Rodentia is the most species-rich order among mammals, making up over 40% of all living mammal species. 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.
Rodents are a very diverse group of mammals, with an exceptional taxonomic and phenotypic diversity. This variability in body shapes and sizes, together with morphological and physiological adaptations, resulted in the successful colonization of most terrestrial environments.
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.
Steppe landscape in South Africa, a common habitat for various rodent species.
Read also: Property Practitioners Regulatory Authority
Rodents in South Africa's Ecosystem
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. 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.
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.
Identification Challenges and Solutions
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 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.
Read also: Discover Thula Thula
More recently, the systematics of many taxa has been partially resolved, with several new species being described and published in taxonomic reference publications, as well as in many systematic studies based upon genetic analyses and field surveys. 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.
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.
Rodent Genera 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. Table 1 provides a list of Pleistocene and modern rodent species from Republic of South Africa (RSA) based on Wilson et al. 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.
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.
Read also: Traditional South African Bread
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.
Cranio-dental features of rodents
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.
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.
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. They were designed using R software version 4.1.0.
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.
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 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.
Dental and Alveolar Formulas
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. 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.
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.
Mammal Skull Identification Lecture
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.
Here is a table summarizing the measurements used for rodent identification:
| Measurement | Description |
|---|---|
| LUTR | Length of the upper toothrow (including molars and premolar(s) if present) |
| WM1 (Upper) | Width of the first upper molar |
| LLTR | Length of the lower toothrow |
| WM1 (Lower) | Width of the first lower molar |
