https://orcid.org/0000-0003-3684-0186
Figures
Abstract
Rodentia is the most species-rich order among mammals. The Republic of South Africa harbours a high rodent diversity whose taxonomy and phylogeny have been extensively studied using genetic tools. Such advances have led to the establishment of new faunal lists for the country. Because rodents are frequently recovered from archaeological cave site material and owl pellets, and constitute prime material for studying both past and present environmental conditions, it is necessary to characterize their osteological remains. The skull and teeth are the most useful diagnostic skeletal elements preserved in modern and fossil accumulations. This key provides updated craniodental criteria for identifying rodent genera found in Quaternary deposits, and modern material from the Republic of South Africa, thus facilitating research on past and present rodent diversity.
Citation: Linchamps P, Avery DM, Cornette R, Denys C, Matthews T, Stoetzel E (2023) Quaternary rodents of South Africa: A companion guide for cranio-dental identification. PLoS ONE 18(11): e0289812. https://doi.org/10.1371/journal.pone.0289812
Editor: Mohmed Isaqali Karobari, University of Puthisastra, CAMBODIA
Received: April 25, 2023; Accepted: July 26, 2023; Published: November 28, 2023
Copyright: © 2023 Linchamps et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting information files.
Funding: Funds were given for accessing specimens from Czech Republic and South Africa by the Partenariat Hubert Curien (PHC) Barrande and the international mobility program Transhumance of the doctoral school 227 “Sciences de la nature et de l’Homme” from the Muséum national d’Histoire naturelle-Sorbonne Université. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
The Republic of South Africa (RSA) has a rich storehouse of Quaternary archaeological sites, spanning the last 2.6 MYA from the Early Pleistocene to the late Holocene. It possesses one of the world’s richest fossil hominin inventories documenting the origin of modern humans, as well as an infinite amount of animal and plant fossils which have enabled a detailed palaeo-reconstruction of the environments in which we evolved. Small mammals, especially rodents, have been widely used for reconstructing past environments from the Pleistocene to the Holocene [e.g., 1–10]. They have an advantage in that they are frequently abundant in the fossil record, and, because they do not migrate and have small home ranges, generally provide a clearer and more detailed picture of local conditions than the larger mammals.
Rodents are a very diverse group of mammals, with an exceptional taxonomic and phenotypic diversity, and constitute the most species-rich order among mammals, making up over 40% of all living mammal species [11–13]. This variability in body shapes and sizes, together with morphological and physiological adaptations, resulted in the successful colonization of most terrestrial environments [12–14]. Rodents are found in almost every habitat around the world; from open, dry deserts, to thick, wet rainforests, and display a wide array of ecomorphological adaptations including fossorial (burrowing), saltatorial (jumping), arboreal, subaquatic, and gliding forms [15].
Rodents are the preferred prey of many predators, including owls and small and meso-carnivores, which are often the agents responsible for the accumulation of micromammal remains in archaeological and palaeontological cave sites [16–19]. For instance, diurnal and nocturnal raptors regurgitate pellets that contain the undigested parts (mostly hair, bones, and teeth) of their prey at roost and nest sites; those pellets may accumulate over a long period of time, and bones and teeth become incorporated into the sediment as they break down. Analysis of modern owl pellets thus provides useful comparative information for identifying cranial fragments and teeth of fossil Pleistocene microfauna.
Whether the remains are fossil or (sub)contemporaneous, the material used for identification are the mandibles and maxillae (these are seldom complete, and rarely retain all the teeth) and also isolated teeth. The most effective method for identification of specimens at species level is a morpho-anatomical comparison of the fossil material with a modern reference collection. However, these are not always easy to access, and modern, comparative collections in museums, for example, may contain specimens which are misidentified, poorly documented, or which have unresolved taxonomies. A solution to these problems is the use of an identification key. There are several available keys for the identification of southern African rodents [20–22] but they generally rely on external characteristics such as length of the body, proportions of the tail or hindfeet, color and pattern of the pelage or number of nipples, and are therefore of little use for identifying cranio-dental remains. The only exhaustive existing keys based on cranial characters are those of Coetzee [23] and De Graaff [24], which are not up to date in terms of taxonomic research, and some other publications cover only specific families or subfamilies [25–27]. More recently, the systematics of many taxa has been partially resolved, with several new species being described and published in taxonomic reference publications [12, 13, 22], as well as in many systematic studies based upon genetic analyses and field surveys [e.g., 28–31]. Many recently described taxa are missing from earlier keys and this, together with the problems related to obtaining reliable comparative collections, has prevented the correct identification, and effective palaeoenvironmental reconstructions of quaternary rodent assemblages.
Here we propose a new identification key of cranio-dental morphological features which are of generic diagnostic importance, and apply them to material extracted from owl pellets and Quaternary fossil assemblages. The key, which follows the systematics of Wilson et al. [12, 13], covers all genera that occurred in South Africa during the Quaternary, including extinct fossil taxa, as well as more recently introduced taxa. Since cryptic diversity has been showed to occur within several genera, and because many uncertainties remain regarding the validity and/or taxonomic status of some extinct species, we chose not to provide identification guidance beyond the genus level, pending further systematic, taxonomic, and morphometric investigations. To assist in the identification of specimens in the field, and to get an idea of the relative size of each genus, some synthetic plates with full-scale photos of modern specimens are provided as supplementary material (S1 Fig for upper jaws, S2 Fig for lower jaws, S1 Table for references of photographed specimens; the plates S1 and S2 Figs should be printed at actual size to maintain full-scale).
As this key is mostly dedicated to researchers working in palaeontology, archaeology, and taphonomy, diagnostic information is based on features and materials that are most often recovered from fossil or taphonomic sites, i.e., upper and lower toothrows, and mandibles and maxillae with, and without, teeth. Some other cranial features which are often useful for identifying rodents, such as the auditory bullae and width of the nasals, are not dealt with in detail here because they are not generally preserved on the material studied.
2. Past and present rodent diversity in South Africa
There are currently 35 genera of rodents living in RSA (see Table 1), with one additional valid fossil genus listed for the Quaternary. These 36 genera belong to nine families:
- The Muridae is the largest family of rodents, including many diverse species such of mice, rats and gerbils.
- Mole-rats of the family Bathyergidae are burrowing rodent with cylindrical bodies and short limbs that have distinct morphological adaptations to subterranean life.
- Dormice of the family Gliridae are small nocturnal rodents, largely arboreal and well adapted to climbing; in sub-Saharan Africa, they are only represented by the genus Graphiurus.
- Porcupines of the family Hystricidae are large, nocturnal rodents that have quills (spines) for defense against predators.
- The family Petromuridae is monospecific, it contains the dassie rat (Petromus typicus)—a medium size rodent which inhabits rocky outcrops.
- Springhares also belong to a single genus which constitutes the family Pedetidae; they are nocturnal rodents whose method of locomotion is hopping with their strong hind legs.
- The family Nesomyidae is diverse, and contains small to medium size morphologically varied rodents endemic to continental Africa and Madagascar.
- South African squirrels of the family Sciuridae are terrestrial (Geosciurus) or semi-arboreal (Paraxerus) diurnal rodents with an elongated body and bushy tail.
- Cane rats of the family Thryonomyidae are large, heavily built rodents that live in marshy areas and along riverbanks.
Taxa are listed in alphabetical order. Published extinct taxa are identified by the symbol † in the table.
In the genus accounts of this work, we provide a short synthesis of the most recent research on the phylogeny and geographical distribution of each genus and species, as well as an overview of the fossil record in South Africa during the Quaternary period.
3. General anatomy and glossary for the lower and upper jaws
Despite the significant number of species, and a great diversity of morphological and ecological adaptations, rodents are remarkably uniform regarding the general morphology of the skull and dentition. The upper and lower jaws (Fig 1) each support a single pair of large, ever-growing scalpriform incisors. The enamel is mainly limited to the outer surface of the incisors. The presence of iron in the mineral phase of incisor enamel can give the surface of some rodent incisors a yellowish to orange appearance which can be diagnostic for some genera; however, the incisor colour criterion is not always usable in an owl pellet and fossil context due to taphonomic alteration (digestion, diagenesis, staining with sediment, etc.). A large diastema is present between the incisors and the cheekteeth, allowing the lips to fold inwards in order to prevent debris interfering with chewing activity when gnawing. The maximum number of cheekteeth is four for the lower jaw, and five for the upper jaw, with murids having only three molars in each jaw and an incisor in each mandible and premaxilla. The pattern on the occlusal surface of molars is remarkably varied and diverse among rodents, and thus is of great diagnostic taxonomic value.
1: foramen magnum; 2: tympanic bulla; 3: zygomatic arch; 4: third upper molar (M3); 5: second upper molar (M2); 6: first upper molar (M1); 7: first or anterior palatal foramen; 8: diastema; 9: incisor; 10: second or posterior palatal foramen; 11: incisor; 12: body of mandible; 13: mental foramen; 14: first lower molar (M1); 15: second lower molar (M2); 16: third lower molar (M3); 17: coronoid process; 18: articular surface; 19: condylar process; 20: angular process.
Below is a glossary of the main terms that will be used in the following anatomical descriptions:
angular process. Also called processus angularis; process at the posterior lower corner of the mandible that serves for muscular attachment.
brachyodonty. Cheekteeth with short crowns.
bunodonty. Cheekteeth in which the cusps are high and rounded on the occlusal surface of the crown.
buno-lophodonty. Cheekteeth in which the cusps tend to connect transversally but remain partially individualized at low wear.
condylar process. Process at the posterior upper corner of the mandible that articulates with the glenoid fossa of the skull, forming the lower hinge of the jaw articulations.
coronoid process. Process at the anterior upper corner of the mandible ramus situated anteriorly to the condylar process. It serves as an attachment point for the temporalis muscle and does not participate in the jaw articulation.
cusp. Occlusal eminence on the surface of a tooth.
cusplet. Small cusp, often located on the edge of a tooth.
diastema. Gap between the incisor and the cheekteeth.
foramen. Orifice in a bone through which nerves and blood vessels pass.
hypselodonty. Ever-growing teeth with very long crown and short roots/rootless.
hypsodonty. Teeth with high crowns.
hystricognathous. Condition of the mandible with the angular process deflected lateral to the plane that includes the alveolus of the incisors (see Fig 2).
hystricomorphy. Type of rodent skull with enlarged infraorbital foramen (see Fig 3).
Arrows show the origin and the insertion of the masseter muscle (medial in red, lateral in blue).
incisor groove. Longitudinal groove running lengthways along the anterior enamel surface of lower and/or upper incisor of several rodent genera. The number of grooves (maximum 3) is often specific to a genus, sometimes to species (e.g., in Otomys and Parotomys).
incisor notch. Notching on the incisal worn surface of the upper incisors (see Fig 4).
incisor procumbency. Orientation of the upper incisors, defined by the position of the cutting edge in relation to the vertical plane of the incisor (see Fig 5).
lophodonty. Cheekteeth in which the cusps are fused to form transverse ridges (lophs).
macrodonty. Large teeth in proportion to the skull.
masseter knob. Small bony process located close to and below the anterior root of the zygoma (see Fig 6).
microdonty. Small teeth in proportion to the skull.
myomorphy. Type of rodent skull that combines a large zygomatic plate and a well-developed infraorbital foramen.
opisthodonty. Of the incisors, when the cutting extends posterior to the vertical plane (incisors are directed posteriorly; see Fig 5).
orthodonty. Of the incisors, when the cutting edge is perpendicular to the plane (incisors are directed more or less vertical plane; see Fig 5).
palatal foramina. Orifices in the bony palate for the transmission of palatine vessels and nerves. Anterior palatal foramina (also called incisive foramina) lie between the incisors and the cheekteeth. Posterior palatal foramina are situated between the two rows of cheekteeth.
posterior cingulum. Small, rounded structure of enamel on the posterior edge of the molar occlusal surface.
proodonty. Of the incisors, when the cutting edge extends anterior to the vertical plane (incisors are directed anteriorly; see Fig 5).
scalpriform. Shaped like a chisel.
sciurognathous. Condition of the mandible with the angular process in the same plane that the alveolus of incisors (see Fig 2).
sciuromorphy. Type of rodent skull with reduced infraorbital foramen and anterior part of the zygomatic arch developed into a large plate (see Fig 3).
stephanodonty. Cheekteeth in which ridges are connecting the various cusps in longitudinal rows.
4. Material and methods for constructing the key
4.1. Material
This key provides a simple step-by-step process for identifying rodents from fossil and owl pellet material in South Africa. For this work we have examined representative specimens from the collections of the Ditsong National Museum of Natural History, Republic of South Africa (DNMNH), Evolutionary Studies Institute of the University of the Witwatersrand, Republic of South Africa (ESI), Muséum national d’Histoire naturelle, France (MNHN), Musée royal de l’Afrique centrale, Belgium (RMCA), and the Institute of Vertebrate Biology, Czech Republic (IVB). List of examined museum specimens with associate catalogue numbers can be found in S1 Checklist. No permits were required for the described study, which complied with all relevant regulations.
This contribution has also relied heavily on information published in other works. For information related to biology, ecology and anatomy of the rodents, the comprehensive works of De Graaff [24], Happold [21], Monadjem et al. [22], and Wilson et al. [12, 13] have been of great value. In addition to these works, a multitude of publications are recommended for supplementing nomenclature and anatomical description of each rodent family; they are explicitly stated in each section through literature citations.
4.2. Illustrations
Teeth and skull photographs were taken using Nikon digital camera D 5500 coupled with AF-S Micro NIKKOR 60 mm and macro extension tubes. Photos were stacked using Helicon Focus 8.1.4 and edited in Adobe Photoshop CC 21.1.3. We have tried to show the intrageneric variability in dental anatomy by including photographs of several species in each genus description. Schematics drawings of rodent craniodental morphology, including illustrations of some distinctive features for identification, were realized with Adobe Photoshop CC 21.1.3 and Adobe Illustrator CC 21.0.0. Distribution maps of each species have also been included. These maps are based on distributional data published in Monadjem et al. [22], Wilson et al. [12, 13], data from IUCN (International Union for Conservation of Nature) red list database and range maps from various expert sources available on the Map of Life’s website at https://mol.org/ datasets/?dt = range&sg = Mammals [32]. They were designed using R software version 4.1.0.
4.3. How to identify families and genera of rodents
The first step of the key selects the name of the family to which a rodent belongs (Tables 2 and 3). For each family, further keys provide readers with a series of statements and two or three choices which will eventually lead to the correct identification of the organism. Some families contain only one genus in South Africa (Gliridae, Hystricidae, Pedetidae, Petromuridae and Thryonomyidae), facilitating easy identification. The keys to the genera (Tables 4 and 5, [32, 33, 43, 44]) are preceded by notes on the habits, preferred habitats, and potential predators, and are followed by a description of each genus and salient morphological characters. It contains dental and alveolar formulas (described below), images of the right upper and right lower molar rows, as well as distribution maps of each species in RSA. The reader may sometimes rely on geographical distributions in distinguishing genera; for example, in the case of Mastomys and Myomyscus, which are morphologically very close and have a limited overlapping distribution. However, the precision of distribution maps is limited for various reasons, such as species being falsely identified in the field or in collections (also morphologically cryptic species may not be distinguished), lack of knowledge of a taxon’s range, errors introduced in the georeferencing procedure, etc. Furthermore, the range of species and genera have been noted to sometimes change significantly over time [19], so these maps should be used with extreme caution when identifying fossil specimens.
We tried to use mainly discriminating criteria, which correspond to diagnostic characters of a genus and are present in all the individuals. As discriminating criteria are not always available in cranio-dental morphology for some taxa, or for broken specimens within the fossil or pellet material, we also propose secondary criteria, i.e. character states that are not absolute in terms of identifying to genus. Secondary criteria are sometimes absent in taxa that exhibit great intraspecific variation in size and shape, or may be subject to subjective interpretation. The presence of several secondary criteria can lead to the confident identification of a taxon, but identification must be based on as many characters as possible. We specify in the key when a listed character is likely to display variability.
A modified version of the key can be accessed online with Xper3, a free collaborative platform designed for computer-aided taxonomic description and identification [33, 34], at: https://rodentsouthafrica.identificationkey.org. The online key will be updated over time and integrate additional photographs for modern and fossil specimens.
We draw attention to the fact that the criteria used in this key are for South African taxa only and may not be applicable extralimitally. Furthermore, in an archaeological and palaeontological context, many specimens have been fragmented or damaged by taphonomic processes. Methods of genetic identification are not always feasible, and taxonomy based on geometric morphometrics is time-consuming and requires a certain amount of expertise. In such situations, the wisest attitude is to limit identification at a higher taxonomic determination level such as the family; it is inevitable that some specimens will be excluded from the fossil record.
4.3.1. Morphological characters.
In rodent jaws, the most important morphological character for genus identification is the tooth cusp pattern. Other diagnostic characters for the upper jaw include: number of cheekteeth, incisor morphology, location of the primary and secondary palatal foramina, origination of the zygomatic, presence of a masseter knob, etc., and for the lower jaw: location of the mental foramen, muscle attachment, shape of the mandible, projection of the mental, coronoid and condylar processes, etc. The number of tooth types is written as a dental formula, with the upper and lower teeth shown consecutively. Incisors (I) are indicated first, canines (C) second, premolars (P) third, and molars (M) fourth, giving the formula: I-C-P-M:I-C-P-M.
As loss of teeth is frequent in predator-derived assemblages, alveolar pattern also provides identification information for several taxa. Avery [6] published a useful key (which relies mostly on the size and number of alveoli) to distinguish mandibles of Wonderwerk micromammals in the absence of diagnostic teeth. We use here a similar alveolar-molar root formula to indicate the number of alveoli of the various Muridae and Nesomyidae genera, but here the small, round alveoli resulting from rootlets are counted as independent alveoli. The alveolar formula provides the number of alveoli of each cheektooth and is written in a similar way to dental formula, with upper and lower alveolar patterns consecutively. Thus, the formula 4-3-3:3-2/3-2 means that the upper molars M1, M2 and M3 have four, three, and three alveoli respectively, and the lower molars M1, M2 and M3 have three, two or three, and two alveoli respectively. There may be variability in the number of roots and rootlets of some taxa, so indication of the alveoli should be used as a guiding, but not absolute, criterion.
4.3.2. Size and measurements.
There is a great variation in size in South African rodents: the smallest (Mus indutus) weighs only 3 to 5 g and the largest (Hystrix africaeaustralis) weighs about 20 kg. The same is true for the size of the cheekteeth, and measurements of length and width can assist in identification at genus level. Fig 7 below presents measurements of the length of the upper toothrow (LUTR) and the length of the lower toothrow (LLTR) of each genus from modern museum collections (except for the extinct Proodontomys whose measurements were taken on fossils):
Red dashes represent medians. Numbers under the x-axis represent the number of specimens used for the measurements. The following genera were not included because both LUTR and LLTR are much greater: Pedetes, Thryonomys, and Hystrix (see measurement details in genus accounts).
We use four types of measurements in this key: LUTR = length of the upper toothrow (including molars and premolar(s) if present); WM1 = width of the first upper molar; LLTR = length of the lower toothrow; WM1 = width of the first lower molar. All measurements were taken with a digital calliper to the nearest 0.01 mm. The length and width of the teeth correspond to the maximum values along the mesiodistal and labiolingual axes of the teeth on the basis of the crown, as illustrated in Fig 8:
LUTR: length of the upper toothrow; WM1: width of the first upper molar; LLTR: length of the lower toothrow; WM1: width of the first lower molar.
Although these measurements may be useful for preliminary identification, they must not be considered absolute as some taxa may present biogeographic or temporal variations in size [9].
4.3.3. Comments on the age of the specimens.
This key is mostly intended for adult and sub-adult specimens, as juvenile or old specimens may not always key out correctly. In juvenile specimens, the size and eruption of the teeth may differ. This is especially true for larger species, which have a slower growth rate than smaller species and therefore reach maturity after a longer period. For instance, in Rhabdomys pumilio (mean weight = 45 g) the M1 and M2 starts erupting at the age of two weeks, while the M3 begins to erupt by four to five weeks of age [35]. In Hystrix africaeaustralis (weight up to 20 kg), deciduous premolars begin erupting at about 14 days of age, the M1 between 2 to 3 months, the M2 between five and six months, and the M3 at about 11 months; the permanent premolars are fully erupted at the age of two years [21]. The age at which species reach sexual maturity is also of concern, since most rodents are nidifugous (i.e., they leave the nest shortly after hatching or birth) and will leave the family nest and disperse once they are sexually mature. For instance, the Cairo spiny mice, Acomys cahirinus, become sexually mature and leaves the nest at the age of 2–3 months [36]. In Cryptomys hottentotus, the pups remain in the maternal burrow system for about 60 days until they are expelled from the burrow by the mother [36]. In Otomys sloggetti, males reach sexual maturity in 11 weeks, and females in 16 weeks, before dispersing [37] while subadult males of Geosciurus inauris do not disperse until eight months of age [38]. In older specimens, wear on the occlusal surface, which results in the removal of enamel and dentine, may obscure or obliterate the cusp pattern, and previously separate cusps may become fused. At a most advanced stage of wear, molars can be reduced to flattened, ovoid lobes with exposed dentine (Fig 9). At this stage, it is often more prudent to limit identification to the level of the family or subfamily.
4.3.4. Effects of digestion on tooth morphology.
The passage through the digestive tract of predators can cause corrosive damage to the teeth, which in turn may sometimes cause difficulties in taxonomic identification. Numerous works have been dedicated to categorizing predators into distinct categories based on their digestion patterns [16, 18, 39, 40]. Predators that cause heavy or extreme digestion modifications, such as diurnal raptors or small carnivores, can alter the enamel outline of molars, resulting in some cusps or other features of diagnostic importance almost eaten away or completely missing (Fig 10).
Enamel is locally removed, dentine is also affected.
5. Key to the families and genera of rodents in South Africa
The keys presented below for upper and lower jaws provides the main identification criteria to identify specimens to family level.
Muridae & Nesomyidae
Following Wilson et al. [12, 13], the family Muridae includes three subfamilies in RSA (Deomyinae, Gerbillinae, Murinae) and the family Nesomyidae includes four subfamilies (Cricetomyinae, Dendromurinae, Mystromyinae, Petromyscinae). Each subfamily of the Nesomyidae is morphologically well characterized, but there are no known morphological dental features that distinguish the family itself from the Muridae. We have therefore grouped the two families together. We use cusp nomenclature from Misonne [26] and Denys et al. [27] for describing the cusps of the lower and upper molars of Murinae and Dendromurinae:
In Muridae and Nesomyidae, the upper molars are usually the most diagnostic teeth. Denys & Michaux [41] grouped some African genera by the structure of the third upper molar; these structures are presented in Fig 11 for genera from South Africa. Unfortunately, skulls are more prone to fragmentation than mandibles, which explains why they are generally proportionately less well-represented in archaeological and paleontological assemblages. In the absence of teeth, the number and structure of alveoli, as well as the length of the anterior palatal foramina, provide useful criteria for identifying Muridae and Nesomyidae taxa (Fig 12).
Lower toothrow: alb: labial anteroconid; alc: anterolabial cingulum; alg: lingual anteroconid; hd: hypoconid; md: metaconid; pd: protoconid; pc: posterior cingulum; pl: prelobe; plc: posterolabial cusplet; tma: anteromedian cusp. Upper toothrow: t1: anterostyle; t2: lingual anterocone; t3: labial anterocone; t4: anterostyle; t5: protocone; t6: paracone; t8: pseudohypocone; t9: metacone (modified from Denys et al., 1992).
1) t3 present, t1 absent, t3 connected to the first row of cusps: Cricetomys, Dendromus, Malacothrix, Petromyscus, Saccostomus, Steatomys = Nesomyidae; 2) t3 present, t1 absent, t3 isolated: Acomys = Deomyinae; 3) t1 present, a distal cusp, t3 absent: some Aethomys, Mastomys, some Micaelamys, Myomyscus, Rattus, Rhabdomys; 4) t1 present, t3 absent, second lobe with 2 fused or distinct distal cusps: some Aethomys, Dasymys, Lemniscomys, some Micaelamys; 5) rather large t1, labial link between first lobe and second lobe: Zelotomys; 6) presence of t1 and tiny t3, trace of median longitudinal link: Grammomys, Thallomys; 7) small molars with t1 and link between first and second lobe, cusps poorly differentiated: Mus.
Subfamily DEOMYINAE Thomas, 1888
Genus Acomys I. Geoffroy, 1838 (Spiny Mice)
1) Upper (a) and lower (b) right toothrow of A. selousi (DNMNH-2833); 2) Upper (a) and lower (b) right toothrow of A. subspinosus (DNMNH-40989).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-3/2-2-2 (Figs 16 and 17).
Alveoli of the molars M2 are indicated in grey. Adapted and modified from [6].
Alveoli of the molars M2 are indicated in grey. Adapted and modified from [6].
Upper jaw.
Upper incisors are opisthodont and ungrooved. The palatal foramina taper to the t4 of the M1. Molars are small and show a superficially Mus-like cusp configuration, with the t1 displaced backwards (sometimes almost in line with t5 and t6) and the t4 subsequently low. M3 is rather small and has a t3 but no t1 (group 2 in Fig 12), in contrast to Dendromurinae and most Murinae, including Mus.
Lower jaw.
The M1 is the longest in the lower toothrow. The lingual anteroconid is approximately the same size as or slightly larger than the labial anteroconid. A posterolabial cusplet is sometimes present on the tooth. In South African species, the posterior cingulum is absent or very small in both the M1 and the M2. The molars M2 and M3 have two rows of cusps. The M2 displays a short anterolophid on its anterolabial zone, while the M3 displays a small anterolabial cusplet (more developed in A. selousi than in A. subspinosus) which tends to obliterate with wear (this cusplet is absent in Mus). The sciurognath mandible has a poorly developed coronoid process. The mandible is small, being of same average size than in Steatomys.
Systematic notes and South African fossil record.
Two species are currently recognized in South Africa:
- A. selousi (Forster, 1778)
- A. subspinosus (Waterhouse, 1838)
The species A. selousi has been described as a split from as A. spinosissimus Peters, 1852 [22, 42], the latter occurring further north in South-Central Africa. Further taxonomic and biogeographic investigation of these species is required [43]. Two additional fossil species have been described:
- †Acomys mabele Denys, 1990 known only from the Pliocene site of Langebaanweg, and that constitutes the first occurrence of the genus in the South African fossil record
- Acomys spinosissimus Peters, 1852 described in various Pleistocene sites
Fossils of this genus have been recorded from many Quaternary fossil deposits. Based on the recent split between A. selousi and A. spinosissimus, which are morphologically undistinguishable, South African fossil specimens previously attributed to A. spinosissimus may be more parsimoniously assigned to A. selousi.
Subfamily GERBILLINAE Gray, 1825
Genus Desmodillus Thomas & Schwann, 1904 (Cape Short-tailed Gerbils)
Upper (a) and lower (b) right toothrow of Desmodillus auricularis (DNMNH-39365).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-1:2-2-1 (Figs 16 and 17).
Upper jaw.
Upper incisors are long with a shallow groove. The anterior palatal foramina end short of the M1, the posterior palatal foramina extend between the second lamina of M1 to the beginning of M3. Molars show a semi-lophodont (or buno-lophodont) condition. M1 has three lobes, M2 has two, and M3 has one lobe only (species of Gerbilliscus (Gerbillurus) have one or two lobes). The central lobe in M1 has two circular cusps. Tympanic bullae are greatly inflated; they can even be seen when the skull is viewed from above. Fossil material that preserves only the toothrow is not always easily distinguished from Gerbillurus, although the length of the toothrow is greater in Desmodillus.
Lower jaw.
Lower incisors are plain. Young specimens have two cusps in the first lobe of the M1 but they fuse with the age to form the typical horseshoe shape of Taterillini; the two cusps of the second lobe of M1 remain unfused, while they are fused in a lamina in the third lobe. M2 has two lobes. M3 is very small and consists of one tiny cusp. The angular process of the mandible is sharp and elongated, and its ventral edge makes an angle with the ventral margin of the mandibular body. The coronoid process is higher than in Gerbillurus.
Systematic notes and South African fossil record.
The genus is monotypic:
- Desmodillus auricularis (Smith, 1834)
An additional fossil species has been described:
- †Desmodillus magnus Denys and Matthews, 2017 from Langebaanweg
Since the Early Pliocene, until the present, fossils of Desmodillus have been recorded from numerous fossil deposits.
Genus Gerbilliscus Thomas, 1897 (Gerbils & Hairy-footed Gerbils)
Subgenus Gerbilliscus (Gerbilliscus)
1) Upper (a) and lower (b) right toothrow of G. afra (DNMNH-21634); 2) Upper (a) and lower (b) right toothrow of G. brantsii (DNMNH-27755); 3) Upper (a) and lower (b) right toothrow of G. leucogaster (DNMNH-44280).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 4-2-1:4-2-1 (Figs 16 and 17).
Upper jaw.
Upper incisors are slightly opisthodont, yellow to orange in colour, and have a single groove. The anterior palatal foramina barely reach the alveolus of M1; the posterior palatal foramina are not as developed as in Desmodillus and Gerbilliscus. Molars show a lophodont condition, with cusps fused in transverse laminae lacking longitudinal connections. The molar M1 has three laminae, the first consisting of a single cusp; molars M2 and M3 both have two laminae when unworn.
Lower jaw.
Molars show lophodont condition, with cusps fused in transverse laminae that lack longitudinal connections. The anteroconid of M1 has a horseshoe shape whose appearance is highly variable. The M2 has two laminae, and the M3 has one. The mandible is elongated and seems vertically compressed, with coronoid and condylar processes projecting backwards.
Systematic notes and South African fossil record.
Species of the genus Gerbilliscus were previously included in Gerbillus, then in Tatera, but are now grouped in their own distinct genus [44, 45]. The genera Gerbilliscus and Gerbillurus have long been treated as two distinct genera, but recent molecular and chromosomal analysis suggest that they should be combined into one genus, Gerbilliscus [46]. In South Africa, three species of Gerbilliscus (excluding Gerbillurus) are currently recognised:
- Gerbilliscus afra (Gray, 1830),
- Gerbilliscus brantsii (Smith, 1836)
- Gerbilliscus leucogaster (Peters, 1852).
The first occurrences of this genus are from Makapansgat [47, 48] and Taung [1, 47, 49] in the Pliocene. Remains of Gerbilliscus are found in many deposits throughout the Quaternary [50].
Subgenus Gerbilliscus (Gerbillurus)
1) Upper (a) and lower (b) right toothrow of G. paeba (DNMNH-32636); 2) Upper (a) and lower (b) right toothrow of G. vallinus (DNMNH-32580).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 4-2-1:4-2-1 (Figs 16 and 17).
Upper jaw.
Upper incisors are strongly opisthodont, yellow to orange, and have a single groove. The anterior palatal foramina end short of the M1, the posterior palatal foramina extend between the second lamina of M1 to the beginning of M3. Molars show semi-lophodont condition, with cusps fused in transverse laminae that lack longitudinal connections. M1 has three lobes, M2 has two lobes, and M3 has two lobes, poorly, or completely, fused (in Desmodillus the second lobe is always much reduced). Fossil material that preserves only the teeth is not always easily distinguishable from Desmodillus, although the length of the toothrow is smaller in Gerbilliscus.
Lower jaw.
As opposed to Gerbilliscus (Gerbilliscus), cusps are high and round and do not fuse together in transverse laminae as in other Gerbilliscus spp. Lower molars are very similar to those of Desmodillus, but the M1 has four alveoli and the incisor is thinner. Mandible is small, vertically compressed and not as high as in Desmodillus.
Systematic notes and South African fossil record.
Until recently, specimens of Gerbillurus were placed in their own genus based on the marked morphological difference with other Gerbilliscus, but recent chromosomal and molecular analyses indicate they should be assigned to Gerbilliscus. In South Africa, two species of Gerbilliscus (Gerbillurus) are currently recognized:
- Gerbilliscus paeba (A. Smith, 1836)
- Gerbilliscus vallinus (Thomas, 1918)
The oldest remains of Gerbillurus in the South African fossil record are found in the Early Pleistocene site of Wonderwerk [6, 51].
Subfamily MURINAE Illiger, 1811
Genus Aethomys Thomas, 1915 (Veld Rats)
1) Upper (a) and lower (b) right toothrow of A. chrysophilus (DNMNH-4659); 2) Upper (a) and lower (b) right toothrow of A. ineptus (DNMNH-46903).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 4-3-3:4-3-3 (Figs 16 and 17).
Upper jaw.
Incisors are ungrooved and opisthodont. The anterior palatal foramina are long and extend beyond the level of the M1. Molars are relatively broad. In M1, the t1 is situated slightly or well behind t2 and t3; there is often a small stephanodont crest uniting t6 and t9 on M1 and on M2; t7 is absent. The configuration of M3 corresponds to group 3 or 4 in Fig 12, with the distal lobe having often one elongated cusp (or two poorly differentiated, as opposed to two well differentiated in Lemniscomys) although this criterion shows variability in few specimens.
Lower jaw.
The M1 sometimes display lateral cusplets or small ridges that can form a stephanodont crest; it has no conspicuous posterior cingulum (a condition similar in Micaelamys and Lemniscomys), although this feature is not constant. In the M1 there is no tma (it is present in Micaelamys) but some specimens may display a tiny anteromedian cusplet. The M2 typically has three alveoli (two isolated and two fused ones).
Systematic notes and South African fossil record.
Species of the genus Micaelamys were previously included in Aethomys, but they were later placed in their own genus based on molecular and morphological data [52]. Two species of Aethomys are currently recognized in South Africa:
- Aethomys chrysophilus (de Winton, 1897)
- Aethomys ineptus (Thomas and Wroughton, 1908)
Additional fossil species have been identified:
- †Aethomys adamanticola Denys, 1990 from the Early Pliocene locality of Langebaanweg
- †Aethomys modernis Denys, 1990 from the Early Pliocene locality of Langebaanweg
Fossils of Aethomys are known from many Pleistocene and Holocene sites from South Africa [50]. Many of these specimens were identified as Aethomys chrysophilus at a time when A. ineptus was not recognized yet as a cryptic species distinct from A. chrysophilus. As these two species appear to be morphologically indistinguishable using the cranio-dental anatomy, caution is recommended when making an osteological identification
Genus Dasymys Peters, 1875 (Shaggy Rats)
1) Upper (a) and lower (b) right toothrow of D. capensis (DNMNH-26274); 2) Upper (a) and lower (b) right toothrow of D. robertsii (DNMNH-30639).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 6-5-3:4/5-3/4-3 (Figs 16 and 17).
Upper jaw.
Upper incisors are broad and ungrooved. The palatal foramina end just before the alveolus of the first root of the M1. Molars are large and heavily cusped, exhibiting macrodonty. Rows of cusps are arranged in transverse rows, with t1 in line with t2 and t3 in M1. With age, rows of tubercles are obliterated, and the spaces between the original rows of cusps may isolate as characteristic enamel islands. The configuration of M3 corresponds to Group 4 in Fig 12.
Lower jaw.
Lower incisors are ungrooved. Molars are large and heavily cusped, showing macrodonty. M1 has three cusps on the prelobe, and no posterior cingulum (or a tiny one). Most of the cusps are fused in laminae that tend to isolate as enamel islands with the age. M2 and M3 have two lobes each and are large. The posterior root of the M3 is large. The mandible is rather massive and high.
Systematic notes and South African fossil record.
The number of species recognised within this genus has changed several times over the past decades, and species limits are still not fully resolved [22]. Currently, three species are described in South Africa:
- Dasymys capensis Roberts, 1936
- Dasymys incomtus (Sundevall, 1847)
- Dasymys robertsii Mullin et al., 2004
Both D. capensis and D. robertsii were previously identified as D. incomtus. Although the three species should be distinguishable on a craniometric basis (if the preservation of the fossils allows it), most of the fossils of Dasymys were identified at a time when only D. incomtus was recognized. This material therefore should be re-evaluated. Additional fossil species from South African deposits have been described:
- †Dasymys bolti Broom [Unpublished]
- †Dasymys broomi Broom [Unpublished]
- †Dasymys lavocati Broom [Unpublished]
D. bolti was described by Denys [47], but the other two fossil species D. broomi and D. lavocati have not been published and are considered as invalid species [4].
Genus Grammomys Thomas, 1915 (Thicket Rats)
1) Upper (a) and lower (b) right toothrow of G. cometes (DNMNH-40467); 2) Upper (a) and lower (b) right toothrow of G. dolichurus (DNMNH-10403).
Note that there is a significant size difference between the two species.
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 4/5-4/5-3:2/4-3/4-3 (Figs 16 and 17).
Upper jaw.
Upper incisors are ungrooved and orthodont. The anterior palatal foramina are long and penetrate between the first upper molars. The palate is broad, and the molars show microdonty. The cusps are well separated from each other and linked by stephanodont crests. In the M1, t1 is very slightly behind t2 and t3, and t4 is slightly behind t5 and t6. The M1 has also a small crestiform t7, and an accessory anterior median cusp is occasionally present. The cusp t3 is reduced or absent in M2, and the t9 is well visible and situated close to the t6. On the M3, presence of a t1 and a tiny t3, and trace of median longitudinal link (group 6 in Fig 12).
Lower jaw.
Lower incisors are ungrooved. Molar cusps are high and well separated. There are three cusps on the first lobe of M1, a stephanodont crest uniting the first lobe with second and third lobes, and a conspicuous posterior cingulum. The M2 has two lobes with two cusps each, a small anterolabial cusp and lateral cusplets or small ridges that can form stephanodont crest, as well as a posterior cingulum. The M3 displays a tiny antero-external cusp.
Systematic notes and South African fossil record.
Grammomys was previously considered as subgenus of Thamnomys but this genus is now elevated to genus rank based on morphological and phylogenetic analysis [14, 21, 53]. According to Monadjem et al. [22], this genus critically needs revision. Two species are currently recognized in South Africa:
- Grammomys cometes Thomas & Wroughton, 1908
- Grammomys dolichurus (Smuts, 1832)
Fossils are known from Pliocene to Holocene deposits, with the oldest occurrence in Makapansgat Limeworks [1, 49].
Genus Lemniscomys Trouessart, 1881 (Grass Mice)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 5-5-4:5-5/6-3 (Figs 16 and 17).
Upper jaw.
Incisors are opisthodont. The anterior palatal foramina generally reach the level of the t2 in M1. Molars are large, with broad central cusps (t2, t5, t8); this feature is mostly patent in the M2. In molar M1, the t4 is not connected to the t8 by a small crest as in Aethomys and is usually placed higher, sometimes in line with the cusp t5. There can be a small stephanodont crest uniting t6 and t9 on M1 and on M2. The M3 is relatively large, its distal lobe has two well differentiated cusps (as opposed to one or two fused cusps in Aethomys or Micaelamys); it belongs to Group 4 in Fig 12.
Lower jaw.
The distinction between Aethomys and Lemniscomys based on the lower toothrow is sometimes not straightforward. In Lemniscomys the lateral cusplets are rarely well marked, while Aethomys often displays lateral cusplets or small ridges that can form some stephanodont crest. Molar size is overlapping between the two genera. As with Micaelamys and Aethomys, the M1 has no clear cingulum posterior. The M2 has two rootlets, resulting in six alveoli.
Systematic notes and South African fossil record.
There is only one species recognized in South Africa:
- Lemniscomys rosalia (Thomas, 1904)
Fossils are known from the Early Pleistocene in various karst deposits from the Sterkfontein Valley [50].
Genus Mastomys Thomas, 1915 (Multimammate Mice)
1) Upper (a) and lower (b) right toothrow of M. coucha (DNMNH-45433); 2) Upper (a) and lower (b) right toothrow of M. natalensis (DNMNH-37511).
Dental formula is 1-0-0-3:1-0-03. Alveolar formula is 3-3-2:2-2-2 (Figs 16 and 17).
Upper jaw.
Mastomys is morphologically very similar to Myomyscus but has a much wider distribution. Upper incisors are smooth and orthodont. The anterior palatal foramina are long and reach the t1 of the M1. Rows of cusps are rather distorted in the M1, with the t1 located behind t2 and t3, and the t4 also located behind t5 and t6. As in Myomyscus and Zelotomys (two other members of the Praomyini tribe), the t9 is well separated from the t6, and located far in the labial side. The M1 has three roots. The M2 has a t1 and a small t3, and the t9 is low and well distinct. The M3 has two lobes and corresponds to Group 3 in Fig 12.
Lower jaw.
Lower incisors are smooth. Molars show a typical Praomyini pattern (as Myomyscus and Zelotomys). Both M1 and M2 have a posterior cingulum. The M1 displays a small posterior cingulum and a posterolabial cusplet (plc). The M2 has both anterolabial and posterolabial cusplets. The mandible is approximately the same size as in Rhabdomys, but both genera can be distinguished based on the number of alveoli when teeth are absent (alveolar formula is 2-2-2 in Mastomys, 4-5-3 in Rhabdomys). As in Myomyscus and Zelotomys, the alveolar region of the mandible is well-developed in proportion to the rest of the mandible.
Systematic notes and South African fossil record.
Mastomys was previously included in Praomys, but is now placed in its own genus [22]. Two species are currently recognized in South Africa:
- Mastomys coucha (Smith, 1834)
- Mastomys natalensis (Smith, 1834)
Fossils of Mastomys have been recovered from many Pleistocene fossil deposits from South Africa. Most of this material is attributed to M. natalensis, but the remains were identified at a time when M. coucha was not recognized. These two species appear undistinguishable using cranio-dental features [54], so specimens from the Early Pleistocene should be conservatively attributed to Mastomys sp.
Genus Micaelamys Ellerman, 1941 (Lesser Veld Rats)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 4/5-3/4-3:3-3-3 (Figs 16 and 17).
Upper jaw.
Incisors are opisthodont. In the M1, cusps are few distorted with t1 slightly behind t2 and t3, and t4 behind t5 and t6. There are sometimes traces of stephanodonty. The M3 lacks the t3 its distal lobe has one elongated cusp (or 2 very few differentiated, as opposed to two well differentiated in Lemniscomys); it corresponds to Group 3 in Fig 12. Molars are smaller than in Aethomys and Lemniscomys.
Lower jaw.
M1 has always three cusps in the anterior lobe (presence of a well-developed tma), as opposed to two in most specimens of Aethomys (some Aethomys display a very small anteromedian cusplet). As Aethomys and Lemniscomys, the M1 has no posterior cingulum or a very small one in few specimens. Similarly, the M2 has a very small pc contrary to Aethomys where there is often a round median pc. A row of additional cusplets often occurs on the labial side of the M1 and M2.
Systematic notes and South African fossil record.
Species of this genus were previously included within Aethomys but are now placed in their own genus based on morphological and molecular ground [28, 52, 55]. Two species are currently recognized in South Africa:
- Micaelamys granti (Wroughton, 1908)
- Micaelamys namaquensis (A. Smith, 1834)
The genus Micaelamys has been identified in many Pleistocene deposits in South Africa, often referred to as Aethomys (for instance, Aethomys namaquensis). The oldest Aethomys cf. namaquensis has been described by Pocock [48] in Makapansgat.
Genus Mus Linnaeus, 1758 (Old World Mice and Pygmy Mice)
1) Upper (a) and lower (b) right toothrow of M. indutus (DNMNH-45737); 2) Upper (a) and lower (b) right toothrow of M. musculus (MNHN-ZM-MO-1994-2350).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-1/2:2-2/3-1/2 (Figs 16 and 17).
Upper jaw.
Upper incisors are ungrooved and opisthodont, with the worn posterior surface characteristically notched for the three South African species. The palatal foramina reach the first or the second row of cusps in M1. The skull is small, and so are the molars. The cusps on M1 are distorted, with t1 situated far posterior of t2 and t3. Molar M3 is small with a reduced cusp configuration: it has a t1, but cusps are poorly differentiated and there is a link between first and second lobe (Group 7 in Fig 12). Presence of a masseteric knob.
Lower jaw.
The M1 has a typically enlarged lingual anteroconid that distorts the tooth, while the labial anteroconid is smaller and close to the protoconid. The M3 is tiny and has one or two lobes with one or two corresponding alveoli (always two lobes and two alveoli in Acomys). The mandible is small, being of the same average size as Dendromus; the M2 is often 3-rooted in Mus whereas it is 2-rooted in Dendromus.
Systematic notes and South African fossil record.
Four species are currently recognized in South Africa:
- Mus indutus (Thomas, 1910)
- Mus minutoides Smith, 1834
- Mus musculus Linnaeus, 1758
- Mus neavei (Thomas, 1910)
M. musculus belongs to the subgenus Mus, while M. indutus, M. minutoides and M. neavei belong to the subgenus Nannomys, formerly named Leggada [e.g., 24, 56]. The genus Mus is known from the Pliocene of Makapansgat [48] and has been recovered from many Pleistocene sites. The species M. musculus was introduced and is thus not found in the Pleistocene deposits. The oldest and single record of M. musculus in Holocene deposits is from Hope Hill Shelter dated to 4,400±100 B.P. [57]
Genus Myomyscus Shortridge, 1942 (Meadow Mice)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-2:2-2-2 (Figs 16 and 17).
Upper jaw.
Myomyscus is morphologically very similar to Mastomys and is mostly diagnosable by its restricted distribution to the Fynbos biome. The anterior palatal foramina reach the t1 of the M1. The rows of cusps are slightly distorted in the M1, with the t1 located behind t2 and t3, and the t4 also located behind t5 and t6. As with other members of the Praomyini tribe, the t9 is well separated from the t6, and located far in the labial side. The M1 has three roots. The M2 has a t1 and a small t3, and the t9 is low and well distinct. Like in Mastomys, the M3 is large with a t1 and a t3 and two lobes and corresponds to Group 3 in Fig 11.
Lower jaw.
As in Mastomys, molars show a typical Praomyini pattern. Both M1 and M2 have a posterior cingulum. The M1 displays a small posterior cingulum and a posterolabial cusplet (plc). The M2 has both anterolabial and posterolabial cusplets. The mandible is of average size.
Systematic notes and South African fossil record.
A single species is recognized in South Africa:
- Myomyscus verreauxii (Smith, 1834)
In the past, this species had been placed in the genera Praomys and Myomys [22]. In South Africa, remains of Myomyscus are known since the Late Pliocene [50, 58].
Genus Otomys Cuvier, 1824 (Vlei Rats or Laminate-toothed Rats)
1) Upper (a) and lower (b) right toothrow of O. angoniensis (MNHN- ZM-2020-574 BO24); 2) Upper (a) and lower (b) right toothrow of O. laminatus (DNMNH-4647); 3) Upper (a) and lower (b) right toothrow of O. sloggetti (DNMNH-7781).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is variable, for instance 3-3-11:7-4-3 (Figs 16 and 17).
Upper jaw.
Incisors are opisthodont and always grooved (the number of grooves varies by species). The anterior palatal foramina rarely reach the M1. The palate is very narrow with broad region of contact between the two halves. The cusps of the cheekteeth are fused into several transverse laminae: M1 has three laminae; M2 has two; M3 is the longest cheekteeth, having four or more laminae. The auditory bullae are hardly, or not, inflated.
Lower jaw.
Lower incisors have one deep or faint groove, or no groove at all. The M1 is the largest molar of the toothrow, having between three (O. unisulcatus) and seven (O. laminatus) laminae. The M2 and the M3 have two laminae each. The mandible is stocky with a broad ascending ramus.
Systematic notes and South African fossil record.
There has been debate about the position of Otomys within its own subfamily Otomyinae, but most recent phylogenetic analyses nest it within the Murinae [28, 55]. There has also been debate about the genus boundaries of Otomys, Parotomys, and Myotomys. The latter genus, which comprised M. unisulcatus and M. sloggetti, has been proved paraphyletic [59]. For this reason, only the two genera, Otomys and Paratomys, are currently recognized [22]. Recent systematic analyses have revealed multiple cryptic species, which resulted in a considerable increase in the total number of species currently recognized in South Africa:
- Otomys angoniensis Wroughton, 1906
- Otomys auratus Wroughton, 1906
- Otomys irroratus (Brants, 1827)
- Otomys karoensis Roberts, 1931
- Otomys laminatus Thomas & Schwann, 1905
- Otomys sloggetti Thomas, 1902
- Otomys unisulcatus F. Cuvier, 1829
Further taxonomic investigation is required for this genus. Additional fossil species have been described from the fossil deposits of South Africa:
- †Myotomys campbelli Broom and Schepers, 1946 from the Pliocene deposit of Taung
- †Otomys gracilis Broom, 1937 identified in many Pleistocene karstic deposits from the Sterkfontein Valley
The fossil species †O. gracilis is regarded by Avery [4] as a synonym of modern O. saundersiae. The Otomys fossils need to be revised in light of recent advancements in systematic research.
Genus Parotomys Thomas, 1918 (Whistling Rats)
1) Upper (a) and lower (b) right toothrow of P. brantsii (DNMNH-22612); 2) Upper (a) and lower (b) right toothrow of P. littledalei (DNMNH-22446).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 4/5-3/4-3/4:4/5-3/4-3/4 (Figs 16 and 17).
Upper jaw.
Incisors are opisthodont, ungrooved in P. littledalei and grooved (a single groove) in P. brantsii. The anterior palatal foramina do not reach the M1. Cusps of the cheekteeth are fused into a number of transverse laminae: M1 has three laminae; M2 has two; M3 is the longest cheekteeth, having three (P. brantsii) or two (P. littledalei) differentiated laminae and two distal laminae fused by enamel ridge (while they remain separate in Otomys). Auditory bullae are distinctively inflated (hardly or not inflated in Otomys).
Lower jaw.
Lower incisors are ungrooved. The M1, which is the longest tooth of the toothrow, has four laminae, the first two most anterior of which are joined by an enamel ridge. The M2 and the M3 have two laminae each.
Systematic notes and South African fossil record.
Currently, there are two species recognized in South Africa:
- Parotomys brantsii (A. Smith, 1834)
- Parotomys littledalei (Thomas, 1918)
Fossils of Parotomys have been recorded in numerous South African deposits since the Middle Pleistocene [50, 58].
1) Upper (a) and lower (b) right toothrow of R. norvegicus (MNHN-ZM-MO-1888-382); 2) Upper (a) and lower (b) right toothrow of R. rattus (MNHN-ZM-MO-1981-352).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 5-4-3/4:4-3/4-3 (Figs 16 and 17).
Upper jaw.
Upper incisors are smooth and orthodont. The anterior palatal foramina are rather short, barely reaching short of the root of M1. The palate is broad, extending well beyond the M3. The dentition is proportionally small with relatively narrow teeth. The central cusps of the upper molars are pronounced. In the M1, the t1 is displaced backwards to the t2 and connected to it by a ridge, and cusp t7 is absent. In both M1 and M2, t9 is reduced and very close to the t8 (forming a single lamina when worn). In M2, the t3 is small or absent. The M3 displays a prominent t1 but lacks t3, corresponding to Group 3 in Fig 12; it is not markedly smaller than the M2.
Lower jaw.
The alveolar region of the mandible is well-developed in relation to the whole mandible. Both the M1 and the M2 have a posterior cingulum (often oval-shaped); they lack anterior cusps but up to three small posterolabial cusps may be present.
Systematic notes and South African fossil record.
This genus was unintentionally introduced to South Africa. Three species are currently recognized:
- Rattus norvegicus (Berkenhout, 1769)
- Rattus rattus (Linnaeus, 1758)
- Rattus tanezumi Temminck, 1844
R. tanezumi was first discovered in South Africa by Bastos et al. [60] from two sites in Limpopo Province, and has been identified since then from further localities. There is debate about the species status of this taxon, and some authors classify it rather as belonging to a lineage of R. rattus [61]. The three species R. norvegicus, R. rattus and R. tanezumi are commensal to humans. According to archaeological data, it seems that R. rattus followed human migrations into southern Africa during Iron Age [62] and that R. norvegicus may have arrived from European ships during the 19th century [63].
Genus Rhabdomys Thomas, 1916 (Four-striped Grass Mice)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 5-5-3:4-4/5-3 (Figs 16 and 17).
Upper jaw.
Upper incisors are ungrooved and orthodont to opisthodont. The anterior palatal foramina reach just to the anterior root alveolus of the M1. Upper cheekteeth display large central cusps. In the M1, the t1 is located behind t2 and t3, and t4 behind t5; the second and third rows of cusps appear to be linked on the lingual and labial sides. In both the M1 and M2, the t9 is small and often reduced to a small ridge that connects with the t6. The M3 lacks the t3 and has one distal cusp. Length of upper cheekteeth is similar to Mastomys and Myomyscus, but the number of roots and alveoli is different (M1 is 5-rooted in Rhabdomys, 3-rooted in Mastomys and Myomyscus).
Lower jaw.
The M1 has two anterior cusps and an additional small tma is often present. Both M1 and M2 have labial cusplets and reduced to absent posterior cingulum. The lower dentition is of similar size than Mastomys but both genera can be distinguished almost unambiguously on the basis of the number of roots and corresponding alveoli: the M1 of Rhabdomys is 4-rooted, with the central pair of roots usually visible on the sides of the tooth in situ, while the M1 of Mastomys is 2-rooted.
Systematic notes and South African fossil record.
Until recently, the genus was considered as monotypic [21] with R. pumilio being the only known species, but new karyotypic and genotypic analyses revealed a more complex diversity [64, 65]. Today, four species are recognized in South Africa:
- Rhabdomys bechuanae (Thomas, 1893)
- Rhabdomys dilectus (De Winton, 1897)
- Rhabdomys intermedius (Wroughton, 1905)
- Rhabdomys pumilio (Sparman, 1784)
The oldest known Rhabdomys have been discovered around 5 MYA in Langebaanweg [66] and 3.3 MYA in Makapansgat [48]. Fossils are known from a variety of Pleistocene localities. This material was identified as R. pumilio at a time when the specific diversity of Rhabdomys had not been recognized and is likely to include material attributable to other species.
Genus Thallomys Thomas, 1920 (Acacia Rats or Tree Rats)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 5-4/5-3/4:4/5-4-3 (Figs 16 and 17).
Upper jaw.
Incisors are orthodont and ungrooved. The palatal foramina are large and penetrate between the second rows of cusps on M1. Molars have an embossed, angular appearance, and show microdonty. In the M1, the t7 is absent or much reduced, and there are small stephanodont crests on t1 and t3 and joining t6 with t9. The central cusps (t2, t5 and t8) are prominent. In the M2 the cusps t2 and t7 are absent. The molar M3 is the smallest of the toothrow, but not greatly reduced; it corresponds to Group 6 in Fig 12.
Lower jaw.
Cusps of lower molars are prominent and sharply defined. The M1 and M2 have equal-sized posterior cingula, and a longitudinal crest that connects the lobes on the labial side. The M1 often displays small labial extra cusplets or ridges, and sometimes a small anteromedian cusplet. Both the M2 and M3 have antero-external cusplets. The toothrow is relatively small compared to the size of the mandible (microdonty).
Systematic notes and South African fossil record.
According to Monadjem et al. [22], this genus requires urgent revision. Currently, three species are listed in South Africa:
- Thallomys nigricauda Thomas, 1882
- Thallomys paedulcus (Sundevall, 1846)
- Thallomys shortridgei Thomas and Hinton, 1923
An additional fossil species has been described in South Africa:
- † Thallomys debruyni Broom, 1948
The oldest known fossil of Thallomys was found in Langebaanweg around 5MYA [66]. Fossils of this genus are known from various Pleistocene localities in South Africa.
Genus Zelotomys Osgood, 1910 (Broad-headed Mice)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-2/3:2-2-2 (Figs 16 and 17).
Upper jaw.
Upper incisors are ungrooved and opisthodont. The palatal foramina reach the second root of the three-rooted M1. The labial row of cusps is well developed and tend to project outwards in a characteristic way. In the M1, t1 and t2 tend to fuse with wear. As with other members of the Praomyiny tribe (Mastomys and Myomyscus), the t9 is well separated from the t6, and located far in the labial side. The M2 is as broad as or broader than long, with a low t9 and a t3 very small, often reduced to a tiny process. The M3 is markedly reduced, with traces of two laminae (group 5 in Fig 12) and a big t1.
Lower jaw.
Lower incisors are plain and ungrooved; they are relatively strong, with the alveolar region of the mandible well-developed, and generally extend far beyond the alveolus. The prelobe of the M1 has two nearly longitudinal, poorly differentiated cusps that are connected to those of the second lobe, together forming a characteristic compact “trefoil” like pattern, while the third lobe is located further below. There is a small cingulum posterior on M1 and M2. The M3 is reduced.
Systematic notes and South African fossil record.
Currently, a single species is recognised in South Africa:
- Zelotomys woosnami (Schwann, 1906)
The oldest known fossil of Zelotomys was found in Langebaanweg around 5MYA [66]. Fossils of this genus are known from various Pleistocene localities in South Africa.
Family NESOMYIDAE Major, 1897
Subfamily CRICETOMYINAE Roberts, 1951
Genus Cricetomys Waterhouse, 1840 (Giant Pouched Rats)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-3:2-2/3-2/3 (Figs 16 and 17).
Upper jaw.
Upper incisors are opisthodont and ungrooved but display characteristic striations on their anterior enamel surface, forming a slightly raised band (this feature is also present in Saccostomus and Mystromys). The anterior palatal foramina are short, located before the basis of the zygomatic arches. Molars are large, with high, well-separated cusps. The M1 has three lobes, The M2 and M3 two lobes each. In the first lobe of M1, the t2 and t3 are large and connected, being well separated from the t1 which is smaller and isolated below; similarly in the second lobe, the t5 and t6 are large and separated from the t4; the third lobe has no t7, but a large posterior cingulum. The M2 has two lobes, with t4 separated from t5 and t6, and a conspicuous posterior cingulum. The M3 is almost as large as the M2 and corresponds to Group 1 in Fig 12.
Lower jaw.
Lower incisors are ungrooved but display the same ridged band as the upper incisors. The molar M1 has three lobes: the first lobe has one elongated cusp, the second and third lobes have two cusps each that are aligned transversally. It has a well-developed posterior cingulum and a variable number (generally 2) of supplementary cusps on the labial side. As in Saccostomus (another member of the subfamily Cricetomyinae), the coronoid process is well-developed while the angular process is not projected backwards.
Systematic notes and South African fossil record.
South African specimens of Cricetomys were previously attributed to C. gambianus Waterhouse, 1840, but recent morphological and molecular analyses revealed the existence of multiple species [67]. Today, the specimens recorded in Southern Africa are attributed to a single species:
- Cricetomys ansorgei Thomas, 1904
Fossils of Cricetomys have been identified from very few Quaternary deposits in South Africa. Its oldest record is from the Late Pleistocene from Sibudu [68] and Rose Cottage Cave [69].
Genus Saccostomus Peters, 1846 (Pouched Mice)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-3:2-2-2 (Figs 16 and 17).
Upper jaw.
Upper incisors are opisthodont and ungrooved, but they display characteristic striations on their anterior enamel surface, forming a slightly raised band (this feature is also present in Cricetomys and Mystromys). The anterior palatal foramina reach the anterior root of the first upper molar. Molars have bulbous cusps that are connected transversally. The M1 has three lobes with only the outer two cusps in the first row (it lacks the t1), and three cusps on the second and third rows. Both the M1 and the M2 have a relatively well-developed posterior cingulum. The M2 and M3 have two lobes each. The M3 is not greatly reduced, as in the Dendromurinae; it corresponds to Group 1 in Fig 12.
Lower jaw.
Lower incisors are ungrooved but display the same ridged band than in the upper incisors. The M1 has three lobes and a small posterior cingulum. The first lobe consists of a single elongated cusp (sometimes two poorly differentiated cusps are visible), which may be connected to the second row of cusps. The second and third lobes have two fused cusps each. The M2 has two lobes, with an antero-external cusps of variable size and a posterior cingulum. The M3 has two lobes with a tiny antero-external cingulum.
Systematic notes and South African fossil record.
One species is currently recognized in South Africa:
- Saccotomus campestris Peters, 1846
This genus seems absent from South Africa until the late Pleistocene, with the oldest remains discovered in Border Cave around 0.2 MYA [70], in Sterkfontein StP6 around 0.1 MYA [5], in younger levels of Makapansgat Cave of Hearths [1] and in Gladysvale Pink Breccia, S18.E6, and S19.6 of uncertain age [71].
Subfamily DENDROMURINAE G.M. Allen, 1939
Genus Dendromus Smith, 1829 (African Climbing Mice)
1) Upper (a) and lower (b) right toothrow of D. mesomelas (DNMNH-2445); 2) Upper (a) and lower (b) right toothrow of Dendromus sp. (ESI modern owl pellet collection).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 4-3-1:2-2-1 (Figs 16 and 17).
Upper jaw.
Upper incisors are opisthodont and have one pronounced groove, located closer to the outer margin than the inner (a character shared with other Dendromurinae). There is a masseteric knob at the lower anterior corner of the zygomatic plate. Anterior palatal foramina extend beyond the first row of cusps of M1. Molars have a typical Dendromurinae dental pattern, but cusps are more individualized than in Steatomys. The M1 is the longest of the toothrow, being twice as large as the M2. It lacks the cusp t1 but has a rather large t4 connected to the t5. There is a groove separating the central (t2, t5 and t8) and labial (t3, t6, t9) cusps (not marked in Steatomys). The M2 lacks the t1 and has a small to tiny t3; the t4 is well-developed and connected to the t5. Molar M3 is very small and corresponds to Group 1 in Fig 12.
Lower jaw.
Lower incisors are ungrooved (one groove in upper incisors). The prelobe of M1 consists of a single median cusp. It is connected to the first row of cusps, which is oblique and constituted by a lingual metaconid connected to the labial protoconid located further back. The second row of cusps is equally oblique, with a well-developed hypoconid and a very large posterior cingulum located in the postero-lingual angle of the tooth. The M2 has two oblique rows of cusps, with conspicuous anterolabial and posterior cingula. The M3 is very small. As in other Dendromurinae, muscle attachment on the mandible is situated rather far from the mental foramen, which can help to distinguish Dendromus from Mus.
Systematic notes and South African fossil record.
The classification and species delimitation of the genus Dendromus has changed considerably over the last decades. Four species of Dendromus are listed in Monadjem et al. [22] for South Africa:
- Dendromus melanotis Smith, 1834
- Dendromus mesomelas (Brain, 1827)
- Dendromus mystacalis Heuglin, 1863
- Dendromus nyikae Wroughton, 1909
Recent molecular systematic assessment published by Voelker et al. [72] led to the resurrection of the genus Poemys. According to this, D. nyikae and D. melanotis would rather be identified as P. nyikae and P. melanotis, with the latter species probably comprising at least four species (P. arenarius, P. basuticus, P. melanotis, P. vulturnus). The two species D. mesomelas and D. mystacalis remain attributed to Dendromus and include many lineages and cryptic species not yet described. Consequently we prefer to keep the traditional taxonomy pending further revisions of South African Dendromus species. Additional fossil species have been described in South Africa:
- †Dendromus antiquus Broom, 1946 from the Late Pliocene of Taung, considered by Avery [4] as nomen nudum
- †Dendromus averyi Denys, 1994 from the Early Pliocene of Langebaanweg [73]
- †Dendromus darti Denys, 1994 from the Early Pliocene of Langebaanweg [73]
Material assigned to this genus has been recovered from many Quaternary fossil deposits in South Africa [50].
Genus Malacothrix Wagner, 1843 (Long-eared Mouse)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-1:2-2-1 (Figs 16 and 17).
Upper jaw.
Upper incisors are opisthodont and have one groove (as other members of Dendromurinae). Anterior palatal foramina extend beyond the first row of cusps of M1. Molars exhibit the typical Dendromurinae pattern, but the M1 is very long, occupying half of the length of the toothrow. The M1 has a well-marked antero-median cusp. The M3 is very small and corresponds to Group 1 in Fig 12. The three alveoli of the M1 are almost in line, with the lingual alveolus less displaced lingually than in Dendromus and Steatomys. There is a ridge-shaped masseter knob on the lower anterior of the zygomatic plate (sub-circular in Dendromus and Steatomys).
Lower jaw.
Lower incisors are narrow and ungrooved. The molars display round, well separated bunodont cusps, with the typical Dendromurinae cusp pattern. The M1 is narrow and very long, occupying more than half of the length of the molar toothrow; it displays seven highly alternated cusps separated by a longitudinal crest. The M2 has an antero-external cusp. The M3 is tiny and has a single cusp. The ventral edge of the mandible body between the ramus and the angular process is markedly curved. The posterior curvature between angular and coronoid process is well pronounced. As in other Dendromurinae, muscle attachment on the mandible is situated rather far from the mental foramen.
Systematic notes and South African fossil record.
This genus is monotypic:
- Malacothrix typica (A. Smith, 1834)
An additional fossil species has been described in South Africa:
- †Malacothrix makapani De Graaff, 1961
The oldest remains of Malacothrix are from Makapansgat around 3.3 MYA [48]. It has been recorded in several Quaternary fossil deposits in South Africa [50].
Genus Steatomys Peters 1846 (Fat Mice)
1) Upper (a) and lower (b) right toothrow of S. pratensis (RMCA-96-037-M-5108); 2) Upper (a) and lower (b) right toothrow of Steatomys sp. (DNMNH-3458).
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-1:2-2-1 (Figs 16 and 17).
Upper jaw.
Upper incisors are opisthodont and have one pronounced groove, located closer to the outer margin than the inner. Anterior palatal foramina are long and extend beyond the first row of cusps of M1. Molars have a typical Dendromurinae cusp pattern (with t1 and t7 missing in M1 and M2) but cusps are rounder than in Dendromus and Malacothrix. The M1 is the longest of the toothrow and has three lobes: the first lobe has two cusps that are connected and project posteriorly; the second lobe has three cusps with a t4 closely connected to the t5 but located slightly behind; the third lobe has two cusps and a posterior cingulum, which is connected to the t8 and delimits a fovea. The M2 has two lobes and a small t3. The M3 is small (although generally not as reduced as in Dendromus) and corresponds to Group 1 in Fig 12.
Lower jaw.
Lower incisors are ungrooved. Like Dendromus, the M1 has a median prelobe connected to the first oblique row of cusps and a median posterior cingulum below the second oblique row. The M2 has two lobes with an anterolabial cingulum and a small to medium posterior cingulum. The M3 is small and consists of a single cusp. As in other Dendromurinae, muscle attachment on the mandible is situated rather far from the mental foramen.
Systematic notes and South African fossil record.
Two species are recognised in South Africa:
- Steatomys krebsii Peters, 1852
- Steatomys pratensis Peters, 1846
In South Africa, the oldest known fossils are from Makapansgat around 3.3 MYA [48]. Remains of this genus have been recorded in several Quaternary fossil deposits [50].
Subfamily MYSTROMYINAE Vorontsov, 1966
Genus Mystromys Wagner, 1841 (African White-tailed Rat)
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-2:2-2-2 (Figs 16 and 17).
Upper jaw.
Upper incisors are orthodont and display typical striations on the anterior enamel surface that form a slightly raised band. The palatal foramina are long and extend past the M1. The cheekteeth have well separated opposed cusps that are connected by a longitudinal crest, resulting in a characteristic zigzag pattern. M1 has three lobes with two lingual and two labial folds; the first lobe has one to two fused cusps, the second and third lobe have two alternated cusps. M2 has two lobes each with two cusps. M3 is small and has two lobes, the last one consisting of a single cusp.
Lower jaw.
Lower incisors display characteristic raised enamel bands. Cusps of lower molars have alternate cusps connected by a longitudinal crest, which results in a distinctive zigzag pattern. Molar M1 has three lingual and two labial folds. Molar M2 has two lingual and two labial folds. Molar M3 has three cusps with a small antero-external cingulum.
Systematic notes and South African fossil record.
This genus is monotypic today:
- Mystromys albicaudatus (A. Smith, 1834)
Additional species have been described in South African fossil deposits:
- †Mystromys hausleitneri Broom, 1937 from various Pleistocene sites, which is often regarded as a chronospecies that closely resembles the present-day form, albeit slightly smaller [4, 47, 74].
- †Mystromys pocockei Denys, 1991 from the Early Pliocene site of Langebaanweg
Fossils of this genus are known from many Pleistocene sites in South Africa [50]. In the karst deposits of the Sterkfontein Valley from the Gauteng Province, it is often one the most abundant taxa from the micromammal assemblages.
† Genus Proodontomys Pocock, 1987
Upper (a) and lower (b) right toothrow of P. cookei (upper ESI, Cooper’s D fossil collection CD1980; lower ESI T.N. Pocock microfaunal fossil collection, EXQRM47). The molar M3 is missing.
Dental formula is 1-0-0-3:1-0-0-3. Alveolar formula is 3-3-1:2-2-1/2 (Figs 16 and 17).
Upper jaw.
Upper incisors are highly proodont and ungrooved; they do not display the typical ridge seen in Mystromys. Anterior palatal foramina are long and penetrate between the second rows of cusps in M1. Molars show a tendency to hypsodonty, with weak cusps that quickly wear down to a plain occlusal surface. As in Mystromys, the cheekteeth lack lateral cusps, but they differ by having opposite, rather than alternate, molar cusps, resulting in a semi-lophodont dentition. The M1 has three lobes, the M2 has two lobes, and the M3 has two poorly differentiated lobes that tend to fuse with wear to form a simple cylindrical tooth.
Lower jaw.
Lower incisors are ungrooved and do not display the typical ridge present in Mystromys. As in upper molars, molars have week cusps that flatten in early wear into a plain, semi-lophodont occlusal surface. The molar M1 shows a trilophodont pattern, and both the M2 and M3 are bilophodont. The coronoid process of the mandible is very large.
Systematic notes and South African fossil record.
This genus got extinct around 1 MYA, and a single species has been described in South Africa fossil deposits:
- †Proodontomys cookei Pocock, 1987
The oldest remains of Proodontomys come from the locality of Limeworks Makapansgat around 3.3 MYA [48].
Subfamily PETROMYSCINAE Roberts, 1951
Genus Petromyscus Thomas, 1926 (Pygmy Rock Mice)
Dental formula is 1-0-0-3-:1-0-0-3. Alveolar formula is: 3-3-1:2-2-1 (Figs 16 and 17).
Upper jaw.
Upper incisors are opisthodont and ungrooved. Cheekteeth are relatively small. Anterior palatal foramina reach the margin or the first lobe of the M1. Molars display round cusps connected by a median longitudinal crest, resulting in a zigzag pattern reminiscent of Mystromys. The M1 has three lobes: the first has one or two fused cusps, the second has three cusps including a t4 attached to the t5 (which distinguishes this genus from the members of Mystromyinae), the third has two cusps. The M2 has two lobes as well as an antero-external cusp (t3) but no t4. The M3 is small, having a tiny t3 (t1 is absent) and two rows of cusps connected lingually. It corresponds to Group 1 in Fig 12.
Lower jaw.
Lower incisors are ungrooved. The molar M1 has three lobes, which are connected in their median part by a longitudinal crest. Molar M3 is small and display a tiny antero-external cingulum.
Systematic notes and South African fossil record.
Three species are presently recognized in South Africa:
- Petromyscus barbouri Shortridge & Carter, 1938
- Petromyscus collinus (Thomas & Hinton, 1925)
- Petromyscus monticularis (Thomas & Hinton, 1925)
Remains of Petromyscus are scanty in the Quaternary fossil record, and the oldest fossils in South Africa are known from Holocene sites [50].
Bathyergidae
Mole-rats of the family Bathyergidae (Tables 32 and 33) are fossorial (adapted to digging or burrowing) rodents and exhibit numerous morphological adaptations for an underground life. Some four genera occur in South Africa. They live in tunnel systems and are almost never seen above the ground’s surface [21]. They mostly become vulnerable during the mound formation, when predators are able to locate them with accuracy. Only juveniles of Bathyergus and Georychus are small enough to be captured by owls such as Tyto alba (Barn owl), and Bubo africanus (Spotted eagle-owl), which are the main accumulators of Quaternary fossil micromammal deposits in South Africa. As a result, it is common to find mainly young individuals among fossil assemblages accumulated by these predators, with specimens showing only two or three completely erupted cheekteeth versus the four cheekteeth found in adult specimens. Members of Bathyergidae display distinctly proodont upper incisors. The lower jaw is hystricomorph in character, showing a characteristic flared and inflected angular process on the mandible. In South African mole-rats, the dental formula is always 1-0-1-3:1-0-1-3 and the anterior palatal foramina are small, ending well before the M1.
Family BATHYERGIDAE Waterhouse, 1841
Genus Bathyergus Illiger, 1811 (Dune Mole-rats)
1) Upper (a) and lower (b) right toothrow of an adult B. janetta (upper DNMNH-543; lower DNMNH-39304); 2) Upper (a) and lower (b) right toothrow of a juvenile B. suillus (DNMNH-2162), with molars M3/M3 not erupted yet.
Dental formula is 1-0-1-3:1-0-1-3. Alveolar formula varies with age of specimen.
Upper jaw.
Upper Incisors are proodont, robust and heavily grooved; they are less protruding than in other genera and are rooted above the anterior cheekteeth. Anterior palatal foramina are short, ending before molar M1. Cheeekteeth are hypsodont and simple (rounded and uncusped), but young animals display re-entrant folds.
Lower jaw.
Lower incisors are plain. Lower cheekteeth are similar to upper cheekteeth. The mandible is hystricognathous and stickily built, with the angular portion extending well posteriorly. The mental foramen is positioned posteriorly to the diastema, aligned with the posterior border of the M1.
Systematic notes and South African fossil record.
Two species of Bathyergus occur in South Africa:
- Bathyergus janetta Thomas & Schwann, 1904
- Bathyergus suillus (Schreber, 1782)
An additional fossil species has been described:
- †Bathyergus hendeyi Denys, 1998 from the Early Pliocene of Langebaanweg
Genus Cryptomys Gray, 1864 (Mole-rats)
The molars M3 are missing.
1) Upper (a) and lower (b) right toothrow of an adult C. hottentotus (DNMNH-13033); 2) Upper (a) and lower (b) right toothrow of a juvenile C. hottentotus (DNMNH-3418), with molars M3/M3 not erupted yet.
Dental formula is 1-0-1-3:1-0-1-3. Alveolar formula varies with age of specimen.
Upper jaw.
Cryptomys and Fukomys appear to be indistinguishable on morphological grounds. Upper incisors are robust, ungrooved and proodont. Anterior palatal foramina are very short, ending before molar M1. Cheekteeth are simple, rounded and uncusped, but young animal display re-entrant folds. The skull is the smallest found in the Bathyergidae.
Lower jaw.
Lower incisors are smooth on their anterior surfaces. Their width is less than 2 mm, even for older specimens (in Bathyergus, the width of the incisor is superior to 2 mm before the M3 is erupted). Lower cheekteeth are similar to upper cheekteeth: simple, rounded and uncusped, although young animals display re-entrant folds. The coronoid process of the hystricognathous mandible is high and projects backwards. The mental foramen is positioned posteriorly to the diastema, aligned with the posterior border of the M1.
Systematic notes and South African fossil record.
The species delimitation of Cryptomys has changed multiple times over last decades [75]. This genus formerly encompassed all the species currently placed within Fukomys. A single species is currently recognised in South Africa:
- Cryptomys hottentotus (Lesson, 1826)
Additional fossil species have been described in South Africa:
- †Cryptomys broomi Denys, 1998 from Langebaanweg
- †Cryptomys robertsi Broom, 1937
Fossils of Cryptomys are known from the Early Pliocene and have been identified in many Quaternary deposits. However, most of this material was identified before the recognition of Fukomys as a distinct genus. It is therefore likely that some of these remains may rather be attributed to this latter genus.
Genus Fukomys Kock et al., 2006 (Mole-rats)
Dental formula is 1-0-1-3:1-0-1-3. Alveolar formula varies with age of specimen.
Upper jaw.
Cryptomys and Fukomys appear to be indistinguishable on morphological grounds. See Cryptomys for anatomical description.
Systematic notes and South African fossil record.
Species of this genus were previously included in Cryptomys until molecular analyses revealed the existence of two well separated clades within this genus [75]. A single species is currently recognized in South Africa:
- Fukomys damarensis (Ogilby, 1838)
No fossil remains of Fukomys have been identified in South Africa, probably due to the fact that F. damarensis has only recently been erected as a separate species from C. hottentotus.
Genus Georychus Illiger, 1811 (Cape Mole-rats)
1) Upper (a) and lower (b) right toothrow of an adult G. capensis (DNNH-9145); 2) Upper (a) and lower (b) right toothrow of a juvenile G. capensis (DNMNH-4159), with molars M3/M3 not erupted yet.
Dental formula is 1-0-1-3:1-0-1-3. Alveolar formula varies with specimen age.
Upper jaw.
Incisors are robust, ungrooved and proodont. Anterior palatal foramina are very short, ending before molar M1. Molars have single inner and outer folds of enamel jutting into the dentine. Folds on M2 persist with age, on M3 also to a lesser extent, and folds on P4 and M1 tend to disappear when teeth are worn. The skull is of intermediate size, being larger than Cryptomys/Fukomys but smaller than Bathyergus; it is robust and has a narrow rostrum.
Lower jaw.
Lower incisors are smooth. Their width is more than 2.3 mm, even for young specimens. Lower molars have one outer fold that persists with age and one inner fold that tends to disappear with wear. The mental foramen is positioned posteriorly to the diastema, aligned with the posterior border of the M1.
Pedetidae
The family Pedetidae contains only the genus Pedetes. It is a very large rodent, surpassed in weight only by Hystrix and large specimens of Thryonomys. Due to its size and high weight, eagle owls (Bubo africanus, Bubo capensis and Bubo lacteus) are the only nocturnal raptors capable of predating adult individuals [21, 76]. The nomenclature of the teeth of fossil and modern representatives of Pedetidae was detailed in Pickford & Mein [77].
Genus Pedetes Fitzinger, 1867 (Springhares)
Dental formula is 1-0-1-3:1-0-1-3.
Upper jaw.
The skull is large and robust. Upper incisors are opisthodont, thick and ungrooved. The anterior palatal foramina are short. The cheekteeth are all roughly the same size. They are hypsodont and open-rooted. The cheekteeth are comprised of two lophs, which are separated by a median transverse valley and uniting in the lingual side.
Lower jaw.
The mandible is sciurognath. The condylar process reaches beyond the level of the angular process, which is poorly defined. The coronoid process consists of a thin ridge. Lower incisors are thick and ungrooved. Lower cheekteeth are bilophed like in upper tooth row but are uniting in the labial side.
Systematic notes and South African fossil record.
A single species of Pedetes occurs in South Africa:
- Pedetes capensis (Forster, 1778)
Two additional fossil species have been listed:
- †Pedetes gracilis Broom, 1934 from the Pliocene locality of Taung
- †Pedetes hagenstadti Dreyer and Lyle, 1931 from the Middle Pleistocene locality of Florisbad
Hystricidae
In South Africa, only the genus Hystrix occurs. It contains one modern species, H. africaeaustralis, and one fossil species, H. makapanensis. H. africaeaustralis is the largest African rodent and has long spines (quills) covering its back and flanks. Large carnivores are the most common predator. The nomenclature of the teeth of fossil and modern representatives of African Hystrix is detailed in Azzarà et al. [78].
Family HYSTRICIDAE G. Fischer, 1817
Genus Hystrix Linnaeus, 1758 (Crested Porcupines)
Dental formula is 1-0-1-3:1-0-1-3.
Upper jaw.
Upper incisors are large, proodont and smooth. The anterior palatal foramina are short and located far forwards from the cheekteeth. Molars are large and tend towards hypsodonty. They display cusps related by lophs when unworn, but the crowns rapidly reduce in height with age, and wear of the tooth and cusps get individualized into small enamel islands. The teeth all have only one fold (also called flexus) on the lingual side and up to three folds on the labial side (this arrangement is revers in the lower teeth).
Lower jaw.
Lower incisors are large and ungrooved. Lower teeth display the same characteristic occlusal pattern with crests of enamel, but there is one labial fold and up to three lingual folds. The mandible is hystricognath.
Systematic notes and South African fossil record.
The following species occurs in South Africa:
- Hystrix africaeaustralis Peters, 1852
An additional species is listed in the South African fossil record:
- †Hystrix makapanensis Greenwood, 1958 found in the Pliocene Limeworks of Makapansgat and several Early Pleistocene deposits from the Sterkfontontein Valley in Gauteng Province [78].
Remains of Hystrix are found in many fossil deposits through the Quaternary.
Thryonomyidae
The family Thryonomyidae includes the single genus, Thryonomys. It is the third largest rodent genus in South Africa, after Hystrix and Pedetes. Like the latter, eagle owls are the only nocturnal raptors susceptible of predating adults [21] but mesocarnivores can prey upon this genus.
Family THRYONOMYIDAE Pocock, 1922
Genus Thryonomys Fitzinger, 1867 (Cane Rats)
Dental formula is 1-0-1-3:1-0-1-3.
Upper jaw.
Upper incisors are broad, opisthodont, and have three grooves on the inner side. The skull is stickily built with a short rostrum. The anterior palatal foramina reach the M1. The cheekteeth have three to four characteristic enamel infoldings, which are single in the lingual side and double in the labial side. The premolar P4 has four lophs, while the molars M1-M3 have only three. The M2 is often the largest tooth in the row, and the M3 erupts late in the life on the individual.
Lower jaw.
Lower incisors are ungrooved. The cheekteeth have two outer and one inner enamel infoldings (it is reversed in the maxilla). As for the lower teeth, there are four lophs in P4 and three in the molars M1-M3. The mandible is robust, and the well-developed angular process projects far backwards, and is hystricognathous.
Petromuridae
The family Petromuridae contains only a single genus, Petromus. These medium-sized rodents are squirrel-like in appearance, and their present-day distribution is restricted to the west coast of Southern Africa. They are associated with rocky habitats, occupying rock crevices in boulders, canyons and mountain slopes [21]. Little is known about its predators, but remains were identified within pellet material produced by Bubo africanus [81].
Family PETROMURIDAE Tullberg, 1899
Genus Petromus A. Smith, 1831 (Nokis or Dassie Rats)
Dental formula is 1-0-1-3:1-0-1-3.
Upper jaw.
The skull is dorsoventrally flat with inflated bulla. Upper incisors are plain and opisthodont. The palate is narrow, about equal to the width of the P4 in its anterior part. The anterior palatal foramina are long and reach the P4. The four cheekteeth have roughly the same size. They are hypsodont with deep lingual enamel infoldings.
Lower jaw.
Lower incisors are ungrooved. There are four cheekteeth, which are hypsodont with deep buccal enamel infoldings. The mandible is very long and vertically compressed with a sharp angular process, showing hystricognathy.
Systematic notes and South African fossil record.
The genus is monotypic:
- Petromus typicus A. Smith, 1831
Two additional fossil species have been described in the South African fossil record:
- †Petromus antiquus Sénégas, 2004 from the Early Pliocene site of Waypoint 160
- †Petromus minor Broom, 1939 from the Late Pliocene site of Taung
Gliridae
Only one genus of Gliridae occur in Southern Africa, the African dormouse Graphiurus. Species of Graphiurus are small-sized, squirrel-like rodents with good climbing abilities. They are predominantly arboreal, or are associated with boulders and rocky outcrops [22]. They are preyed upon by various predators, including owls.
Family GLIRIDAE Thomas, 1897
Subfamily GRAPHIURINAE Winge, 1887
Genus Graphiurus Smuts, 1832 (Dormice)
1) Upper (a) and lower (b) right toothrow of G. platyops (DNMNH-4360); 2) Upper (a) and lower (b) right toothrow of G. ocularis (DNMNH-27470).
Dental formula is 1-0-1-3:1-0-1-3.
Upper jaw.
Incisors are ungrooved, and orthodont to slightly opisthodont. Palatal foramina are small and reach far before the toothrow. The palate is wide, about equal to the length of the upper toothrow. The premolar P4 is roughly the same size or a bit smaller than M3, except in G. ocularis where P4 is much reduced. The occlusal surface of the teeth displays faint transverse ridges.
Lower jaw.
Incisors are ungrooved. Lower cheekteeth also display faint transverse ridges. The mandible is elongated with a long coronoid process and a tilted angular process. The mental foramen is well marked.
Systematic notes and South African fossil record.
Five species of Graphiurus are currently described in South Africa:
- Graphiurus microtis (Noack, 1887)
- Graphiurus murinus (Desmerest, 1822)
- Graphiurus ocularis (Smith, 1829)
- Graphiurus platyops Thomas, 1897
- Graphiurus rupicola (Thomas & Hinton, 1925)
The oldest remains of Graphiurus in South Africa are found in several Early Pleistocene deposits from the Sterkfontein Valley, in Gauteng Province [5, 47, 48, 74, 82].
Sciuridae
Some two genera of Sciuridae (Tables 43 and 44) occur in RSA: Paraxerus and Geosciurus (which was included until recently in the genus Xerus). Species of the genus Paraxerus are small to medium-sized tree squirrels, which have predominantly diurnal and arboreal habits. They typically nest in tree holes and occupy forests and woodland areas. They are preyed upon by owls [24]. Geosciurus is a larger ground-dwelling squirrel, being terrestrial and resting in burrows, piles of rocks and termite mounds [21]. Members of Geosciurus are too big to be hunted by most nocturnal raptors, but Spotted eagle owl (Bubo africanus) and giant eagle owl (Bubo lacteus) have been reported preying upon them [83, 84]. The nomenclature of the teeth of fossil and modern representatives of African sciurids is provided in Denys et al. [85], along with a list of distinctive dental and skulls characters presented below (see Fig 119).
Family SCIURIDAE Fischer de Waldheim, 1817
Genus Geosciurus Smith, 1834 (Ground squirrels)
1) Upper (a) and lower (b) right toothrow of G. inauris (DNMNH-15143); 2) Upper (a) and lower (b) right toothrow of G. princeps (DNMNH-6327).
Dental formula is 1-0-1-3-:1-0-1-3.
Upper jaw.
Upper incisors are ungrooved and opisthodont. The palatine bone extends well behind the molars. There is a well-developed masseter knob. A single premolar (P4) is present in the upper jaw (there are two premolars in Paraxerus). Molars have three transverse lophs (anteroloph, metaloph, posteroloph) relating bunodont cusps on the occlusal surface. They display a big protocone on the whole lingual part of the molar, while the hypocone is hardly visible. In M1 and M2, the metaloph is short and is connected to the posteroloph. The presence of a mesostyle varies among specimens.
Lower jaw.
Lower incisors are ungrooved. Molars have well individualized bunodont cusps (a prominent hypoconid and protoconid, a metaconid and a crestiform entoconid) connected by transverse ridges (protolophid, entolophid, posterolophid) that develop with wear. The molars have a prominent hypoconid In M1 and M2, the cusps protoconid and metaconid are fused by their distal side (anterior side in Paraxerus). The mandible is sciurognath, with its angular part stockily built.
Systematic notes and South African fossil record.
This genus was previously included within Xerus, until recent phylogenetic reconstructions led to the recognition of the genus Geosciurus for specimens from Southern Africa [88]. Two species are currently recognised:
- Geosciurus inauris (Zimmermann, 1780)
- Geosciurus princeps Thomas, 1929
Remains of this genus are scanty in the Quaternary fossil record. The oldest record of Geosciurus is from the Middle Pleistocene locality of Florisbad [89, 90].
Genus Paraxerus Forsyth Major, 1893 (Bush Squirrels)
1) Upper (a) and lower (b) right toothrow of P. cepapi (DNMNH-4360); 2) Upper (a) and lower (b) right toothrow of P. palliatus (DNMNH-6214).
Dental formula is 1-0-2-3:1-0-1-3.
Upper jaw.
Upper incisors are orthodont and ungrooved. The palatine bone extends only the posterior edge of the M3. The anterior palatal foramina are short and located far forward from the teeth. There are five upper cheekteeth, with the presence of a minute P3. Molars are of approximately the same size and display transverse lophs (anteroloph, protoloph, metaloph, posteroloph) relating bunodont cusps on the occlusal surface. M1 and M2 both have three transverse valleys, M3 has two. In M1-M2, the metaloph is long and connected to the protocoone, and the mesostyle is absent. In the M3 the hypocone and protocone are connected.
Lower jaw.
Lower incisors are ungrooved. Molars are ridged more or less transversely and display four main cusps which are not always visible with wear: a big protoconid fused to a metaconid by the anterior side (posterior in Geosciurus), a hypoconid and a transverse entoconid. The M3 is the largest tooth of the molar row. The mandible is sciurognath; it has a high coronoid process, well segregated from the condylar process.
Systematic notes and South African fossil record.
Two species are currently recognized in South Africa:
- Paraxerus cepapi (A. Smith, 1836)
- Paraxerus palliates (Peters, 1852)
Fossils of this genus are rare in the Quaternary fossil record, with a single Pleistocene record from Border Cave [91] and a few Holocene records [50].
6. Discussion and conclusion
Recent advances in rodent taxonomy at the subfamily, genus, and species level, together with the need for more detailed description of their dental anatomy, has encouraged us to provide an updated version of identification keys previously published by Coetzee [23] and De Graaff [24], based on standard, and new, morphological characters. This contribution has endeavored to assemble the latest information regarding rodent systematics, taxonomy, and palaeontology. We aim to support and facilitate the identification of rodent specimens from Quaternary and modern coprocenoses by providing a simple identification tool based on a traditional dichotomous system and scaled photographs of the skull and dentition.
We have attempted to compile the most reliable characters based on collection specimens identified on cytogenetic or molecular grounds mainly. Some published questionable characters were discarded (for instance, the presence of an accessory anterior median cusp on M1 was previously used to distinguish Thallomys from Grammomys) and characters likely to display variability have received less emphasis in this work than in previous identification keys (such as the presence of accessory cusplets or posterior cingulum). Instead, we have retained mainly those features that we were able to observe consistently in the collection specimens ourselves.
Regrettably, however, isolated teeth and fragmented specimens from owl pellet or fossil material may remain unidentifiable, even at the genus level. Most of the criteria used in our key involve having the complete dental row preserved. In a similar way, this key does not include genera found in the savannahs and deserts of Botswana, Zambia, Namibia and Zimbabwe that were possibly found further south or west during the Pleistocene period. This remark concerns, for instance, the genus Pelomys, which was identified by some authors among the fossil remains from Makapansgat, Border Cave, and Sterkfontein [47, 70, 92].
Accurate taxonomic identification is essential for a wide range of scientific applications, from ecology, conservation biology and pest management, to archaeozoology and palaeoenvironmental reconstruction. The species level generally constitutes the fundamental unit of investigation. For well-preserved modern specimens, one may have little difficulty in identifying features characteristic of a species using criteria available in literature coupled with biogeographic data. Works of this kind are to be found in general monographs or more specialised journal articles, and we have elaborated on the factors that led us to restrict our key to genus level. Problems arise with broken specimens from modern and palaeo coprocenoses, which lack diagnostic criteria. Moreover, several species cannot be distinguished on the basis of craniodental anatomy unless time-intensive techniques such as geometric morphometrics are employed, and this may not always be feasible. Finally, when it comes to fossils, the use of distribution data is extremely risky, and may explain the disagreement that exists between some researchers on the status of different fossil species, especially those described during the last century (this is the case, for example, of the large number of species described by Broom in the 1930s and 1940s).
Ironically, the huge progress made in systematics and species identification through the use of numerical taxonomy and molecular taxonomy has eroded our confidence in identifying species based on bone and teeth remains. However, the discovery of new fossil rodents in South Africa is continuing at a rapid pace, and taxonomic identification of craniodental material remains the foundation of palaeontological and archaeological research. Identification skills rely mainly on experience and repeated observations of key features. In this respect, this key should be a valuable resource for both professionals and amateurs alike. It is particularly beneficial regarding the numerous genera which have similar morphological features, and which are challenging to differentiate. Nevertheless, we must emphasize that the identification key is intended to complement the use of official natural history collections in order to facilitate accurate identification.
Supporting information
S1 Fig. Comparative osteological plates.
Upper jaws.
https://doi.org/10.1371/journal.pone.0289812.s001
(PDF)
S2 Fig. Comparative osteological plates.
Lower jaws.
https://doi.org/10.1371/journal.pone.0289812.s002
(PDF)
S1 Checklist. References of specimens used for osteological plates.
https://doi.org/10.1371/journal.pone.0289812.s003
(DOCX)
S1 Table. Material examined for measurements.
https://doi.org/10.1371/journal.pone.0289812.s004
(XLSX)
Acknowledgments
This study could not have been attempted without the help of collection curators and researchers who helped us and permitted us access to their collections. We would like to thank here Josef Bryja and Ondřej Mikula from the Institute of Vertebrate Biology (Czech Republic), Violaine Nicolas from the Muséum national d’Histoire naturelle (France), Teresa Kearney from the Ditsong National Museum of Natural History (Republic of South Africa) and Emmanuel Gilissen from the Royal Museum for Central Africa (Belgium).
References
- 1. De Graaff G. A preliminary investigation of the mammalian microfauna in Pleistocene deposits of caves in the Transvaal system. Palaeontologia Africana. 1961; 7: 79–118.
- 2. Thackeray JF. Late Quaternary environmental changes inferred from small mammalian fauna, southern Africa. Climatic Change. juill 1987;10(3): 285–305.
- 3. Denys C. Les analyses multivariées: une aide à l’interprétation des paléoenvironnements. L’exemple des rongeurs Plio-Pléistocènes d’Afrique Australe. Geobios. 1992;25: 209–17.
- 4. Avery DM. An assessment of the Lower Pleistocene micromammalian fauna from Swartkrans members 1–3, Gauteng, South Africa. Geobios. 1998; 31(3): 393–414.
- 5. Avery DM. The Plio-Pleistocene vegetation and climate of Sterkfontein and Swartkrans, South Africa, based on micromammals. Journal of Human Evolution. 2001; 41(2): 113–32. pmid:11437522
- 6. Avery DM. Pleistocene micromammals from Wonderwerk Cave, South Africa: practical issues. Journal of Archaeological Science. 2007; 34: 613–25.
- 7. Matthews T, Denys C, Parkington JE. The palaeoecology of the micromammals from the late middle Pleistocene site of Hoedjiespunt 1 (Cape Province, South Africa). Journal of Human Evolution. 2005; 49(4): 432–51. pmid:16045969
- 8. Matthews T, Rector A, Jacobs Z, Herries AIR, Marean CW. Environmental implications of micromammals accumulated close to the MIS 6 to MIS 5 transition at Pinnacle Point Cave 9 (Mossel Bay, Western Cape Province, South Africa). Palaeogeography, Palaeoclimatology, Palaeoecology. 2011; 302(3–4): 213–29.
- 9. Matthews T, Marean CW, Cleghorn N. Past and present distributions and community evolution of Muridae and Soricidae from MIS 9 to MIS 1 on the edge of the Palaeo-Agulhas Plain (south coast, South Africa). Quaternary Science Reviews. 2020; 235.
- 10. Nel TH, Henshilwood CS. The Small Mammal Sequence from the c. 76–72 ka Still Bay Levels at Blombos Cave, South Africa–Taphonomic and Palaeoecological Implications for Human Behaviour. PLoS ONE. 2016; 11(8).
- 11.
Lacher T, Murphy W, Rogan J, Smith A, Upham N. Evolution, phylogeny, ecology and conservation of the Clade Glires: Lagomorpha and Rodentia. In: Handbook of the Mammals of the World: Volume 6 Lagomorphs and Rodents I. Lynx Ediciones, Barcelona. 2016. p. 15–26.
- 12.
Wilson DE, Lacher TE, Mittermeier RA. Handbook of the Mammals of the World–Vol. 6, Lagomorphs and Rodents I. Lynx Edicions. 2016. 987 p.
- 13.
Wilson DE, Lacher TE, Mittermeier RA. Handbook of the Mammals of the World–Vol. 7, Rodents II. Lynx Edicions. 2017. 1008 p.
- 14. Schenk JJ, Rowe KC, Steppan SJ. Ecological Opportunity and Incumbency in the Diversification of Repeated Continental Colonizations by Muroid Rodents. Systematic Biology. 2013; 62(6): 837–64. pmid:23925508
- 15. Fabre PH, Hautier L, Dimitrov D, Douzery EJ P. A glimpse on the pattern of rodent diversification: a phylogenetic approach. BMC Evolutionary Biology. 2012; 12(1): 88. pmid:22697210
- 16.
Andrews P. Owls, Caves and Fossils: Predation, Preservation and Accumulation of Small Mammal Bones in Caves, with an Analysis of the Pleistocene Cave Faunas From Westbury-Sub-Mendip, Somerset, U.K. University of Chicago Press; 1990. 252 p.
- 17. Avery DM. Micromammals as palaeoenvironmental indicators of the southern African Quaternary. Transactions of the Royal Society of South Africa. 2007; 62(1): 17–23.
- 18. Matthews T. Taphonomic Characteristics of Micromammals Predated by Small Mammalian Carnivores in South Africa: Application to Fossil Accumulations. Journal of taphonomy. 2006; 4(3): 143–61.
- 19. Matthews T, Marean CW, Cleghorn N. Past and present distributions and community evolution of Muridae and Soricidae from MIS 9 to MIS 1 on the edge of the Palaeo-Agulhas Plain (south coast, South Africa). Quaternary Science Reviews. 2020; 235: 105774.
- 20.
Skinner JD, Chimimba CT. The Mammals of the Southern African Subregion (3rd edn). Cambridge, UK: Cambridge University Press; 2005.
- 21.
Happold DCD. Mammals of Africa. Vol. III: Rodents, Hares and Rabbits. Bloomsbury Publishing. London; 2013.
- 22. Monadjem A, Taylor PJ, Denys C, Cotterill FPD. Rodents of Sub-Saharan Africa. Rodents of Sub-Saharan Africa. De Gruyter; 2015.
- 23. Coetzee CG. The identification of southern African small mammal remains in owl pellets. Cimbebasia. 1972; A(2): 53–64.
- 24.
De Graaff G. The Rodents of Southern Africa: Notes on their identification, distribution, ecology and taxonomy. Butterworths. Durban; 1981.
- 25. Davis DHS. Classification problems of African Muridae. Zoologica Africana. 1965; 1(1): 121–45.
- 26. Misonne X. African and Indo-Australian Muridae: evolutionary trends. Annales du Musée royal de l’Afrique centrale. 1969; Série 4(172): 1–219.
- 27. Denys C, Michaux J, Petter F, Aguilar JP, Jaeger JJ. Molar morphology as a clue to the phylogenetic relationship of Acomys to the Murinae. Israel Journal of Zoology. 1992; 38: 253–262.
- 28. Alhajeri B, Steppan S. Disparity and Evolutionary Rate Do Not Explain Diversity Patterns in Muroid Rodents (Rodentia: Muroidea). Evolutionary Biology. 2018; 45
- 29. D’Elía G, Fabre PH, Lessa E. Rodent systematics in an age of discovery: Recent advances and prospects. Journal of Mammalogy. 2019; 100: 852–71.
- 30. Swanson M, Oliveros C, Esselstyn J. A phylogenomic rodent tree reveals the repeated evolution of masseter architectures. Proceedings Biological sciences. 2019; 286. pmid:31064307
- 31. Upham N, Esselstyn J, Jetz W. Inferring the mammal tree: Species-level sets of phylogenies for questions in ecology, evolution, and conservation. PLOS Biology. 2019; 17. pmid:31800571
- 32. Marsh C, Sica Y, Burgin C, Dorman W, Anderson R, Mijares I, et al. Expert range maps of global mammal distributions harmonised to three taxonomic authorities. Journal of Biogeography. 2022; 49. pmid:35506011
- 33. Ung V, Dubus G, Zaragüeta-Bagils R, Vignes-Lebbe R. Xper2: introducing e-taxonomy. Bioinformatics. 2010; 26(5): 703–4. pmid:20053842
- 34. Kerner A, Bouquin S, Portier R, Vignes Lebbe R. The 8 Years of Existence of Xper3: State of the art and future developments of the platform. Biodiversity Information Science and Standards. 2021; 5.
- 35. Henschel J, David JHM, Jarvis JUM. Age determination and age structure of a striped fieldmouse, Rhabdomys pumilio, population from the Cape Flats. South African Journal of Zoology. 1982; 17: 136–42.
- 36. Dieterlen F. Beitrage zur Biologie der Stachelmaus, Acomys cahirinus imidiatus Cretzschmar. Zeitschrift fur Säugetierkunde. 1961; 26: 1–13.
- 37. Willan K. Reproductive biology of the southern African ice rat. Acta Theriol. 1990;35: 39–51.
- 38. Waterman J. Reproductive Biology of a Tropical, Non-Hibernating Ground Squirrel. Journal of Mammalogy. 1996;77: 134.
- 39. Fernández-Jalvo Y, Andrews P, Denys C, Sesé C, Stoetzel E, Marin-Monfort D, et al. Taphonomy for taxonomists: Implications of predation in small mammal studies. Quaternary Science Reviews. 2016;139: 138–57.
- 40.
Fernández-Jalvo Y, Andrews P. Atlas of Taphonomic Identifications: 1001+ Images of Fossil and Recent Mammal Bone Modification. Springer. 2016.
- 41. Denys C, Michaux J. La troisième molaire supérieure chez les Muridae d’Afrique tropicale et le cas des genres Acomys, Uranomys et Lophuromys. Bonner Zoologische Beiträge. 1992;43(3).
- 42. Verheyen W, Hulselmans J, Wendelen W, Leirs H, Corti M, Backeljau T, et al. Contribution to the systematics and zoogeography of the East-African Acomys spinosissimus Peters 1852 species complex and the description of two new species (Rodentia: Muridae). Zootaxa. 2011; 3059(1): 1–35.
- 43. Aghová T, Palupčíková K, Šumbera R, Frynta D, Lavrenchenko LA, Meheretu Y, et al. Multiple radiations of spiny mice (Rodentia: Acomys) in dry open habitats of Afro-Arabia: evidence from a multi-locus phylogeny. BMC Evol Biol. 2019;19:69. pmid:30832573
- 44.
Musser G, Carleton MD. Family Muridae. In: Mammal Species of the World: A Taxonomic and Geographic Reference (2nd edn). John Hopkins University Press. Baltimore; 1993.
- 45. Pavlinov IJ, Dubrowsky AY, Potapova EG, Rossolimo OL. Gerbils of the World. J. Nauka Publications. Moscow; 1990.
- 46. Granjon L, Colangelo P, Tatard C, Colyn M, Dobigny G, Nicolas V. Intrageneric relationships within Gerbilliscus (Rodentia, Muridae, Gerbillinae), with characterization of an additional West African species. Zootaxa. 2012; 3325: 1–25.
- 47.
Denys C. Implications paléoécologiques et paléobiogéographiques de létude de rongeurs plio-plaistocènes d’Afrique orientale et Austrake [Thèse de 3e cycle]. Université Pierre et Marie Curie; 1990.
- 48. Pocock TN. Plio-Pleistocene fossil mammalian microfauna of Southern Africa—a preliminary report including description of two new fossil muroid genera (Mammalia: Rodentia). Palaeontologia Africana. 1987; 26(7): 70–91.
- 49. De Graaff G. A short survey of investigations of fossil rodents in African deposits. South African Journal of Science. 1961; 57: 191–6.
- 50.
Avery DM. A Fossil History of Southern African Land Mammals. Cambridge: Cambridge University Press; 2019.
- 51. Fernández-Jalvo Y, Avery DM. Pleistocene Micromammals and Their Predators at Wonderwerk Cave, South Africa. Afr Archaeol Rev. 2015; 32(4): 751–91.
- 52. Chimimba CT. Phylogenetic relationships in the genus Aethomys (Rodentia: Muridae). African Zoology. 2005; 40(2): 271–84.
- 53. Lecompte E, Aplin K, Denys C, Catzeflis F, Chades M, Chevret P. Phylogeny and biogeography of African Murinae based on mitochondrial and nuclear gene sequences, with a new tribal classification of the subfamily. BMC Evolutionary Biology. 2008; 8(1): 199. pmid:18616808
- 54. Dippenaar NJ, Swanepoel P, Gordon H. Diagnostic morphometrics of two medically important southern African rodents, Mastomys natalensis and M. coucha (Rodentia, Muridae). S Afr J Sci. 1993; 89: 300–3.
- 55. Steppan SJ, Schenk JJ. Muroid rodent phylogenetics: 900-species tree reveals increasing diversification rates. PLoS One. 2017; 12(8): e0183070. pmid:28813483
- 56. Meester JAJ, Rautenbach IL, Dippenaar NJ, Baker CM. Classification of Southern African Mammals. Transvaal Museum Monogr. 5. 1986. 359 p.
- 57. Wadley L, Turner G. Hope Hill Shelter: a later stone age site in the southern Transvaal. South African Journal of Science. 1987; 83(2): 98.
- 58.
Winkler AJ, Denys C, Avery DM. Fossil rodents of Africa. In: Werdelin L, Sanders W, éditeurs. Cenozoic Mammals of Africa. Berkeley: California University Press; 2010.
- 59. Taylor PJ, Denys C, Mukerjee M. Phylogeny of the African murid tribe Otomyini (Rodentia), based on morphological and allozyme evidence. Zoologica Scripta. 2004; 33: 389–402.
- 60. Bastos AD, Nair D, Taylor PJ, Brettschneider H, Kirsten F, Mostert E, et al. Genetic monitoring detects an overlooked cryptic species and reveals the diversity and distribution of three invasive Rattus congeners in south Africa. BMC Genetics. 2011; 12(1): 26. pmid:21324204
- 61. Aplin KP, Suzuki H, Chinen AA, Chesser RT, ten Have J, Donnellan SC, et al. Multiple Geographic Origins of Commensalism and Complex Dispersal History of Black Rats. PLOS ONE. 2011; 6(11): e26357. pmid:22073158
- 62. Avery DM. Micromammals and the Late Quaternary of southern Africa. South African Journal of Geology. 2021; 124(4): 1073–82.
- 63. Avery DM. The dispersal of brown rats Rattus norvegicus and new specimens from 19th century Cape Town. Mammalia. 1985; 49: 573–6.
- 64. Castiglia R, Solano E, Makundi R, Hulselmans J, Verheyen E, Colangelo P. Rapid chromosomal evolution in the mesic four-striped grass rat Rhabdomys dilectus (Rodentia, Muridae) revealed by mtDNA phylogeographic analysis. Journal of Zoological Systematics and Evolutionary Research. 2011.
- 65. Ganem G, Dufour CMS, Avenant NL, Caminade P, Eiseb SJ, Tougard C, et al. An update on the distribution and diversification of Rhabdomys sp. (Muridae, Rodentia). Journal of Vertebrate Biology. 2020; 69(2): 1.
- 66. Matthews T, Denys C, Parkington JE. Community evolution of Neogene micromammals from Langebaanweg ‘E’ Quarry and other west coast fossil sites, south-western Cape, South Africa. Palaeogeography, Palaeoclimatology, Palaeoecology. 2007; 245(3–4): 332–52.
- 67. Olayemi A, Nicolas V, Hulselmans J, Missoup A, Fichet-Calvet E, Amundala D, et al. Taxonomy of the African giant pouched rats (Nesomyidae: Cricetomys): molecular and craniometric evidence support an unexpected high species diversity. Zoological Journal of the Linnean Society. 2012; 165(3): 700.
- 68. Wadley L. The Howieson’s Poort Industry of Sibudu Cave. Goodwin Series. 2008; 10: 122–32.
- 69.
Welbourne RG. Appendix 1: an analysis of the animal remains recovered from Rose Cottage Cave. In: Kohary J, éditeur. Rose Cottage Cave. Johannesburg: Archaeological Research Unit, University of the Witwatersrand. 1988. p. 133–40.
- 70. Avery DM. The environment of early modern humans at Border Cave, South Africa: micromammalian evidence. Palaeogeography, Palaeoclirnatology, Palaeoecology. 1992; 91: 71–87.
- 71. Avery DM. A preliminary assessment of the micromammalian remains from Gladysvale Cave, South Africa. Palaeontologia Africana. 1 janv 1995; 32: 1–10.
- 72. Voelker G, Huntley JW, Bryja J, Denys C, Šumbera R, Demos TC, et al. Molecular systematics and biogeographic history of the African climbing-mouse complex (Dendromus). Molecular Phylogenetics and Evolution. 2021; 161: 107166. pmid:33798668
- 73. Denys C. Nouvelles espèces de Dendromus (Rongeurs, Muroidea) à Langebaanweg. Conséquences stratigraphiques et paléoécologiques. Paleovertebrata. 1994; 23(1–4): 153–76.
- 74.
Sénégas F. Les faunes de rongeurs (Mammalia) plio-pléistocènes de la province de Gauteng (Afrique du Sud): mises au point et apports systématiques, biochronologiques et précisions paléoenvironnementales [Thèse de Doctorat]. Université de Montpellier 2, Montpellier, France; 2000.
- 75. Faulkes CG, Verheyen E, Verheyen W, Jarvis J, Bennett NC. Phylogeographical patterns of genetic divergence and speciation in African mole-rats (Family: Bathyergidae). Molecular ecology. 2004; 13(3). pmid:14871365
- 76. Butynski TM, Mattingly R. Burrow structure and fossorial ecology of the springhare Pedetes capensis in Botswana. African Journal of Ecology. 1979;1 7(4): 205–15.
- 77. Pickford M, Mein P. New Pedetidae (Rodentia: Mammalia) from the Mio-Pliocene of Africa. Estudios Geológicos. 2011; 67(2): 455–69.
- 78. Azzarà B, Cherin M, Adams J, Boschian G, Crotti M, Denys C, et al. The Thorny Issue of African Porcupines: a New Mandible of Hystrix makapanensis from Olduvai Gorge (Tanzania) and Rediagnosis of the Species. J Mammal Evol. 2022; 29(2): 447–74. pmid:35079214
- 79. Kaplan J. The Umhlatuzana Rock Shelter sequence: 100 000 years of Stone Age history. Southern African Humanities. 1990.
- 80. Clark J, Plug I. Animal exploitation strategies during the South African Middle Stone Age: Howiesons Poort and post-Howiesons Poort fauna from Sibudu Cave. Journal of human evolution. 2008; 54: 886–98. pmid:18234284
- 81. Tilson RL, LeRoux P. Resource partitioning in coexisting Namib Desert owls, Bubo africanus and Tyto alba. Madoqua. 1983; 13(3): 221–7.
- 82. Sénégas F, Paradis E, Michaux J. Homogeneity of fossil assemblages extracted from mine dumps: An analysis of Plio-Pleistocene fauna from South African caves. Lethaia. 2005;38: 315–22.
- 83.
Dean WRJ. The ecology of owls at Barberspan, Transvaal. In: Proc symp African Predatory Birds. Pretoria: Transvaal Ornithological Society; 1977. p. 25–45.
- 84. Pitman CRS, Adamson J. Notes on the ecology and ethology of the giant owl Bubo lacteus (Temminck). Part 2. Honeyguide. 30 nov 1978; 96: 26–43.
- 85. Denys C, Viriot L, Daams R, Pelaez-Campomanes P, Vignaud P, Andossa L, et al. A New Pliocene Xerine Sciurid (Rodentia) from Kossom Bougoudi, Chad. Journal of Vertebrate Paleontology. 2003; 23(3): 676–87.
- 86. Cuenca-Bescós G. Revision de los Sciuridae del Aragoniense y del Rambliense en la fosa de Calatayud-Montalbán. Scripta Geologica. 1988; 87: 1–116.
- 87. Viriot L, Peláez-Campomanes P, Vignaud P, Andossa L, Mackaye H, Brunet M. A New Xerinae (Rodentia, Sciuridae) from the Late Miocene of Toros-Menalla (Chad). Journal of Vertebrate Paleontology. 2011; 31: 844–8.
- 88. Kryštufek B, Mahmoudi A, Tesakov AS, Matějů J, Hutterer R. A review of bristly ground squirrels Xerini and a generic revision in the African genus Xerus. Mammalia. 2016; 80(5): 521–40.
- 89. Cooke HBS. Some fossil mammals in the South African Museum collections. Annals of the South African Museum Annale van die Suid-Afrikaanse Museum. 1955; 42: 161–8.
- 90. Kuman K, Clarke RJ. Florisbad—New investigations at a middle stone age Hominid Site in South Africa. Geoarchaeology. 1986; 1(2): 103–25.
- 91. Avery DM. The micromammalian fauna from Border Cave, Kwazulu, South Africa. Journal of Archaeological Science. 1 juin 1982;9(2): 187–204.
- 92.
Lavocat R. Sur l’âge des faunes de rongeurs des grottes à australopithèques. In: Proceedings of Third Pan-African Congress on Prehistory. Livingstone; 1957.