighted the presence of transitional morphologies in late Middle Pleistocene and early Upper Pleistocene Arvicola (Heinrich, 1978, 1982, 1990, 1997; Fejfar and Heinrich, 1983; Agadzhanyan, 1983; Rekovets, 1990, 1994; Kolfschoten, 1992; Marquet, 1993; von Koenigswald and Van Kolfschoten, 1996; Abbassi et al., 1998; Desclaux et al., 2000; Maul et al., 2000; Kalthoff et al., 2007; Escudé et al., 2008a, 2008b; Ruddy, 2011).

The MIS 3 sites - Abauntz, Cora Eirós, and Trou Al’Wesse - showed a marked shift toward fully modern morphologies. In Abauntz, nearly all (96-97%) specimens are confidently classified as Arvicola amphibius sensu stricto, with rare occurrences of A. sapidus (3-4%). The low discordance rate (3.4%) and high classification probabilities indicate that by MIS 3, dental morphology had largely stabilised into modern species phenotypes, in sharp contrast to the 41.7% discordance observed at MIS 6 Lazaret. These results support the evolutionary transition from archaic A. cantianus forms in the Middle Pleistocene to fully differentiated modern species in the Late Pleistocene. The Cova Eirós assemblage documents sympatry between A. amphibius and A. sapidus, while Trou Al’Wesse yields a clearly monospecific Arvicola amphibius with perfect probabilistic assignment concordance. These results demonstrate that by MIS 3, modern Arvicola phenotypes had become fully established across Western and Central Europe, with morphological ambiguity almost entirely absent. The strong convergence among probabilistic assignment methods (up to 100% agreement) at these sites contrasts sharply with the mixed morphologies at MIS 6, underscoring evolutionary transition between the late Middle or Late Pleistocene. From MIS 3 onwards, morphotypes become fully congruent with modern species, as demonstrated by the near-perfect classification rates at Abauntz Cave (LDA: 96.6% AA), Cova Eiros (mixed AA/AS assemblage confirmed) and Trou Al’Wesse (LDA and RF: 100% AA).

This temporal progression is independently validated by the increasing convergence between LDA and RF probabilistic assignments through time - from partial concordance at MIS 6 to full concordance at MIS 3 - which we interpret as reflecting the progressive morphological stabilization of modern Arvicola phenotypes. This pattern constitutes one of the most original evolutionary findings of the present study.

Our outline-based EFA framework demonstrates that, despite the considerable individual variability species in Arvicola m/1 morphology that complicated interspecific discrimination, which has been already mentioned by Escudé et al. (2008b), species-level discrimination remains highly reliable when global outline shape is captured in its entirety. This suggests that outline methods are particularly well suited to taxa with high intraspecific morphological variation such as Arvicola.

Our work shows a highly consistent picture: at four out of five sites (Lazaret Cave, Baume Moula-Guercy, Cova Eiros and Trou Al’Wesse), 2D morphometric results are fully concordant with prior identifications, providing independent validation of both the paleontological determinations and our methodological framework. The most significant result in this respect concerns Abauntz Cave: while the assemblage had been previously identified as monospecific (A. amphibius), our EFA-based analysis suggests the possible presence of a single A. sapidus specimen (specimen 1054_abauntz_spain, LDA: AS 81.37%; RF: AS 77.8%; Mahalanobis closer to AS). This finding, supported by the convergence of three independent methods, illustrates precisely the added value of outline-based morphometrics over traditional qualitative identification: it can detect minority taxa that might otherwise be overlooked in monospecific assemblages.

CONCLUSION

A comprehensive 2D geometric morphometric protocol based on m/1 outline analysis was applied to a modern reference dataset of 390 specimens representing four Arvicola species: A. amphibius (N = 231), A. sapidus (N = 145), A. italicus (N = 5), and A. persicus (N = 9). Using a multivariate analytical framework combining PCA, MANOVA, LDA, CDA, Random Forest probabilistic assignment, and Mahalanobis distance analysis, we show that all modern species exhibit significant and diagnostic differences in molar outline shape. These findings confirm that species-level discrimination based on m/1 morphology is reliable across the genus.

Applying this validated framework, with minor specific adaptations, to fossil assemblages spanning MIS 6 to MIS 3 reveals a clear temporal evolutionary trajectory. The MIS 6 assemblage from Lazaret Cave displays high morphological variability, with intermediate morphotypes and complex morphologies characteristic of archaic Arvicola lineages (A. cantianus sensu lato, syn. A. mosbachensis). Specimens fall into two morphological groups - Arvicola ex gr. cantianus/amphibius and Arvicola ex gr. cantianus/sapidus - representing transitional forms ancestral to modern species. Comparison with other MIS 6 assemblages and forthcoming aDNA data will help refine this terminology. By MIS 5.5 (Baume Moula-Guercy), assemblages are dominated by A. amphibius -like morphologies, although several specimens retain plesiomorphic traits. In contrast, MIS 3 assemblages (Abauntz, Cova Eirós, Trou Al’Wesse) show predominantly or entirely modern phenotypes, with high concordance among probabilistic assignment methods and minimal ambiguity. Together, these results document the progressive morphological divergence and eventual stabilization of the modern Arvicola amphibius and A. cantianus lineages during the late Middle to Late Pleistocene.

The data also supports a more complex biogeographic history for Arvicola sapidus than previously inferred from modern genetic evidence alone, suggesting multiple dispersal events into Western Europe during the Middle and early Late Pleistocene, followed by local extinctions and later recolonization. Overall, this study demonstrates that m/1 outline based geometric morphometrics is a powerful and cost-effective tool for resolving taxonomic identities and exploring evolutionary processes in both modern and fossil contexts (Rekovets and Kovalchuk, 2017). By integrating modern and Pleistocene datasets within a unified framework, we provide new insights into the morphological evolution, lineage differentiation, and biogeographic dynamics within the genus Arvicola across the European Pleistocene.

FUNDING

This research forms part of the ERC project PrimiGenomes - Investigating mammalian evolution using million-year genomic transects. It is also supported by the collective research project Paleoecology of the Lazaret Cave: human-environment interactions on the coast of the meridional Alps during the Late Middle Pleistocene (MIS 6), funded by the DRAC PACA (French Ministry of Culture) and the Département des Alpes-Maritimes. IR-R is supported by the grant Axudas de apoio á etapa de formación postdoutoral (ED481B-2023-040) from the Consellería de Cultura, Educación, Formación Profesional e Universidades, Xunta de Galicia. M.F.-G. is currently supported by a Marie Skłodowska-Curie grant (HORIZON-MSCA-2021-PF-01; Ref. 101205968).

ACKNOWLEDGEMENTS

The authors thank B. Kryštufek for providing modern specimens of Arvicola italicus and for his advice.

DATA AVAILABILITY STATEMENT

The R script used for statistical analyses is publicly available on Huma-Num:
https://sharedocs.huma-num.fr/wl/?id=SIaouFJjYm2dsYg6zEU9qDKALn3qd3kl

REFERENCES

Abbassi, M., Desclaux, E., Marquet, J-C., and Chaline, J. 1998. Répartition et évolution des Arvicola Lacépède, 1799 (Rodentia, Mammalia) au cours du Pléistocène moyen et supérieur en France et en Ligurie. Quaternaire, 9:105-116.
https://doi.org/10.3406/quate.1998.1594 

Agadzhanyan, A.K. 1983. Eine altertümliche Arvicola (Rodentia, Mammalia) aus dem Mittelpleistozän der Russischen Ebene. Schriftenreihe für Geologische Wissenschaften, 19/20:9-29.

Arbez, L., Royer, A., Gilg, O., Pohl, B., Buoncristiani, J., and Montuire, S. 2026. Living on the edge: Pleistocene ice‐free refugia and collared lemming (Dicrostonyx sp.) in the North American High Arctic. Boreas, 55:216-230.
https://doi.org/10.1111/bor.70026

Arlegi, M., García-Sagastibelza, A., Veschambre-Couture, C., and Gómez-Olivencia, A. 2022. Sexual dimorphism in the first rib of Homo sapiens. Journal of Anatomy, 240:959-971.
https://doi.org/10.1111/joa.13594

Arteaga-Brieba, A., Arlegi, M., Alonso-Llamazares, C., Arriolabengoa, M., Burguet-Coca, A., Cubas, M., Del Cojo, F., Fernández-García, M., Gómez-Olivencia, A., Hermoso De Mendoza, A., López-Polín, L., Morales, J.I., Ollé, A., Pablos, A., Pantoja-Pérez, A., Rico, A., Rios-Garaizar, J., Rodríguez-Almagro, M., Rodríguez-Hidalgo, A., Sala, N., Santamaría-Díaz, U., Soares-Remiseiro, M., Val-Peón, C., and Villalba De Alvarado, M. 2025. Intervención arqueológica en la cueva de Abauntz (Arraitz-Orkin). Resumen de la campaña 2024 y nuevas dataciones. Trabajos de Arqueología Navarra, 36.
https://doi.org/10.35462/tan36.7

Balmori-de La Puente, A., Ventura, J., Miñarro, M., Somoano, A., Hey, J., and Castresana, J. 2022. Divergence time estimation using ddRAD data and an isolation-with-migration model applied to water vole populations of Arvicola. Sci. Rep. 12, 4065. https://doi.org/10.1038/s41598-022-07877-y

Bal-García, H., Rey-Rodríguez, I., Fernández-Rodríguez, C., Díaz-Rodríguez, M., Ibáñez-Encinas, Á., Mosquera-Castro, T., De Lombera-Hermida, A., Rodríguez-Álvarez, X.P., and Fábregas Valcarce, R. 2026. Palaeoecological dynamics during the Middle and Upper Palaeolithic in northwest Iberia: the Cova Eirós archaeological site. Palaeogeography, Palaeoclimatology, Palaeoecology, 699:114026.
https://doi.org/10.1016/j.palaeo.2026.114026

Barakat, S., Williams, K., Willmes, M., Desclaux, E., and Britton, K. 2025. Multi-isotope analysis of mammalian fauna from the Middle Pleistocene (MIS6) Lazaret Cave: palaeoecological, palaeoenvironmental and archaeological implications. Journal of Archaeological Science: Reports, 68:105456.
https://doi.org/10.1016/j.jasrep.2025.105456

Bergmann, C. 1847. Über die verhältnisse der Wärmeökonomie der Thiere zu ihrer Grösse. Göttingen Studies, 1:595-708.
https://doi.org/10.1007/978-3-662-30049-7_1

Bonhomme, V., Picq, S., Gaucherel, C., and Claude, J. 2014. Momocs: Outline Analysis Using R, Journal of Statistical Software, 56:1-24.
https://www.jstatsoft.org/v56/i13/

Bonhomme, V., Terral, J.-F., Zech-Matterne, V., Ivorra, S., Lacombe, T., Deborde, G., Kuchler, P., Limier, B., Pastor, T., Rollet, P., and Bouby, L. 2021. Seed morphology uncovers 1500 years of vine agrobiodiversity before the advent of the Champagne wine. Scientific Reports, 11:2305.
https://doi.org/10.1038/s41598-021-81787-3

Bouby, L., Bonhomme, V., Ivorra, S., Bacilieri, R., Ben Makhad, S., Bonnaire, E., Cabanis, M., Derreumaux, M., Dietsch-Sellami, M.F., Durand, F., Evin, A., Figueiral, I., Flottes, L., Hallavant, C., Jedrusiak, F., Lacombe, T., Marinval, P., Martin, L., Matterne, V., Pagnoux, C., Pastor, T., Pinaud, R., Pradat, B., Preiss, S., Ros, J., Rovira, N., Ruas, M.P., Schaal, C., Tillier, M., Toulemonde, F., Wiethold, J., and Terral, J.F. 2024. Seed morphometrics unravels the evolutionary history of grapevine in France. Scientific Reports, 14:22207.
https://doi.org/10.1038/s41598-024-72692-6

Brace, S., Ruddy, M., Miller, R., Schreve, D.C., Stewart, J.R., and Barnes, I. 2016. The colonization history of British water vole (Arvicola amphibius (Linnaeus, 1758)): origins and development of the Celtic fringe. Proceedings of the Royal Society B: Biological Sciences, 283:20160130.
https://doi.org/10.1098/rspb.2016.0130

Carro-Rodríguez, P.M., López-Guerrero, P., and Álvarez-Sierra, M.Á. 2020. Fourier analysis applied to Cricetodon and Hispanomys (Rodentia, Mammalia): its use as a taxonomical criterion. Historical Biology, 32:104-127.
https://doi.org/10.1080/08912963.2018.1472254

Centeno-Cuadros, A., Delibes, M., and Godoy, J.A. 2009. Phylogeography of southern water vole (Arvicola sapidus): evidence for refugia within the Iberian glacial refugium? Molecular Ecology, 18:3652-3667.
https://doi.org/10.1111/j.1365-294X.2009.04297.x

Chaline, J. 1972. Les rongeurs du Pleistocène moyen et supérieur de France. Cahiers de Paléontologie CNRS, 1-410.

Chevret, P., Renaud, S., Helvaci, Z., Ulrich, R.G., Quéré, J., and Michaux, J.R. 2020. Genetic structure, ecological versatility, and skull shape differentiation in Arvicola water voles (Rodentia, Cricetidae). Journal of Zoological Systematics and Evolutionary Research, 58:1323-1334.
https://doi.org/10.1111/jzs.12384

Cucchi, T., Vigne, J.-D., Auffray, J.-C., Croft, P., and Peltenburg, E. 2002. Introduction involontaire de la souris domestique (Mus musculus domesticus) à Chypre dès le Néolithique précéramique ancien (fin IXe et VIIIe millénaires av. J.-C.). Comptes Rendus Palevol, 1:235-241.
https://doi.org/10.1016/S1631-0683(02)00033-7

Cucchi, T., Orth, A., Auffray, J.-C., Renaud, S., Fabre, L., Catalan, J., Hadjisterkotis, E., Bonhomme, F., and Vigne, J.-D. 2005. A new endemic species of the subgenus Mus (Rodentia, Mammalia) on the Island of Cyprus.
https://doi.org/10.5281/ZENODO.172873

Cucchi, T., Papayianni, K., Cersoy, S., Aznar-Cormano, L., Zazzo, A., Debruyne, R., Berthon, R., Bălășescu, A., Simmons, A., Valla, F., Hamilakis, Y., Mavridis, F., Mashkour, M., Darvish, J., Siahsarvi, R., Biglari, F., Petrie, C.A., Weeks, L., Sardari, A., Maziar, S., Denys, C., Orton, D., Jenkins, E., Zeder, M., Searle, J.B., Larson, G., Bonhomme, F., Auffray, J.-C., and Vigne, J.-D. 2020. Tracking the Near Eastern origins and European dispersal of the western house mouse. Scientific Reports, 10:8276.
https://doi.org/10.1038/s41598-020-64939-9

Cuenca-Bescós, G., Agustí, J., Lira, J., Rubio, M.M., and Rofes, J. 2010. A new species of water vole from the Early Pleistocene of Southern Europe. Acta Palaeontologica Polonica, 55:565-580.
https://doi.org/10.4202/app.2009.0027

Delgado, M.N., Gamarra, B., Nadal, J., Mercadal, O., Olesti, O., Guàrdia, J., Pérez-Pérez, A., and Galbany, J. 2015. Dental shape variability in cercopithecoid primates: A model for the taxonomic attribution of macaques from Roman archaeological contexts. Folia Primatologica (Basel), 85:361-378.
https://doi.org/10.1159/000371633

Desclaux, E., Abbassi, M, Marquet, J-C., Chaline, J., and Van, K. 2000. Distribution and evolution of Arvicola Lacépède, 1799 (Mammalia, Rodentia) in France and Liguria (Italy) during the Middle and the Upper Pleistocene. Acta zoologica cracoviensia, 43:107-125.

Dykes, S.J. 2016. A morphometric analysis of hominin teeth attributed to Australopithecus, Paranthropus and Homo. South African Journal of Science, 112:15.
https://doi.org/10.17159/sajs.2017/20160136

Escudé, É., Montuire, S., and Desclaux, E. 2008a. Variabilité morphologique de l’espèce Arvicola cantiana (Arvicolinae, Rodentia) du Pléistocène moyen au Pléistocène supérieur de France et de Ligurie (Italie). Quaternaire, 19:31-41.
https://doi.org/10.4000/quaternaire.1632

Escudé, E., Montuire, S., Desclaux, E., Quéré, J.-P., Renvoisé, E., and Jeannet, M. 2008b. Reappraisal of ‘chronospecies’ and the use of Arvicola (Rodentia, Mammalia) for biochronology. Journal of Archaeological Science, 35:1867-1879.
https://doi.org/10.1016/j.jas.2007.11.018

Escudé, É., Renvoisé, É., Lhomme, V., and Montuire, S. 2013. Why all vole molars (Arvicolinae, Rodentia) are informative to be considered as proxy for Quaternary paleoenvironmental reconstructions. Journal of Archaeological Science, 40:11-23.
https://doi.org/10.1016/j.jas.2012.03.003

Evin, A., Cucchi, T., Cardini, A., Strand Vidarsdottir, U., Larson, G., and Dobney, K. 2013. The long and winding road: identifying pig domestication through molar size and shape. Journal of Archaeological Science, 40:735-743.
https://doi.org/10.1016/j.jas.2012.08.005

Fejfar, O. and Heinrich, W.-D. 1989. Muroid rodent biochronology of the Neogene and Quaternary in Europe, p. 91-116. In Lindsay, E.H., Fahlbusch, V., and Mein, P. (eds.), European Neogene Mammal Chronology. New York, London (Plenum Press).

Fox, N.S., Veneracion, J.J., and Blois, J.L. 2020. Are geometric morphometric analyses replicable? Evaluating landmark measurement error and its impact on extant and fossil Microtus classification. Ecology and Evolution, 10:3260-3275.
https://doi.org/10.1002/ece3.6063

Galli, E., Fidalgo, D., García-Martínez, D., and Rosas, A. 2021. Geometric morphometrics applied to molars of Pleistocene Bos and Bison of Europe, Comunicações Geológicas, 108:71-76.
https://doi.org/10.34637/TH8W-HP63

Gomes Rodrigues, H., Chabot, J., Cucchi, T., and Billet, G. 2025. Accommodation of dental variations during jaw growth in ungulate mammals. Journal of Experimental Zoology B, 344:487-504.
https://doi.org/10.1002/jez.b.23321

Gomez Cano, A.R., Kimura, Y., Blanco, F., Menéndez, I., Álvarez-Sierra, M.A., and Hernández Fernández, M. 2017. Ecomorphological characterization of murines and non-arvicoline cricetids (Rodentia) from south-western Europe since the latest Middle Miocene to the Mio-Pliocene boundary (MN 7/8-MN13). PeerJ, 5:e3646.
https://doi.org/10.7717/peerj.3646

Gómez-Robles, A., Bermúdez De Castro, J.M., Martinón-Torres, M., Prado-Simón, L., and Arsuaga, J.L. 2015. A geometric morphometric analysis of hominin lower molars: Evolutionary implications and overview of postcanine dental variation. Journal of Human Evolution, 82:34-50.
https://doi.org/10.1016/j.jhevol.2015.02.013

Hanquet, C., Valensi, P., Bailon, S., Desclaux, E., Guennouni, K.E., Roger, T., and de Lumley, H. 2010. Caractérisation du climat et de la biodiversité au Pléistocène moyen final,d’après les faunes de vertébrés de la grotte du Lazaret (Nice, France). Quaternaire, 21:203-214.
https://doi.org/10.4000/quaternaire.5579

Heinrich, W.-D. 1978. Zur biometrischen Erfassung eines Evolutionstrends bei Arvicola (Rodentia, Mammalia) aus dem Pleistozän Thüringens. Säugetierkundliche Informationen, 2:3-21.

Heinrich, W.-D. 1982. Zur Evolution und Biostratigraphie von Arvicola (Rodentia, Mammalia) im Pleistozän Europas. Zeitschrift für Geologische Wissenschaften, 10:683-735.

Heinrich, W.-D. 1987. Neue Ergebnisse zur Evolution und Biostratigraphie von Arvicola (Rodentia, Mammalia) im Quartär Europas. Zeitschrift für Geologische Wissenschaften, 15:389-406.

Heinrich, W.-D. 1990. Some aspects of the evolution and biostratigraphy of Arvicola (Mammalia, Rodentia) in the Central European Pleistocene, p. 165-182. In Fejfar, O. and Heinrich, W.-D. (eds.), International Symposium Evolution, Phylogeny, and Biostratigraphy of Arvicolids (Rodentia, Mammalia). Pfeil-Verlag, München.

Hinton, M.A.C. 1926. Monograph of the Voles and Lemmings (Microtinae) living and extinct. Vol. 1. British Museum (Natural History), London.
https://doi.org/10.5962/bhl.title.8319

Hurth, E., Montuire, S., Schmittbuhl, M., Le Minor, J.-M., Schaaf, A., Viriot, L., and Chaline, J. 2003. Examination of the tooth morphospace of three Mimomys lineages (Arvicolinae, Rodentia) by elliptical Fourier methods. Coloquios de Paleontologia, 1:325-334.

Kaiser, T.M., Müller, D.W.H., Fortelius, M., Schulz, E., Codron, D., and Clauss, M. 2012. Characterization of ungulate buccal cusp shape using outline-based geometric morphometrics and its implication for mesowear analyses. Palaios, 27:396-405.

Kalthoff, D.C., Mörs, T., and Tesakov, A., 2007. Late Pleistocene small mammals from the Wannenköpfe volcanoes (Neuwied Basin, western Germany) with remarks on the stratigraphic range of Arvicola terrestris. Geobios, 40:609-623.
https://doi.org/10.1016/j.geobios.2006.09.003

Khidas, K., Schmidt, E.A., and Fauteux, D. 2020. Comparative 2D-shape analyses of collared lemmings in the zone of possible sympatry between Dicrostonyx groenlandicus and Dicrostonyx richardsoni (Rodentia, Arvicolinae). Arctic Science, 6:423-436.
https://doi.org/10.1139/as-2019-0023

Kimura, Y., Jacobs, L.L., and Flynn, L.J. 2013. Lineage-specific responses of tooth shape in murine rodents (Murinae, Rodentia) to late Miocene dietary change in the Siwaliks of Pakistan. PLoS ONE, 8:e76070.
https://doi.org/10.1371/journal.pone.0076070

Koenigswald, W. von, Sander, P.M., Leite, M.B., Mörs, T., and Santel, W. 1994. Functional symmetries in the schmelzmuster and morphology of rootless rodent molars. Zoological Journal of the Linnean Society, 110:141-179.
https://doi.org/10.1111/j.1096-3642.1994.tb01474.x

Koenigswald, W. von and Kolfschoten, T. van. 1996. The Mimomys-Arvicola boundary and the enamel thickness quotient (SDQ) of Arvicola as stratigraphic markers in the Middle Pleistocene, p. 211-226. In Turner, C. (ed.), The Early Middle Pleistocene in Europe.

Kolfschoten, T. van. 1992. Aspects of the migration of mammals to northwestern Europe during the Pleistocene, in particular the reimmigration of Arvicola terrestris. Courier Forschungsinstitut Senckenberg, 153:213-220.

Kovarovic, K., Aiello, L.C., Cardini, A., and Lockwood, C.A. 2011. Discriminant function analyses in archaeology: are classification rates too good to be true? Journal of Archaeological Science, 38:3006-3018.
https://doi.org/10.1016/j.jas.2011.06.028

Krokhmal’, O., Rekovets, L., and Kovalchuk, O. 2021. An updated biochronology of Ukrainian small mammal faunas of the past 1.8 million years based on voles (Rodentia, Arvicolidae): a review. Boreas, 50:619-630.
https://doi.org/10.1111/bor.12511

Krokhmal’, O., Rekovets, L., and Kovalchuk, O. 2023. Biochronological scheme of the Quaternary of the south of Eastern Europe and its substantiation based on arvicoline teeth morphometrics. Quaternary International, 674-675:5-17.
https://doi.org/10.1016/j.quaint.2022.12.003

Kryštufek, B. and Shenbrot, G. 2022. Voles and Lemmings (Arvicolinae) of the Palaearctic Region. University of Maribor, University Press.
https://doi.org/10.18690/um.fnm.2.2022

Kuhl, F.P. and Giardina, C.R. 1982. Elliptic Fourier features of a closed contour. Computer Vision, Graphics, and Image Processing, 18:236-258.
https://doi.org/10.1016/0146-664X(82)90034-X

Kuhn, M. 2008. Building Predictive Models in R Using the caret Package. Journal of Statistical Software, 28:1-26.
https://doi.org/10.18637/jss.v028.i05https://doi.org/10.18637/jss.v028.i05

López-García, J.M., Blain, H.-A., Sánchez-Bandera, C., Cohen, J., Lebreton, L., Montuire, S., Stewart, J.R., and Desclaux, E. 2021. Multi-method approach using small vertebrate assemblages to reconstruct the Marine Isotope Stage 6 climate and environment of the Lazaret cave sequence (Maritime Alps, Nice, France). Palaeogeography, Palaeoclimatology, Palaeoecology, 577:110529.
https://doi.org/10.1016/j.palaeo.2021.110529

Machado, R.S.R., Bonhomme, V., Soteras, R., Jeanty, A., Bouby, L., Evin, A., Fernandes Martins, M.J., Gonçalves, S., Antolín, F., Salavert, A., and Oliveira, H.R. 2025. The origins and spread of the opium poppy (Papaver somniferum L.) revealed by genomics and seed morphometrics. Philos. Transactions of the Royal Society B Biological Sciences, 380:20240198.
https://doi.org/10.1098/rstb.2024.0198

Mahmoudi, A., Maul, L.C., Khoshyar, M., Darvish, J., Aliabadian, M., and Kryštufek, B. 2020. Evolutionary history of water voles revisited: confronting a new phylogenetic model from molecular data with the fossil record. Mammalia, 84:171-184.
https://doi.org/10.1515/mammalia-2018-0178

Marcolini, F. 2007. Fourier Analysis Applied to Mimomys (Arvicolidae, Rodentia, Mammalia) First Lower Molars - Biochronological Implications. Mathematical Geology, 38:667-678.
https://doi.org/10.1007/s11004-006-9043-y

Markova, A.K. 2007. Pleistocene mammal faunas of Eastern Europe. Quaternary International, 160:100-111.
https://doi.org/10.1016/j.quaint.2006.09.011

Markova, A.K. and Puzachenko, A. 2017. Small mammal fauna in Europe during the second half of the Middle Pleistocene. Fossil Imprint, 73:48-66.
https://doi.org/10.2478/if-2017-0002 

Markova, A.K. and Puzachenko, A. 2018. Middle Pleistocene small mammal faunas of Europe: Evolution, biostratigraphy, correlations. Geography, Environment, Sustainability, 11:21-38.
https://doi.org/10.24057/2071-9388-2018-11-3-21-38

Marquet, J.-C. 1993. Paléoenvironnement et chronologie des sites du domaine atlantique français d’âge Pléistocène moyen et supérieur d’après l’étude des rongeurs. Cahiers de la Claise, 627.

Martinón-Torres, M., Bastir, M., Bermúdez De Castro, J.M., Gómez, A., Sarmiento, S., Muela, A., and Arsuaga, J.L. 2006. Hominin lower second premolar morphology: evolutionary inferences through geometric morphometric analysis. Journal of Human Evolution, 50:523-533.
https://doi.org/10.1016/j.jhevol.2005.12.004

Masini, F., Maul, L.C., Abbazzi, L., Petruso, D., and Savorelli, A. 2020. Independent water vole (Mimomys savini, Arvicola: Rodentia, Mammalia) lineages in Italy and Central Europe. Fossil Imprint, 76:59-83.
https://doi.org/10.37520/fi.2020.005

Matos, A.S., Matthews-Cascon, H., and Chaparro, O.R. 2020. Morphometric analysis of the shell of the intertidal gastropod Echinolittorina lineolata (d’Orbigny, 1840) at different latitudes along the Brazilian coast. Journal of the Marine Biological Association of the United Kingdom, 100:725-731.
https://doi.org/10.1017/S0025315420000624

Maul, L.C., Rekovets, L., Heinrich, W.-D., Keller, T., and Storch, G. 2000. Arvicola mosbachensis (Schmidtgen 1911) of Mosbach 2: a basic sample for the early evolution of the genus and a reference for further biostratigraphical studies. Senckenberg. Lethaea, 80:129-147.
https://doi.org/10.1007/BF03043667

Mayr, E., 1956. Geographical character gradients and climatic adaptation. Evolution, 10:105.
https://doi.org/10.2307/2406103

Meiri, S. 2011. Bergmann’s Rule - what’s in a name? Global Ecology and Biogeography, 20:203-207.
https://doi.org/10.1111/j.1466-8238.2010.00577.x

Menéndez, I., Zelditch, M.L., Tejero-Cicuéndez, H., Swiderski, D.L., Carro-Rodríguez, P.M., Hernández Fernández, M., Álvarez-Sierra, M.Á., and Gómez Cano, A.R. 2023. Dietary adaptations and tooth morphology in squirrels: Insights from extant and extinct species. Palaeogeography, Palaeoclimatology, Palaeoecology, 629:111788.
https://doi.org/10.1016/j.palaeo.2023.111788

Michaux, J., Cucchi, T., Renaud, S., Garcia‐Talavera, F., and Hutterer, R. 2007. Evolution of an invasive rodent on an archipelago as revealed by molar shape analysis: the house mouse in the Canary Islands. Journal of Biogeography, 34:1412-1425.
https://doi.org/10.1111/j.1365-2699.2007.01701.x

Miller, R., Stewart, J.R., and Otte, M. 2007. Résultats préliminaires de l'étude de la séquence paléolithique au Trou Al'Wesse (comm. de Modave). Notae Praehistoricae, 27:41-49.

Moclán, A., Domínguez-García, Á.C., Stoetzel, E., Cucchi, T., Sevilla, P., and Laplana, C. 2023. Machine learning interspecific identification of mouse first lower molars (genus Mus Linnaeus, 1758) and application to fossil remains from the Estrecho Cave (Spain). Quaternary Science Reviews, 299:107877.
https://doi.org/10.1016/j.quascirev.2022.107877

Montuire, S., Royer, A., Lemanik, A., Gilg, O., Sokolova, N., Sokolov, A., Desclaux, E., Nadachowski, A., and Navarro, N. 2019. Molar shape differentiation during range expansions of the collared lemming (Dicrostonyx torquatus) related to past climate changes. Quaternary Science Reviews, 221:105886.
https://doi.org/10.1016/j.quascirev.2019.105886

Natarajan, S., Ahmed, J., Shetty, S., Jose, N.P., Chowdappa, S., and Chengappa, K. 2024. Unveiling the third dimension of tooth shape: 2D versus 3D geometric morphometry of human post-canine dentition. Journal of Oral Maxillofacial Pathology, 28:716-724.
https://doi.org/10.4103/jomfp.jomfp_451_23

Navarro, N., Zatarain, X., and Montuire, S., 2004. Effects of morphometric descriptor changes on statistical classification and morphospaces: Effects of morphometric descriptor changes. Biological Journal of the Linnean Society, 83:243-260.
https://doi.org/10.1111/j.1095-8312.2004.00385.x

Noerwidi, S., Vialet, A., Widianto, H., Kurniawan, I., Zaim, J., Suriyanto, R.A., Joordens, J., Lorenzo, C., Simanjuntak, T., and Sémah, F. 2020. Exploring the diversity of fossil hominin dental patterns in the western Indonesian archipelago during the Quaternary by Geometric Morphometric Analysis. Application on second upper and lower molars. L’Anthropologie, 124:102791.
https://doi.org/10.1016/j.anthro.2020.102791

Pelletier, M. 2025. Origin, evolution and biogeographic dynamics of the European rabbit (Oryctolagus cuniculus) in Southwestern Europe. The Anatomical Record, ar.70072.
https://doi.org/10.1002/ar.70072

Popova, L., Veklych, Y., Kovalchuk, O., Mishta, A., Stachowicz-Rybka, R., Gorobets, L., Yanenko, V., Nezdolii, Ye., Stupak, A., Stakhiv, M., Yevstropov, A., Lyshenko, M., Rekovets, L., and Barkaszi, Z. 2025. Within the boundaries of the Dnipro ice lobe: Biotic dynamics in the Middle Dnipro area (Ukraine) during deglaciation and postglacial stages. Journal of Quaternary Science, 40:53-70.
https://doi.org/10.1002/jqs.3659

Rekovets, L.I. 1990. Principal developmental stages of the water vole genus Arvicola (Rodentia, Mammalia) from the Eastern Europe Pleistocene, p. 369-384. In Fejfar, O. and Heinrich, W.-D. (eds.), International Symposium Evolution, Phylogeny, and Biostratigraphy of Arvicolids (Rodentia, Mammalia). Pfeil-Verlag, München.

Rekovets, L.I. 1994. Small mammals of the Anthropogene of the south of Eastern Europe. Naukova dumka, Kyiv. (In Russian)

Rekovets, L.I. and Kovalchuk, О.М. 2017. Phenomenon in the evolution of voles (Mammalia, Rodentia, Arvicolidae). Vestnik zoologii, 51:99-110.

Renaud, S., Michaux, J., Jaeger, J.-J., and Auffray, J.-C. 1996. Fourier analysis applied to Stephanomys (Rodentia, Muridae) molars: nonprogressive evolutionary pattern in a gradual lineage. Paleobiology 22:255-265.
https://doi.org/10.1017/S0094837300016201

Rey-Rodríguez, I., López-García, J.-M., Bennàsar, M., Bañuls-Cardona, S., Blain, H.-A., Blanco-Lapaz, Á., Rodríguez-Álvarez, X.-P., de Lombera-Hermida, A., Díaz-Rodríguez, M., Ameijenda-Iglesias, A., Agustí, J., and Fábregas-Valcarce, R. 2016. Last Neanderthals and first Anatomically Modern Humans in the NW Iberian Peninsula: Climatic and environmental conditions inferred from the Cova Eirós small-vertebrate assemblage during MIS 3. Quaternary Science Reviews, 151.
https://doi.org/10.1016/j.quascirev.2016.08.030

Rey-Rodríguez, I., Arnaud, J., López-García, J.-M., Stoetzel, E., Denys, C., Cornette, R., and Bazgir, B. 2021. Distinguishing between three modern Ellobius species (Rodentia, Mammalia) and identification of fossil Ellobius from Kaldar Cave (Iran) using geometric morphometric analyses of the first lower molar. Palaeontologia Electronica, 24:a01.
https://doi.org/10.26879/1122

Rey-Rodríguez, I., Gamarra, B., Arnaud, J., Golovanov, S., Kandel, A.W., Gasparyan, B., Wilkinson, K.N., Adler, D.S., and Weissbrod, L. 2024. Climatic variability in the Armenian Highlands as the backdrop to hominin population dynamics 50-25 ka. Palaeogeography, Palaeoclimatology, Palaeoecology, 648:112285.
https://doi.org/10.1016/j.palaeo.2024.112285

Rodríguez Álvarez, X. P., De Lombera Hermida, A., Fábregas Valcarce, R., and Lazuén Fernández, T. 2011. The Upper Pleistocene Site of Cova Eirós. In Hermida, A. De Lombera and Valcarce, R. Fábregas (eds.), To the West of Spanish Cantabria:the Palaeolithic Settlement of Galicia (British Ar, Issue Figure 1, pp. 123-132).

Román, J., Cabezas-León, S.M., Calzada, J., Giménez, J., Juste, J., Matutano, J., Palomo, L.J., De Paz, Ó., Puig-Montserrat, X., Rodríguez-Rodríguez, E.J., Rouco, C., Saavedra, C., Suarez-Bregua, P., and Tena, E. 2025. An Updated Checklist of the Mammal Species of Spain with Notes on Exclusions, Additions, and Taxonomic Changes. Galemys: Spanish Journal of Mammalogy, 37:29-61.
https://doi.org/10.7325/Galemys.2025.A4

Rufino, M.M., Gaspar, M.B., Pereira, A.M., and Vasconcelos, P. 2006. Use of shape to distinguish Chamelea gallina and Chamelea striatula (Bivalvia: Veneridae): Linear and geometric morphometric methods. Journal of Morphology, 267:1433-1440.
https://doi.org/10.1002/jmor.10489

Sagonas, K., Pafilis, P., Lymberakis, P., Donihue, C.M., Herrel, A., and Valakos, E.D. 2014. Insularity affects head morphology, bite force and diet in a Mediterranean lizard: Head morphology in Lacerta trilineata. Biological Journal of the Linnean Society, 112:469-484.
https://doi.org/10.1111/bij.12290

Seetah, K., Cucchi, T., Dobney, K., and Barker, G. 2014. A geometric morphometric re-evaluation of the use of dental form to explore differences in horse (Equus caballus) populations and its potential zooarchaeological application. Journal of Archaeological Science, 41:904-910.
https://doi.org/10.1016/j.jas.2013.10.022

Somoano, A. 2024. Update on the distribution of the fossorial water vole, Arvicola scherman, in Asturias and Leon, NW Spain. Galemys: Spanish Journal of Mammalogy, 36:1-5.
https://doi.org/10.7325/Galemys.2024.N7

Souto-Lima, R.B. and Millien, V. 2014. The influence of environmental factors on the morphology of red-backed voles Myodes gapperi (Rodentia, Arvicolinae) in Québec and western Labrador: Morphology of Red-backed Voles. Biological Journal of the Linnean Society, 112:204-218.
https://doi.org/10.1111/bij.12263

Stanley, S.M. 1973. An Explanation for Cope’s Rule. Evolution, 27:1.
https://doi.org/10.2307/2407115

Stoetzel, E., Denys, C., Michaux, J., and Renaud, S. 2013. Mus in Morocco: a Quaternary sequence of intraspecific evolution: Quaternary Evolution of Mus in Morocco. Biological Journal of the Linnean Society, 109:599-621.
https://doi.org/10.1111/bij.12065

Stoetzel, E., Cornette, R., Lalis, A., Nicolas, V., Cucchi, T., and Denys, C. 2017. Systematics and evolution of the Meriones shawii/grandis complex (Rodentia, Gerbillinae) during the Late Quaternary in northwestern Africa: Exploring the role of environmental and anthropogenic changes. Quaternary Science Reviews, 164:199-216.
https://doi.org/10.1016/j.quascirev.2017.04.002

Valenzuela-Lamas, S., Baylac, M., Cucchi, T., and Vigne, J.-D. 2011. House mouse dispersal in Iron Age Spain: a geometric morphometrics appraisal. Biological Journal of the Linnean Society, 102:483-497.
https://doi.org/10.1111/j.1095-8312.2010.01603.x