Home Assessment of craniometric sexual dimorphism and ontogenetic variation in invasive Rattus norvegicus and R. rattus from urban and peri-urban areas of Gauteng Province, South Africa
Article Publicly Available

Assessment of craniometric sexual dimorphism and ontogenetic variation in invasive Rattus norvegicus and R. rattus from urban and peri-urban areas of Gauteng Province, South Africa

  • Gordon Ringani ORCID logo EMAIL logo , Tsungai Zengeya ORCID logo , Christian Pirk and Christian Chimimba
Published/Copyright: June 27, 2022
Download Article (PDF)
Mammalia
From the journal Mammalia Volume 86 Issue 6

Abstract

We evaluated craniometric sexual dimorphism and ontogenetic (age) variation in invasive Rattus norvegicus and R. rattus from urban and peri-urban areas of Gauteng Province, South Africa, using univariate and multivariate analyses. Two-way analysis of variance (ANOVA), percent contribution of the sum of squares (%SSQs) of each source of variation, principal components analysis (PCA) and unweighted pair-group method with arithmetic averages (UPGMA) cluster analysis showed no sexual dimorphism in both species, however in both species, significant age variation between five age classes based on maxillary molar toothrow cusp eruption and wear was found and the age classes were pooled into juveniles (i.e., individuals of tooth-wear class I), sub-adults (II–III), and adults (IV–V). Few variables showed statistically significant sex-age interaction. The largest %SSQs to the total variance were due to error (i.e., residual), suggesting that apart from sex, age, and their interaction, there were other components that are responsible for the variation. Our approach may be useful for partitioning the effect of sexual dimorphism and ontogenetic variation in other studies, such as our stable isotope analysis-based trophic ecological studies of Rattus species from urban and peri-urban areas of Gauteng Province, South Africa.

Keywords: craniometrics; invasive; muridae; ontogenesis; synanthropic

1 Introduction

Three invasive synanthropic murid species, the brown rat (Rattus norvegicus), the black rat (Rattus rattus), and the Tanezumi rat (Rattus tanezumi) have been reported to occur in South Africa (Bastos et al. 2011) where they cause negative health and socio-economic impacts, such as zoonotic diseases and damage to agricultural products and infrastructure (Kay and Hoekstra 2008). R. norvegicus is considered to have originated from eastern Siberia and northern China (Long 2003) and was first recorded in South Africa in 1832 after the arrival of the first Europeans (Avery 1982). R. rattus is considered to have originated from India and southern Asia, and spread to almost all the continents, except Antarctica (Aplin et al. 2003). The native range of R. tanezumi is considered to stretch from Afghanistan to the islands of Japan (Long 2003) and was only recorded to occur in South Africa (and Africa) ∼15 years ago following routine genetic profiling of rodents in the country (Bastos et al. 2011). Rattus rattus and R. tanezumi are morphologically indistinguishable, leading to both species being placed within the R. rattus species complex (Musser and Carleton 2005), and may explain why R. tanezumi was not detected in South Africa (and Africa) until recently (Bastos et al. 2011).

R. norvegicus was considered to be restricted to the coastal regions of South Africa, however recent species distribution models (SDMs) have predicted that it may not only be restricted to the coast but also occurs in inland urban areas (Ringani et al. 2022). Rattus rattus was considered to be widely distributed throughout South Africa, except in the drier parts (Skinner and Smithers 1990), and this distribution is also supported by recent SDMs (Ringani et al. 2022). Although the distribution of R. tanezumi remains poorly documented (Ramatla et al. 2019), recent SDMs predicted it to mainly occur in most inland urban areas and along the coast (Ringani et al. 2022). All these SDMs showed that all three species of Rattus have wide distributions in South Africa that overlap extensively, suggesting that their establishment and spread may be influenced by similar factors such as proximity to urban areas under wet and moderate climatic conditions (Ringani et al. 2022). In Gauteng Province, all three species occur in urban and peri-urban landscapes that are related to general lack of sanitation and municipal services particularly in informal human settlements (Jassat et al. 2013).

A stable isotope analysis (SIA) study was initiated to assess the trophic ecology of the three Rattus species in urban and peri-urban areas of Gauteng Province and evaluate how access to available food resources and associated niche complementarity may aid their establishment. However, given that sexual dimorphism and ontogeny can affect patterns of food resource use and trophic inter-relationships among rodents (Kajiura and Rollo 2011), it was necessary to determine the nature and extent of sexual dimorphism and ontogenetic variation within the species based on craniometric data which forms the subject of the current study.

Sexual dimorphism and ontogenetic variation represent critical life history traits of rodents (Schulte-Hofstedde 2007), and together with other factors such as habitat and season can affect the trophic niche utilised by a species (Hobson and Quirk 2014). While pelage colouration and body mass have been used to age small mammals, these are considered inappropriate, as they may be influenced by factors such as the concentration of humic substances in the soil that are created during the decomposition of organic matter (Rios and Álvarez-Castañeda 2007) and the availability and accessibility of quality food (van Rensburg et al. 2004), respectively. Maxillary molar eruption and wear, and their associated craniometric analysis, however, are considered more appropriate for ageing small mammals (Chimimba and Dippenaar 1994). While craniometric analysis of sexual dimorphism and ontogenetic variation based on a range of univariate analyses can be used for ageing small mammals, Leamy (1983) suggested the use of the more computationally simplistic percent contribution of the sum of squares (%SSQ) of each source of variation that can be computed from a two-way analysis of variance (ANOVA) table.

The aim of the present study therefore, was to first independently assess the nature and extent of sexual dimorphism and ontogenetic variation in R. norvegicus and R. rattus, two of the three invasive synanthropic species that occur in urban and peri-urban landscapes in Gauteng Province in South Africa that had adequate samples for analysis, using two-way ANOVA and %SSQ for comparison, and multivariate analyses based on craniometric data. This was undertaken in order to ascertain: 1) whether the sexes should be pooled or analysed separately; and 2) which samples represent different relative age groups (i.e., juveniles, sub-adults, and adults) within the sampled populations of the two species.

2 Materials and methods

2.1 Study area and sampling

Genetically-identified (through Cyt b mtDNA sequencing) R. norvegicus and R. rattus in this study were from urban Alexandra (−26.1038° S; 28.0962° E) and Tembisa (−26° S; 28.214° E) Townships, and peri-urban University of Pretoria (UP) Experimental Farm (−25.749° S; 28.238° E), Gauteng Province, South Africa (Figure 1; Supplementary Appendices 1–3) and were sourced from previous studies (Bastos et al. 2011; Julius 2013; Lithole 2015; Mostert 2009). These were either live-trapped using Sherman traps (H.B. Sherman Traps Inc. Florida, U.S.A.), obtained opportunistically from members of the public and pest control companies, or through an invasive synanthropic rodent extermination programme by the Johannesburg Metropolitan Municipality, Gauteng Province, South Africa. Specimens were weighed and dissected in a biosecurity level 2 laboratory in the Department of Zoology and Entomology, University of Pretoria, Pretoria, South Africa. All specimens are currently housed in the Department of Zoology and Entomology at the University of Pretoria, Pretoria, South Africa and their voucher numbers are provided in Supplementary Appendices 1–3, and will eventually be deposited in the Ditsong Museum of Natural History (formerly the Transvaal Museum), Pretoria, South Africa.

Figure 1: A map (with an insert of Africa and South Africa) showing the sampling sites of R. norvegicus and R. rattus in Gauteng Province, South Africa.
Figure 1:

A map (with an insert of Africa and South Africa) showing the sampling sites of R. norvegicus and R. rattus in Gauteng Province, South Africa.

2.2 Skull preparation, ageing, and cranio-dental measurements

Skulls were soaked in a 90 °C detergent/water solution to loosen connective tissue and dissolve adipose tissue (Mairs et al. 2004), and loosened tissue was removed, and dried skulls examined under a Vickers stereomicroscope (Vickers Instruments Ltd., York, England). Skulls were allocated to five relative age classes based on the degree of maxillary molar toothrow cusp eruption and wear (Chimimba and Dippenaar 1994) as exemplified by R. norvegicus from Alexandra Township (Figure 2) that had the largest sample of both sexes and all five tooth wear classes (Table 1). As modified from Chimimba and Dippenaar (1994), and exemplified by R. norvegicus, sexual dimorphism and ontogenetic variation were assessed using 13 linear cranio-dental measurements (Figure 3) recorded to the nearest 0.05 mm using a pair of Mitutoyo digital callipers (Mitutoyo American Corporation, Aurora, Illinois, U.S.A.).

Figure 2: (A) Left maxillary molar toothrow showing the degree of molar eruption and wear and definitions of five tooth-wear classes (I–V) used for relative ageing of R. norvegicus and R. rattus from Gauteng Province, South Africa. Age class I: M3 in the process of erupting, with no erosion of enamel on either M1 or M2, and if erosion was apparent, it was slight with no connections between the dentine areas; Age class II: The enamel was lightly worn in some areas and dentine was slightly apparent where erosion of enamel occurred, and dentine areas may have joined cusps together; Age class III: The cusps were connected by dentine areas either longitudinally or transversely where enamel was eroded; Age class IV: The enamel was completely worn, although outlines of the cusps still remained; and Age class V: The enamel was completely worm and only the underlying dentine was visible, with the cusp outlines being barely visible. Scale = 1 mm. (B) Ventral view and specimen voucher number of each representative skull positioned under its corresponding left maxillary molar toothrow. Scale = 10 mm.
Figure 2:

(A) Left maxillary molar toothrow showing the degree of molar eruption and wear and definitions of five tooth-wear classes (I–V) used for relative ageing of R. norvegicus and R. rattus from Gauteng Province, South Africa. Age class I: M3 in the process of erupting, with no erosion of enamel on either M1 or M2, and if erosion was apparent, it was slight with no connections between the dentine areas; Age class II: The enamel was lightly worn in some areas and dentine was slightly apparent where erosion of enamel occurred, and dentine areas may have joined cusps together; Age class III: The cusps were connected by dentine areas either longitudinally or transversely where enamel was eroded; Age class IV: The enamel was completely worn, although outlines of the cusps still remained; and Age class V: The enamel was completely worm and only the underlying dentine was visible, with the cusp outlines being barely visible. Scale = 1 mm. (B) Ventral view and specimen voucher number of each representative skull positioned under its corresponding left maxillary molar toothrow. Scale = 10 mm.

Table 1:

Sample size (n) of sexes and relative age classes of R. norvegicus from Alexandra and Tembisa Townships, and R. rattus from the University of Pretoria (UP) Experimental Farm, Gauteng Province, South Africa used for craniometric analyses. Age classes are defined and illustrated in Figure 2.

Species Sampling site Sex Age class (n) Total
I II III IV V
R. norvegicus Alexandra Township Male 1 2 6 20 1 30
Female 1 2 8 15 6 32
Tembisa Township Male 1 – 2 4 – 7
Female – – 3 8 1 12
R. rattus UP Experimental Farm Male 1 1 5 3 – 10
Female – – 1 4 – 5
Figure 3: Dorsal (A), ventral (B), lateral (C), posterior (D) views of the cranium, and medial (E), and dorsal (F) views of the dentary showing abbreviations and reference points of cranio-dental measurements recorded from specimens of R. norvegicus and R. rattus from Alexandra and Tembisa Townships, and the University of Pretoria (UP) Experimental Farm, Gauteng Province, South Africa. 1) GLS: Greatest length of skull, taken from the anterior edge of the nasal bone to the posterior edge of the skull; 2) FRO: Greatest length of the frontals; 3) NPP: Distance measured from the anterior edge of the nasal bone to the anterior edge of the zygomatic arch; 4) BBC: Breath of brain-case; 5) IOB: Interorbital constriction, measured as the least distance between the orbits; 6) GHS: Greatest height of the skull, measured perpendicular to the horizontal plane; 7) FMH: Foramen magnum height; 8) LFM: Length of M1; 9) WSM: Width of M2; 10) AFA: Distance from the angular process to the mandibular condyle; 11) MAF: Distance from the mandibular foramen to the mandibular condyle; 12) IML: Distance from the posterior incisor to M3; 13) WMS: Width of M2.
Figure 3:

Dorsal (A), ventral (B), lateral (C), posterior (D) views of the cranium, and medial (E), and dorsal (F) views of the dentary showing abbreviations and reference points of cranio-dental measurements recorded from specimens of R. norvegicus and R. rattus from Alexandra and Tembisa Townships, and the University of Pretoria (UP) Experimental Farm, Gauteng Province, South Africa. 1) GLS: Greatest length of skull, taken from the anterior edge of the nasal bone to the posterior edge of the skull; 2) FRO: Greatest length of the frontals; 3) NPP: Distance measured from the anterior edge of the nasal bone to the anterior edge of the zygomatic arch; 4) BBC: Breath of brain-case; 5) IOB: Interorbital constriction, measured as the least distance between the orbits; 6) GHS: Greatest height of the skull, measured perpendicular to the horizontal plane; 7) FMH: Foramen magnum height; 8) LFM: Length of M1; 9) WSM: Width of M2; 10) AFA: Distance from the angular process to the mandibular condyle; 11) MAF: Distance from the mandibular foramen to the mandibular condyle; 12) IML: Distance from the posterior incisor to M3; 13) WMS: Width of M2.

2.3 Statistical analysis

Craniometric data were first tested for normality and homoscedasticity, and subjected to a two-way analysis of variance (ANOVA) to test for sexual dimorphism, age variation, and their interaction. Where statistically significant differences were detected, Tukey’s post hoc tests were used to identify non-significant subsets of relative age classes (P > 0.05). The percent sum of squares (%SSQs) from the derived two-way ANOVA table were used to identify sources of variation (Leamy 1983) with reference to sex, relative age, their interaction, and error (i.e., residual). The %SSQs were computed by dividing the SSQ associated with each source of variation by the total SSQ. The samples however, were characterised by small sample sizes in some sampling localities and age classes, such that not all age classes could be subjected to univariate analysis.

Because of equivocal results of some univariate analyses, multivariate analyses were also used to further evaluate sexual dimorphism and age variation. This included principal components analysis (PCA) of standardized variables based on correlation coefficients among variables (Sneath and Sokal 1973). However, given some relatively small sample sizes, the data for each of the three populations were first subjected to Kaiser–Meyer–Olkin (KMO) tests in order to assess their suitability for PCA (Kaiser 1970). These exploratory analyses revealed that unlike the data for R. rattus from UP Experimental Farm (KMO = 0.41), the data for R. norvegicus from Alexandra (KMO = 0.82) and Tembisa (KMO = 0.68) Townships were at least suitable for PCA. Additional exploratory PCAs also revealed relatively low levels of variation explained by successive principal components to the total variance (e.g., the first two PCA axes of R. norvegicus from Alexandra (60.52%); see Results section).

Consequently, the results of the exploratory analyses and for comparison, the data for the three populations were further subjected to Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster analysis based on Euclidean distances and correlation coefficients among groups (Sneath and Sokal 1973) which are not constrained by the data suitability requirements of PCA. Unlike the univariate analyses, the multivariate analyses also allowed the analysis of a wider range of localities, sexes, and age classes regardless of sample size (see Table 1). All univariate and multivariate analyses were based on the 13 cranio-dental measurements and were based on algorithms in R (R Core Team 2021) and in Paleontological Statistics Software package (PAST) (Hammer et al. 2001).

3 Results

3.1 Univariate assessment

The results of two-way ANOVA and %SSQs of sexual dimorphism and ontogenetic variation of R. norvegicus from Alexandra and Tembisa Townships, and R. rattus from UP Experimental Farm (Table 2) were broadly similar. The two-way ANOVA showed more cranio-dental measurements with statistically significant F-values for age than for sex in all the three samples (Table 2). Age also gave higher %SSQ values (range %SSQ = 0.76–56.55 for R. norvegicus from Alexandra Township; range %SSQ = 3.08–64.96 for R. norvegicus from Tembisa Township; range %SSQ = 4.84–66.63 for R. rattus from UP Experimental Farm than for sex [range %SSQ = 0.01–6.39 for R. norvegicus from Alexandra Township; range %SSQ = 0.18–15.93 for R. norvegicus from Tembisa Township; and range %SSQ = 1.52–25.05 for R. rattus from UP Experimental Farm (Table 2)].

Table 2:

F-values and percent sum of squares (%SSQ) of each source of variation from a two-way analysis of variance (ANOVA) of five relative age classes (I–V) of male and female R. norvegicus from Alexandra and Tembisa Townships, and R. rattus from the University of Pretoria (UP) Experimental Farm, Gauteng Province, South Africa.

R. norvegicus (Alexandra) R. norvegicus (Tembisa) R. rattus (UP experimental farm)
F-values %SSQ F-values %SSQ F-values %SSQ
Cranio-dental measurement Sex (S) Age (A) S × A Sex (S) Age (A) S × A Error Sex (S) Age (A) S × A Sex (S) Age (A) S × A Error Sex (S) Age (A) S × A Sex (S) Age (A) S × A Error
GLS 0.59 18.08c 2.13 0.78 56.55 6.00 36.67 1.36 7.97a 0.56 6.15 50.88 2.07 40.90 0.43 11.36b 2.96 5.55 57.02 9.25 28.17
FRO 0.35 0.90 0.97 1.54 4.99 6.49 86.98 0.36 8.85a 0.04 1.70 51.66 0.16 46.49 0.26 7.73a 1.64 4.33 51.09 6.86 37.73
NPP 2.25 15.26c 3.13 0.22 50.04 9.65 40.09 2.28 19.56c 0.27 5.52 64.96 0.71 28.81 1.80 9.15a 0.02 10.78 44.89 0.10 44.23
BBC 2.28 5.13c 2.49 0.01 23.44 12.31 64.25 0.24 1.93 0.67 2.32 24.88 4.15 68.64 0.74 5.02 0.09 6.64 35.55 0.54 57.27
IOB 0.31 2.71a 1.37 0.61 17.61 7.78 74.00 0.28 0.68 1.77 0.78 12.87 11.97 74.39 1.03 0.68 0.74 15.90 4.84 6.04 73.22
GHS 2.28 12.90c 3.17 0.30 44.85 10.75 44.10 2.33 12.37b 0.00 7.63 56.91 0.00 35.46 0.70 2.84 4.09 12.83 24.38 19.64 43.16
FMH 0.12 0.10 0.15 2.13 0.76 1.08 96.03 0.87 0.77 0.15 6.17 6.79 1.18 85.86 0.34 0.19 0.04 4.86 5.47 0.41 89.26
LFM 1.11 1.08 1.10 0.38 8.42 7.13 84.06 0.01 1.08 0.07 0.18 8.33 0.59 90.89 0.33 6.40a 0.02 1.52 54.68 0.08 43.71
WSM 0.29 0.07 0.53 1.05 1.28 3.81 93.86 1.93 1.32 0.46 15.93 3.08 3.26 77.74 2.55 0.04 5.16a 4.54 9.44 31.36 54.66
AFA 0.60 11.72c 1.33 3.11 49.50 4.39 43.00 0.27 4.98a 1.44 2.17 42.71 6.38 48.73 0.81 17.00b 0.86 4.89 66.63 2.50 25.98
MAF 0.00 8.61c 1.66 6.39 34.68 6.67 52.25 0.19 4.95a 0.02 1.02 35.76 0.14 63.08 1.51 12.51b 0.48 9.26 54.83 1.80 34.11
IML 16.98c 27.66c 8.20c 3.89 55.52 15.70 24.90 0.73 4.23 0.56 4.67 38.29 2.78 54.27 1.24 4.91 0.00 9.64 30.23 0.01 60.13
WMS 2.62 2.79a 1.89 0.34 15.80 10.67 73.19 1.80 1.90 0.55 10.06 15.51 3.57 70.86 2.18 0.54 1.25 25.05 9.92 7.93 57.10
Mean 1.60 27.96 7.88 62.57 4.95 31.74 2.84 60.47 8.91 34.54 6.65 49.90

  1. Cranio-dental measurements are defined and illustrated in Figure 3. Statistical significance: a P < 0.05; b P < 0.01; c P < 0.001.

The greatest length of the skull (GLS), the distance from nasal bone to the anterior edge of the zygomatic arch (NPP), and the distance from the angular process to the mandibular condyle (AFA) showed statistically significant age variation in all the three samples (Table 2). These cranio-dental measurements also contributed relatively more to the total variance due to age variation (Table 2). Only one cranio-dental measurement, the distance from the posterior incisor to M3 (IML) in R. norvegicus from Alexandra Township, was statistically significant for both sexual dimorphism and age variation, and together with the width of M2 (WSM) in R. rattus from UP Experimental Farm, were the only two cranio-dental measurements that showed a statistically significant interaction between sex and age (Table 2). The interaction between sex and age showed a negligible contribution in all three samples as indicated by the relatively small mean %SSQs (mean %SSQ = 7.88 for R. norvegicus from Alexandra Township); mean %SSQ = 2.84 for R. norvegicus from Tembisa Township; and mean %SSQ = 6.65 for R. rattus from UP Experimental Farm) (Table 2). In all the three samples, the largest %SSQs were due to error (i.e., residual) (mean % SSQ = 62.57 for R. norvegicus from Alexandra Township; mean %SSQ = 60.47 for R. norvegicus from Tembisa Township; and mean %SSQ = 49.90 for R. rattus from UP Experimental Farm (Table 2).

Tukey’s post hoc test of R. norvegicus from Alexandra, the only locality with adequate age class samples for analysis showed that the following three groupings of age classes were not significantly different from each other (Table 3): 1) individuals of age class I being mainly in its own non-significant subset, with few individuals of age classes II–V; 2) individuals of age class II being in the same non-significant subset with individuals of age class III, with few individuals of age classes I, IV–V; and 3) individuals of age classes IV and V being mainly in the same non-significant subset of age classes, with a few individuals of age classes I–III. Although equivocal, the general pattern that emerged from the Tukey’s post hoc tests of this sample suggested that age class I may represent juveniles, age classes II and III may be sub-adults, while age classes IV and V may represent adult specimens. The progression of cranial-dental dimensions from juveniles to adult individuals is best exemplified by the standard descriptive statistics of R. norvegicus from Alexandra (Table 4A) and Tembisa (Table 4B) Townships, and R. rattus from UP Experimental Farm (Table 4C) that showed a direct relationship between cranial-dental measurement magnitude and relative age.

Table 3:

Tukey’s post hoc tests of cranio-dental measurements (mm) of individuals of R. norvegicus of different age classes (I–V) that were sampled from Alexandra Township, Gauteng Province, South Africa.

Cranio-dental Age class
I (n = 2) II (n = 4) III (n = 14) IV (n = 36) V (n = 7)
GLS 33.98 ± 3.1I 39.12 ± 4.1II 41.69 ± 2.8II,III 44.92 ± 2.7IV,V 47.31 ± 1.2IV,V
FRO 11.64 ± 0.2I,II 12.30 ± 1.3I,II 20.26 ± 0.9III 14.38 ± 0.7IV,V 14.57 ± 1.0IV,V
NPP 24.21 ± 2.4I 26.37 ± 4.1I,II 29.17 ± 2.5II,III 31.79 ± 2.2III,IV,V 33.23 ± 0.9IV,V
BBC 14.43 ± 0.8I,II 15.27 ± 1.7I,II 15.94 ± 1.2III 16.56 ± 0.7IV,V 16.45 ± 1.0IV,V
IOB 6.71 ± 0.9I,II 6.34 ± 0.1I,II 6.45 ± 0.7III 6.93 ± 0.5IV,V 7.24 ± 0.3IV,V
GHS 10.70 ± 0.9I,II 12.11 ± 1.1I,II 11.78 ± 0.7I,II,III 12.88 ± 0.6IV,V 13.14 ± 0.6IV,V
FMH 5.02 ± 0.3I,II 5.43 ± 0.3I,II 5.26 ± 0.8II,III 5.99 ± 4.5IV 5.14 ± 0.6I,II,III
LFM 3.24 ± 0.3I,III 2.94 ± 0.5II,III,IV 2.77 ± 0.4II,III 3.03 ± 0.4I,II,IV 2.73 ± 0.3II,III,V
WSM 2.08 ± 0.1I–V 2.06 ± 0.1I–V 1.99 ± 1.0I–V 2.01 ± 0.2I–V 1.88 ± 0.4I–V
AFA 6.29 ± 0.4I 6.80 ± 0.7II 7.29 ± 0.9III 8.56 ± 0.8IV,V 9.51 ± 1.0III,IV,V
MAF 3.85 ± 0.3I 4.40 ± 0.8II 5.19 ± 1.0III,IV,V 5.81 ± 0.7III,IV,V 6.29 ± 0.9III,IV,V
IML 8.40 ± 5.7I 14.17 ± 0.7II,III 14.89 ± 1.7II,III 16.02 ± 1.0 17.00 ± 1.0IV,V
WMS 1.54 ± 0.6I–V 1.98 ± 0.1II–V 1.81 ± 0.3II–V 2.03 ± 0.1II–V 1.85 ± 0.4II–V

  1. Values are presented as mean ± 1 standard deviation (SD). Prescripts I–V represent non-significant subsets of relative age classes I–V (P > 0.05). n = sample size. Cranio-dental measurements are defined and illustrated in Figure 3.

Table 4:

Standard descriptive statistics of cranio-dental measurements (mm) of males and females and age classes (I–V) of R. norvegicus from Alexandra (A) and Tembisa (B) Townships, and R. rattus from the University of Pretoria (UP) Experimental Farm (C), Gauteng Province, South Africa.

(A) Alexandra township
Cranio-dental measurement
Sex Age class Statistics GLS FRO NPP BBC IOB GHS FMH LFM WSM AFA MAF IML WMS
Female I (n = 1) X ‾ 31.75 11.45 22.49 13.83 6.05 10.06 4.78 3.46 2.22 5.99 3.59 12.50 1.10
SD – – – – – – – – – – – – –
CV – – – – – – – – – – – – –
II (n = 2) X ‾ 36.86 11.40 23.29 14.31 6.20 11.20 5.40 3.33 2.01 6.47 3.88 13.59 2.05
SD 3.09 0.35 0.16 2.30 0.06 0.18 0.49 0.18 0.01 0.33 0.54 0.40 0.07
CV 8.38 3.10 0.70 16.06 1.03 1.58 9.04 5.32 0.35 5.03 14.05 2.97 3.45
III (n = 8) X ‾ 42.72 13.65 29.90 16.38 6.70 12.01 5.36 2.84 2.23 7.41 5.62 15.69 1.87
SD 3.15 1.04 2.60 1.03 0.79 0.79 0.81 0.47 1.37 1.14 0.90 1.48 0.38
CV 7.37 7.61 8.69 6.28 11.85 6.55 15.17 16.63 61.75 15.42 15.95 9.44 20.11
IV (n = 15) X ‾ 45.43 14.47 32.27 16.64 6.95 13.07 5.18 2.95 2.03 8.84 6.20 16.15 2.05
SD 2.32 0.84 1.97 0.65 0.66 0.65 0.38 0.30 0.27 0.89 0.52 1.16 0.18
CV 5.12 5.80 6.09 3.91 9.51 5.00 7.37 10.23 13.07 10.06 8.41 7.16 8.70
V (n = 6) X ‾ 47.61 14.79 33.32 16.32 7.29 13.17 5.01 2.71 1.87 9.80 6.31 17.11 1.80
SD 1.11 0.89 1.05 1.06 0.42 0.67 0.55 0.38 0.50 0.74 1.09 1.15 0.45
CV 2.34 6.03 3.15 6.50 5.72 5.11 11.00 14.19 26.60 7.57 17.31 6.71 24.87
Male I (n = 1) X ‾ 36.21 11.83 25.92 15.02 7.36 11.34 5.26 3.02 1.94 6.58 4.10 4.30 1.98
SD – – – – – – – – – – – – –
CV – – – – – – – – – – – – –
II (n = 2) X ‾ 41.40 13.20 29.50 16.20 6.50 13.00 5.50 2.60 2.10 7.10 4.90 14.80 1.90
SD 4.70 1.30 3.50 0.70 0.00 0.90 0.20 0.50 0.30 1.00 0.80 0.50 0.00
CV 11.30 10.00 12.00 4.10 0.10 7.20 3.10 21.00 15.00 13.70 15.60 3.50 1.80
III (n = 6) X ‾ 40.30 29.30 28.20 15.40 6.10 11.50 5.10 2.70 1.70 7.10 4.60 13.80 1.70
SD 2.00 41.50 2.50 1.40 0.60 0.60 1.00 0.60 0.40 0.80 0.90 1.50 0.40
CV 4.90 141.60 8.80 9.20 10.00 4.90 19.10 20.70 23.60 10.60 20.20 10.60 23.40
IV (n = 20) X ‾ 45.00 14.4 31.70 16.60 7.00 12.80 6.70 3.10 2.00 8.40 5.60 16.10 2.00
SD 2.50 0.50 2.10 0.70 0.60 0.60 6.10 0.60 0.30 0.80 0.70 0.80 0.20
CV 5.50 3.70 6.60 4.50 8.10 5.00 91.10 18.40 13.40 9.00 12.50 4.70 9.60
V (n = 1) X ‾ 45.50 13.20 32.70 17.30 7.00 13.00 5.90 2.80 1.90 7.80 6.20 16.40 2.20
SD – – – – – – – – – – – – –
CV – – – – – – – – – – – – –
(B) Tembisa township
Sex Age class Statistics Cranio-dental measurement
GLS FRO NPP BBC IOB GHS FMH LFM WSM AFA MAF IML WMS
Female III (n = 3) X ‾ 37.39 11.53 25.71 15.36 6.38 11.32 5.26 2.89 2.16 6.39 4.20 13.08 2.01
SD 3.21 0.11 1.82 1.05 0.67 0.74 0.26 0.26 0.29 0.89 0.84 1.21 0.03
CV 8.57 0.92 7.07 6.85 10.55 6.53 4.97 8.9 13.25 13.99 20.10 9.24 1.25
IV (n = 8) X ‾ 43.90 13.88 29.21 16.48 6.85 12.84 5.48 3.06 2.11 8.29 5.20 14.71 2.16
SD 3.02 1.08 1.58 0.88 0.27 0.44 0.29 0.34 0.13 0.92 0.88 0.93 0.14
CV 6.88 7.78 5.41 5.35 3.91 3.46 5.33 11.08 6.30 11.07 16.87 6.35 6.31
Male III (n = 2) X ‾ 36.63 11.23 23.9 15.53 6.78 10.65 5.49 2.85 2.09 6.77 3.92 13.02 2.16
SD 3.92 2.80 0.33 0.33 0.06 1.76 0.04 0.49 0.04 1.12 0.43 0.83 0.04
CV 10.72 24.93 1.39 2.09 0.83 16.54 0.77 17.37 2.03 16.62 11.02 6.36 1.64
IV (n = 4) X ‾ 40.41 13.29 28.32 15.82 6.67 12.19 5.58 3.15 1.91 7.34 5.06 13.78 2.20
SD 0.65 0.74 0.01 0.10 0.13 0.35 0.42 0.64 0.06 0.67 0.01 0.71 0.14
CV 1.61 5.53 0.05 0.63 1.91 2.84 7.48 20.20 2.96 9.16 0.28 5.18 6.43
C) UP experimental farm
Sex Age class Statistics Cranio-dental measurements
GLS FRO NPP BBC IOB GHS FMH LFM WSM AFA MAF IML
Female III (n = 1) X ‾ 38.28 12.35 24.52 15.58 5.80 11.42 5.26 3.04 1.70 6.95 4.18 12.32 1.80
SD – – – – – – – – – – – – –
CV – – – – – – – – – – – – –
IV (n = 4) X ‾ 40.48 12.83 27.77 16.62 5.79 11.32 5.12 2.58 1.81 8.01 5.14 13.34 1.98
SD 2.39 0.36 1.61 1.07 0.21 0.63 0.51 0.21 0.06 0.67 0.62 0.43 0.23
CV 5.91 2.77 5.79 6.46 3.68 5.58 10.04 8.34 3.35 8.34 12.14 3.23 11.83
Male I (n = 1) X ‾ 30.35 11.13 18.24 12.05 5.08 9.30 5.36 3.02 1.73 5.52 3.48 10.25 1.74
SD – – – – – – – – – – – – –
CV – – – – – – – – – – – – –
II (n = 1) X ‾ 40.86 12.98 28.79 16.87 6.01 11.16 5.77 3.07 1.86 7.91 5.2 13.53 1.90
SD – – – – – – – – – – – – –
CV – – – – – – – – – – – – –
III (n = 5) X ‾ 36.86 12.11 25.87 15.88 5.84 11.08 5.34 3.18 1.87 6.94 4.36 12.83 1.77
SD 2.12 0.37 2.01 0.66 0.60 0.33 0.12 0.23 0.07 0.43 0.53 0.90 0.06
CV 5.74 3.04 7.79 4.17 10.28 2.99 2.22 7.24 3.71 6.19 12.16 6.99 3.32
IV (n = 3) X ‾ 43.64 13.40 29.43 17.23 6.33 12.15 5.29 2.66 1.78 8.61 5.79 13.88 1.73
SD 0.58 0.75 0.88 0.52 0.24 0.17 0.25 0.45 0.07 0.26 0.18 0.57 0.03
CV 1.32 5.58 2.99 3.00 3.75 1.36 4.68 16.98 3.89 3.05 3.04 4.08 2.00

  1. n = sample size; X ‾  = arithmetic mean; SD = ±1 standard deviation; CV = coefficient of variation. Cranio-dental measurements are defined and illustrated in Figure 3.

3.2 Multivariate assessment

Because of some equivocal results of the Tukey’s post hoc tests, multivariate PCA and UPGMA cluster analyses were also used to further evaluate sexual dimorphism and age variation in R. norvegicus and R. rattus. These analyses allowed the evaluation of the two species from the three sampling sites irrespective of within-cell sample size limitations of the sexes and age classes. Similar to the results of the univariate analyses, the results of the multivariate analyses were broadly similar.

The PCA of all three populations of Rattus generated 13 principal components (PC) axes, with most eigenvalues that explain the variables responsible for the variability, and the percent contribution to the total variance being attributed to the first two PC axes (R. norvegicus from Alexandra Township: eigenvalue for the first two PC axes = 7.87, total %variance contribution of the first two PC axes = 60.25%); R. norvegicus from Tembisa Township: 9.17, 70.48%; R. rattus from UP Experimental Farm: 9.14, 70.35%) (Table 5). The PCA scattergram of R. norvegicus from Alexandra Township, which was represented by all sexes and all five age classes, reflects age rather than sex as a major source of variation (Figure 4). The single individual of age class I tended to separate from the rest of the age classes. While there was no separation between individuals of age classes II and III, the individuals of these two age classes, with some overlaps, tended to separate from individuals of age classes IV and V. The PCA scattergram of R. norvegicus from Tembisa Township, which was represented by all sexes and age classes I (by a single individual), III, IV and V (by a single individual), also reflects age rather than sex as a major source of variation (Figure 4). The single individual of age class I tended to separate from the rest of the age classes. With a slight overlap (by two individuals), individuals of age classes III and IV were separate from each other. The single individual of age class V tended to separate from the rest of the age classes. The PCA scattergram of R. rattus from UP Experimental Farm (an analysis solely undertaken for comparison with the results of UPGMA cluster analysis because of the data’s low KMO value), which was represented by all sexes and age classes I (by a single individual), II (by a single individual), III and IV, also reflects age rather than sex as a major source of variation (Figure 4). The single individual of age class I separated from the rest of the age classes. The single individual of age class II also separated from the rest of the age classes. Individuals of age classes III and IV were also separated from each other.

Table 5:

The number of principal components (PC) axes, eigenvalues and percent (%) variances explained by each PC axis generated from each principal components analysis (PCA) of R. norvegicus from Alexandra and Tembisa Townships, and R. rattus from the University of Pretoria (UP) Experimental Farm, Gauteng Province, South Africa.

Cranial measurement R. norvegicus (Alexandra township) R. norvegicus (Tembisa township) R. rattus (UP experimental farm)
Principal components Principal components Principal components
I II I II I II
GLS 0.39 −0.16 0.32 −0.03 0.31 −0.04
FRO 0.04 0.00 0.32 0.05 0.31 −0.03
NPP 0.38 −0.13 0.32 −0.06 0.31 0.05
BBC 0.32 0.09 0.29 0.29 0.31 0.04
IOB 0.27 0.05 0.30 0.13 0.31 0.04
GHS 0.36 −0.02 0.29 −0.18 0.30 −0.06
FMH 0.04 0.46 0.06 0.70 0.03 0.62
LFM 0.04 0.59 0.32 −0.03 −0.13 0.62
WSM 0.07 0.45 0.16 −0.59 0.22 0.46
AFA 0.36 −0.09 0.32 0.07 0.31 −0.07
MAF 0.35 −0.08 0.32 0.09 0.31 −0.09
IML 0.30 −0.06 0.31 −0.13 0.31 0.04
WMS 0.19 0.40 −0.04 0.03 0.28 −0.02
%Variance 43.12 17.17 74.59 15.17 79.22 15.60
Figure 4: The first two axes from a principal components analysis (PCA) of cranio-dental data of Rattus norvegicus from Alexandra and Tembisa Townships, and R. rattus from the University of Pretoria Experimental Farm, Gauteng Province, South Africa. Tooth wear classes I: ; II: ; III: ; IV: ; V: . Convex polygons enclose individuals of the same tooth wear class.
Figure 4:

The first two axes from a principal components analysis (PCA) of cranio-dental data of Rattus norvegicus from Alexandra and Tembisa Townships, and R. rattus from the University of Pretoria Experimental Farm, Gauteng Province, South Africa. Tooth wear classes I: ; II: ; III: ; IV: ; V: . Convex polygons enclose individuals of the same tooth wear class.

There were some cranio-dental measurements with relatively high loadings on the first principal components axis that were common in the PCAs of R. norvegicus from Alexandra (43.37% of the total variance) and Tembisa (58.59%) Townships, and R. rattus from UP Experimental Farm (58.10%) (Table 5), and these included: the greatest length of skull (GLS), the distance from nasal bone to the anterior edge of the zygomatic arch (NPP), the distance from the angular process to the mandibular condyle (AFA), the distance from the mandibular foramen to the mandibular condyle (MAF), and the distance from the posterior incisor to M3 (IML). There were some cranio-dental measurements with relatively high loadings on the second principal components axis that were common in the PCAs of R. norvegicus from Alexandra (17.15% of the total variance) and Tembisa (11.89%) Townships, and R. rattus from UP Experimental Farm (12.25%), and these included: foramen magnum height (FMH) and width of M2 (WSM). Collectively, these cranio-dental measurements also featured prominently either as statistically significantly different (except for the distance from the posterior incisor to M3 (IML) in R. norvegicus from Tembisa and R. rattus from UP Experimental Farm) or contributed highly towards the total %SSQs in the univariate analyses (Table 2).

Because of a low KMO value for the R. rattus population from UP Experimental Farm, the relatively low levels of variation explained by successive principal components to the total variance, and for comparative purposes, data for the three Rattus populations were further assessed by UPGMA cluster analysis. Similar to the PCA scattergram, the distance phenogram (Figure 5) of the UPGMA cluster analysis of R. norvegicus from Alexandra Township (cophenetic correlation coefficient (CCC) = 0.99), which was represented by all sexes and all five age classes, reflects age rather than sex as a major source of variation. It showed the following three main clusters: 1) Cluster I – comprising individuals of age class I (n = 2) and a mixture of two individuals of age class II, and three individual of age class III, suggesting that the young age class individuals are generally separated from the older individuals; 2) Cluster II – comprising mainly of individuals of age class III and a mixture of two individuals of age class II and two individuals of age class IV; and 3) Cluster III – except for one individual of age class II and three individuals of age class III, all individuals in this cluster comprise older individuals of age classes IV and V. The phenogram showed no discrete groupings of the sexes.

Figure 5: A distance phenogram from an unweighted pair group method with arithmetic mean (UPGMA) cluster analysis of male (M) and female (F) R. norvegicus from Alexandra (Cophenetic correlation coefficient (CCC) = 0.99) and Tembisa (CCC = 0.73) Townships, and R. rattus from the University of Pretoria (UP) Experimental Farm (CCC = 0.86), Gauteng Province, South Africa. Tooth-wear classes I: ; II: ; III: ; IV: ; V: .
Figure 5:

A distance phenogram from an unweighted pair group method with arithmetic mean (UPGMA) cluster analysis of male (M) and female (F) R. norvegicus from Alexandra (Cophenetic correlation coefficient (CCC) = 0.99) and Tembisa (CCC = 0.73) Townships, and R. rattus from the University of Pretoria (UP) Experimental Farm (CCC = 0.86), Gauteng Province, South Africa. Tooth-wear classes I: ; II: ; III: ; IV: ; V: .

Similar to the PCA scattergram, the distance phenogram of the UPGMA cluster analysis of R. norvegicus from Tembisa Township (CCC = 0.73) (Figure 5), which was represented by all sexes and age classes I (by a single individual), III, IV and V (by a single individual), also reflects age rather than sex as a major source of variation. It showed three main clusters largely separating one individual of age class I, individuals of age class III (except for one individual of age classes IV, and individuals of age classes IV–V (except for one individual of age class III), with no discrete groupings of the sexes. Similarly, the distance phenogram of the UPGMA cluster analysis of R. rattus from UP Experimental Farm (CCC = 0.86) (Figure 5), which was represented by all sexes and age classes I (by a single individual), II (by a single individual), III and IV, also reflects age rather than sex as a major source of variation. It also showed three clusters showing the single individual of age class I, a cluster of individuals of age class II (except for two individuals of age class IV), and a cluster separating individuals of age class IV (except for one individual of age class III), with no discrete groupings of the sexes.

4 Discussion

The results of the univariate and multivariate analysis of R. norvegicus and R. rattus from urban and peri-urban areas of Gauteng Province, South Africa in this study showed no evidence of craniometric sexual dimorphism, however, both species showed significant ontogenetic variation between five age classes based on maxillary molar toothrow cusp eruption and wear. Several studies have indicated that the proximal variables that may exert an influence on the morphology of urban-dwelling rodents include ambient temperature (Porter et al. 2015; Rae et al. 2006), the lack of competition and predation (Angerbjörn 1986), and the quantity and quality of food available (Ventura and Lopez-Fuster 2000).

Wild populations of R. norvegicus and R. rattus are more likely to be exposed to a broader range of temperatures owing to the habitats in which they occur. Previous studies (e.g., Porter et al. 2015; Rae et al. 2006) have alluded to the effect of variation in ambient temperature and latitude on the cranio-dental morphology of Rattus. Laboratory strains of R. norvegicus that had been raised under colder ambient temperatures (5 °C) had subtle cranio-facial morphological differences on their facial shape and the volume of their maxillary sinuses when compared to a control group in which the ambient temperature had been kept at 22 °C (Rae et al. 2006).

Porter et al. (2015) compared R. norvegicus populations from temperate and tropical regions and demonstrated that latitude and seasonal variability could have an influence on the body size of male and female R. norvegicus. The sampled Rattus specimens in our study were sourced from Gauteng Province, which has a sub-tropical climate. Our craniometric analyses indicated that R. norvegicus and R. rattus lack sexually dimorphic cranio-dental morphological traits as similarly found by Ben-Faleh et al. (2012) in their study on the cranial morphology of R. rattus. The opportunistic manner in which our samples were acquired precluded further investigation into the effect that variables such as latitude, altitude, seasonal variability could have on the cranial morphological characteristics of male and female Rattus.

The results of the PCA (Figure 4; Table 5) generally reflected the pattern shown by the UPGMA cluster analysis (Figure 5). When univariate and multivariate results are collectively considered, the pooling of sexes and individuals into juveniles (i.e., tooth-wear class I), sub-adults (tooth-wear classes II and III), and adults (tooth-wear classes IV–V) for our trophic ecological study of Rattus from urban and peri-urban areas of Gauteng Province was justified. The broad pattern that emerged from the univariate and multivariate analyses is that if the tooth wear classes were to be visualised on a hypothetical growth curve, then individuals of age class I would be considered as juveniles, individuals of age classes II and III would be on the growth curve just before it stabilizes and therefore, representing sub-adults, while individuals of age classes IV and V would be considered as adults.

The results of the univariate Tukey’s post hoc tests in our study (Table 3) were equivocal in supporting the case of which age classes to pool in our trophic ecological study of Rattus from urban and peri-urban areas of Gauteng Province. This constraint however, was resolved when the results of the univariate and multivariate analyses were interpreted collectively, showing that there is a progressive increase in cranio-dental dimensions from age classes I to V in R. norvegicus and R. rattus from urban and peri-urban South African populations (Table 4). The initial allocation of individuals into relative age classes indicated that individuals of R. norvegicus and R. rattus in this study can be assigned to five relative age classes, and if necessary, into much broader ageing categories to reflect juvenile, sub-adult, and adult individuals.

The assessment of ontogenetic variation in this study involved the assessment of maxillary molar eruption and wear to establish the relative age of individuals of R. norvegicus and R. rattus from Gauteng Province, South Africa. The erosion of the enamel overlaying the crown of molars can be correlated with relative age in laboratory strains of Norway rats (Sengupta 2011). The potential limitation in the use of molar eruption and wear to assess relative age in rats however, is that differences in their diet may lead to the over- or under-estimation of their relative age (Sengupta 2011). The results of the present study, however, suggest that the orderly tooth-wear structure observed may not have compromised the conclusions reached on relative ageing in R. norvegicus and R. rattus samples.

Only two cranio-dental measurements (the distance from the posterior incisor to M3 (IML) in R. norvegicus from Alexandra Township and width of M2 (WSM) in R. rattus from UP Experimental Farm) in the present study showed statistically significant interaction between sex and age (Table 2). Prado and Percequillo (2011) found that Aegialomys xanthaeolus (Family: Cricetidae) that they assigned to age class III exhibited inadequately defined differences in the molar tooth wear patterns and their broad range of cranio-dental measurements. Male A. xanthaeolus were found to grow larger than females resulting in the observed differences in their morphological features. Males and females from age classes IV to V, however, did not exhibit sexually dimorphic features (Prado and Percequillo 2011). The hypothesized mechanisms that resulted in the differences between the sexes and relative age included the misidentification and mis-assignment of individuals into correct age classes, as well as differences in their diet. Patton and Rogers (1983) concluded that the misidentification and mis-assignment of individuals may have resulted in the observed variation in individuals of age class III. Shine (1978) hypothesized that, at least in snakes, competition between males and females may result in morphological differences between them. Brandt and Pessôa (1994) found sexual dimorphism in relative age class III that was intermediate between juveniles and adults of the Cerradomys langguthi (Family: Cricetidae).

The largest %SSQ contribution to the total %SSQ in this study was due to error (i.e., residual) (Table 2), suggesting that apart from sex, age and their interaction, there may be other components of variation that may be responsible for the nature and extent of variation in R. norvegicus and R. rattus from Gauteng Province. Analyses based on the computationally simplistic %SSQs may be more appropriate for establishing the relative importance of cranio-dental variables when assessing sexual dimorphism and ontogenetic, and other sources of variation in rodents (Leamy 1983). Subsequent studies on partitioning sexual dimorphism and ontogenetic variation should focus on comparing individuals across populations while limiting or adjusting for additional sources of variation that may not be due to these two components of variation. For the analysis of sexual dimorphism and ontogenetic variation, cranio-dental variables with the smallest variances, and with large error components, may be the most appropriate for such analyses (Leamy 1983).

In conclusion, although R. norvegicus from Tembisa Township and R. rattus from UP Experimental Farm were represented by relatively small sample sizes compared to R. norvegicus from Alexandra Township which was represented by the largest sample size, all sexes and all age classes, craniometric analysis in this study (Table 1) showed the general lack of sexual dimorphism in the two species of Rattus from South Africa. These analyses however, showed evidence of statistically significant ontogenetic variation in R. norvegicus and R. rattus. Despite some of the relatively small sample sizes, the results in this study suggest that the sexual dimorphism and ontogenetic variation observed may not have compromised the conclusions reached on relative ageing in the two species of Rattus. Collectively, both univariate and multivariate results justified the pooling of the sexes as well as individuals into juveniles (i.e., individuals of tooth-wear class I), sub-adults (tooth-wear classes II–III), and adults (tooth-wear classes IV–V) for our trophic ecological study of Rattus in urban and peri-urban South African populations. Very few of the cranio-dental variables used in the present study showed statistically significant interaction between sex and age, and the interaction between these two components of variation did not account substantially to the total variance (Table 2). Previous studies have attributed the interaction between sexual dimorphism and age variation in rodents to differences in growth rates and competition for food resources among the sexes and age classes. The analysis based on SSQs revealed that the largest %SSQs to the total %SSQ was due to error (= residual) (Table 2), suggesting that apart from sex, age and the interaction between them, there may be other components of variation that may be responsible for the overall nature and extent of variation in R. norvegicus and R. rattus from urban and peri-urban areas of Gauteng Province, South Africa.

Corresponding author: Gordon Ringani, Department of Zoology and Entomology, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa, E-mail:

Funding source: National Research Foundation

Award Identifier / Grant number: 103602

Acknowledgements

We thank Ms Elmarie Mostert, Dr Rolanda Julius, and Ms Asiashu Lithole for procuring the samples that were used in the present study.

  1. Author contributions: CC, CP and TZ conceptualized the study; GR collected and analysed the data, and all authors draughted and revised the manuscript.

  2. Research funding: This study was funded by the South African Department of Science and Innovation-National Research Foundation (DSI-NRF) Centre of Excellence for Invasion Biology (CIB) (CC; TZ) and the South African National Research Foundation (NRF) (CC; CP, TZ). TZ acknowledges funding from the South African Department of Forestry, Fisheries and the Environment (DFFE). This publication does not, however, necessarily represent the views or opinions of the funders or their employees.

  3. Competing interests: The authors declare that they have no known conflicts of interest or personal relationships that could have influenced this study.

  4. Research ethics: Samples were obtained under permit number CPF6 0032 from Gauteng Department of Nature Conservation, Johannesburg, South Africa. Permit number 13788 to conduct this study was granted under the terms of section 20 of the Animal Diseases Act (Act no. 35 of 1984) of the South African Department of Agriculture, Forestry and Fisheries. An animal ethics approval number ECO25-10 to conduct this study was granted by the Animal Ethics Committee of the University of Pretoria, Pretoria, South Africa.

References

Angerbjörn, A. (1986). Gigantism in island populations of wood mice (Apodemus) in Europe. Oikos 47: 47–56, https://doi.org/10.2307/3565918.Search in Google Scholar

Aplin, K.P., Chesser, T., and Ten Have, J. (2003). Evolutionary biology of the genus Rattus: profile of an archetypal rodent pest. In: Singleton, G.R., Hinds, L.A., Krebs, C.J., and Spratt, D.M. (Eds.). Rats, mice and people: rodent biology and management. Canberra, ACIAR, pp. 487–498.Search in Google Scholar

Avery, D.M. (1982). The dispersal of the brown rat, Rattus norvegicus, and new specimens from 19th century Cape Town. Mammalia 4: 573–592.Search in Google Scholar

Bastos, A.D.S., Deenadayalan, N., Taylor, P., Brettschneider, H., Kirsten, F., Mostert, E., von Maltitz, E., Lamb, J.M., Belmain, S.T., Contrafato, G., et al.. (2011). Genetic monitoring detects an overlooked cryptic species and reveals the diversity and distribution of three invasive Rattus congeners in South Africa. BMC Genet. 12: 1–32, https://doi.org/10.1186/1471-2156-12-26.Search in Google Scholar PubMed PubMed Central

Ben-Faleh, A., Annabi, A., and Said, K. (2012). Morphometric variation in black rat Rattus rattus (Rodentia: Muridae) from Tunisia. Acta Zool. Bulg. 64: 381–387.Search in Google Scholar

Brandt, R.S. and Pessôa, L.M. (1994). Intrapopulation variability in cranial characters of Oryzomys subflavus (Wagner 1842) (Rodentia: Cricetidae) in north-eastern Brazil. Zool. Anz. 233: 45–55.Search in Google Scholar

Chimimba, C.T. and Dippenaar, N.J. (1994). Non-geographic variation in Aethomys chrysopilus (De Winton, 1987) and A. namaquensis (A Smith 1834) (Rodentia: Muridae) from southern Africa. S. Afr. J. Zool. 29: 107–117, https://doi.org/10.1080/02541858.1994.11448335.Search in Google Scholar

Hammer, Ø., Harper, A.T., and Ryan, P.D. (2001). Past: paleaontological statistical software package for education and data analysis. Paleontol. Electron. 4: 1–9.Search in Google Scholar

Hobson, K.A. and Quirk, T. (2014). Effect of age and ration on diet-tissue isotopic (Δ13C, Δ15N) discrimination in striped skunks (Mephitis mephitis). Isot. Environ. Health Stud. 5: 300–306, https://doi.org/10.1080/10256016.2014.867852.Search in Google Scholar PubMed

Jassat, W., Naicker, N., Naidoo, S., and Mathee, A. (2013). Rodent control in urban communities in Johannesburg, South Africa: from research to action. Int. J. Environ. Health Res. 23: 1–10, https://doi.org/10.1080/09603123.2012.755156.Search in Google Scholar PubMed

Julius, R. (2013). Molecular prevalence and diversity of zoonotic bacteria of invasive Rattus from South Africa, with emphasis on the genera Rickettsia and Streprobacillus, MSc Dissertation. Pretoria, University of Pretoria.Search in Google Scholar

Kaiser, H.F. (1970). A second-generation little jiffy. Psychometrika 35: 401–415, https://doi.org/10.1007/bf02291817.Search in Google Scholar

Kajiura, L.J. and Rollo, D.C. (2011). The ontogeny of resource allocation in giant transgenic rat growth hormone mice. Can. J. Zool. 74: 492–507, https://doi.org/10.1139/z96-058.Search in Google Scholar

Kay, H.E. and Hoekstra, H.E. (2008). Rodents. Curr. Biol. 18: 406–410, https://doi.org/10.1016/j.cub.2008.03.019.Search in Google Scholar PubMed

Leamy, L. (1983). Variance partitioning and effects of sex and age on morphometric traits in random bred house mice. J. Mammal. 64: 55–61, https://doi.org/10.2307/1380750.Search in Google Scholar

Lithole, A. (2015). Transmission dynamics of Bartonella in invasive Rattus from South Africa, MSc Dissertation. Pretoria, University of Pretoria.Search in Google Scholar

Long, J. (2003). Introduced mammals of the world: their history, distribution and influence. CSIRO Publishing, Melbourne, pp. 172–199.10.1071/9780643090156Search in Google Scholar

Mairs, S., Swift, B., and Rutty, G.N. (2004). Detergent: an alternative approach to traditional bone cleaning methods for forensic practice. Am. J. Forensic Med. Pathol 25: 276–284, https://doi.org/10.1097/01.paf.0000147320.70639.41.Search in Google Scholar PubMed

Mostert, M.E. (2009). Molecular and morphological assessment of invasive, inland Rattus (Rodentia: Muridae) congenerics in South Africa and their reservoir host potential with respect to Helicobacter and Bartonella, MSc Dissertation. Pretoria, University of Pretoria.Search in Google Scholar

Musser, G.G. and Carleton, M.D (2005). Superfamily Muroidea. In: Wilson, D.E and Reeder, D.M (Eds.), Mammal species of the world: a taxonomic and geographic reference, 3rd ed., Vol. 1. Johns Hopkins Press, Baltimore, pp. 894–1531.Search in Google Scholar

Patton, J.L. and Rogers, M.A. (1983). Systematic implications of non-geographic variation in the spiny rat genus Proechimys (Echymyidae). Zeitschrift für Säugetierkunde 48: 363–370.Search in Google Scholar

Porter, F.H., Costa, F., Rodriguez, G., and Farias, H. (2015). Morphometric and demographic differences between tropical and temperate Norway rats (Rattus norvegicus). J. Mammal. 96: 317–325, https://doi.org/10.1093/jmammal/gyv033.Search in Google Scholar

Prado, J.R. and Percequillo, A.R. (2011). Ontogenetic and sexual variation in cranial characters of Aegialomys xanthaeolus (Thomas, 1894) (Cricetidae: Sigmodontinae) from Ecuador and Peru. Pap. Avulsos Zool. 51: 155–177, https://doi.org/10.1590/s0031-10492011000900001.Search in Google Scholar

Rae, T.C.T., Viðarsdóttir, U.S., Jeffery, N.A., and Steegmann, A.T. (2006). Developmental response to cold stress in cranial morphology of Rattus: implications for the interpretation of climatic adaptation in fossil hominins. Proc. Royal. Soc. B 273: 2605–2610, https://doi.org/10.1098/rspb.2006.3629.Search in Google Scholar PubMed PubMed Central

Ramatla, T., Mputhi, N., Gofaone, K., Taioe, M.O., Thekisoe, O.M.M., and Syakalima, M. (2019). Identification of rodent species that infest poultry houses in Mafikeng, North West Province, South Africa. Int. J. Zool. 1: 1–8, https://doi.org/10.1155/2019/1280578.Search in Google Scholar

R Core Team. (2021). R: a language and environment for statistical computing. R Foundation for Statistical Computing. Vienna. Available at: <https://www.r-project.org/>.Search in Google Scholar

Ringani, G.V., Julius, R.S., Chimimba, C.T., Pirk, C.W.W., and Zengeya, T.A. (2022). Predicting the potential distribution of a previously undetected cryptic invasive synanthropic Asian house rat (Rattus tanezumi) in South Africa. J. Urban Econ. 8: 1–9, https://doi.org/10.1093/jue/juac005.Search in Google Scholar

Rios, E. and Álvarez-Castañeda, S.T. (2007). Environmental responses to altitudinal gradients and subspecific validity in pocket gophers (Thomomys bottae) from Baja California Sur, Mexico. J. Mammal. 88: 926‒934, https://doi.org/10.1644/06-mamm-a-226r1.1.Search in Google Scholar

Schulte-Hofstedde, A.I. (2007). Sexual size dimorphism in rodents. In: Wolff, J.O., and Sherman, P.W. (Eds.), Rodent societies: an ecological and evolutionary perspective. University of Chicago Press, Chicago, pp. 115–128.Search in Google Scholar

Sengupta, P. (2011). A scientific review of age determination for a laboratory rat: how old is it in comparison to human age? BioMed Res. Int. 2: 82–89.Search in Google Scholar

Shine, R. (1978). Sexual size dimorphism and male combat in snakes. Oecologia 33: 269–277, https://doi.org/10.1007/bf00348113.Search in Google Scholar PubMed

Skinner, J.D. and Smithers, R.H.N. (1990). The mammals of the Southern African sub-region. University of Pretoria, Pretoria.Search in Google Scholar

Sneath, P.H.A. and Sokal, R.R. (1973). Numerical taxonomy. Freeman WH and Company, San Francisco, California.Search in Google Scholar

van Rensburg, L.J., Chimimba, C.T., van der Merwe, M., Schoeman, A.S., and Bennett, N.C. (2004). Relative age and reproductive status in Cryptomys hottentotus pretoriae (Rodentia: Bathyergidae) from South Africa. J. Mammal. 85: 1225‒1232, https://doi.org/10.1644/ber-113.1.Search in Google Scholar

Ventura, J. and Lopez-Fuster, M.J. (2000). Morphometric analysis of the black rat Rattus rattus from Congreso island (Chafarinas Archipelago, Spain). Orisis 15: 91–102.Search in Google Scholar

Supplementary Material

The online version of this article offers supplementary material (https://doi.org/10.1515/mammalia-2021-0191).

Received: 2021-12-03
Accepted: 2022-06-02
Published Online: 2022-06-27
Published in Print: 2022-11-25

© 2022 Walter de Gruyter GmbH, Berlin/Boston

Articles in the same Issue

  1. Frontmatter
  2. Obituary
  3. François Catzeflis (1953–2021)
  4. Ecology
  5. Hunting and water scarcity affect habitat occupancy by peccaries (Tayassu pecari and Pecari tajacu) in Calakmul, Mexico
  6. Camera trapping arboreal mammals in Argentina’s Atlantic Forest
  7. Activity patterns and burrowing ecology of the giant pouched rat (Cricetomys emini) in Tshuapa Province, D. R. Congo
  8. Is resource partitioning between two sympatric species of Gracilinanus (Didelphimorphia: Didelphidae) related to trophic and spatial niches?
  9. Gone before it’s known? Camera-trapping shows alarming levels of putative hybrids in the wildcat (Felis silvestris) population of the Gargano National Park (Southern Italy)
  10. A note on bat faunal assessment in Karuna bat cave, Baglung, Nepal
  11. Long-distance dispersal of two species of shrews (Sorex caecutiens Laxmann, 1788 and Sorex minutus Linnaeus, 1766)
  12. Conservation
  13. A new threat for the rarest neotropical carnivore: the vulnerable Colombian weasel, Neogale felipei
  14. Evolutionary Biology
  15. Assessment of craniometric sexual dimorphism and ontogenetic variation in invasive Rattus norvegicus and R. rattus from urban and peri-urban areas of Gauteng Province, South Africa
  16. A geometric morphometric analysis of geographic variation in the Cape Short-eared gerbil, Desmodillus auricularis (Rodentia: Gerbillinae)
  17. Taxonomy/Phylogeny
  18. Cephalic shield morphology as species diagnostic trait and individual natural mark in three-banded armadillos (Tolypeutes; Cingulata: Chlamyphoridae)
  19. Annotations on the taxonomy of the opossums (Didelphimorphia: Didelphidae) of Honduras
  20. Clarifying the taxonomic status of Crocidura cantabra Cabrera, 1908 (Eulipotyphla: Soricidae: Crocidurinae)
https://doi.org/10.1515/mammalia-2021-0191
Keywords for this article
craniometrics; invasive; muridae; ontogenesis; synanthropic

Articles in the same Issue

  1. Frontmatter
  2. Obituary
  3. François Catzeflis (1953–2021)
  4. Ecology
  5. Hunting and water scarcity affect habitat occupancy by peccaries (Tayassu pecari and Pecari tajacu) in Calakmul, Mexico
  6. Camera trapping arboreal mammals in Argentina’s Atlantic Forest
  7. Activity patterns and burrowing ecology of the giant pouched rat (Cricetomys emini) in Tshuapa Province, D. R. Congo
  8. Is resource partitioning between two sympatric species of Gracilinanus (Didelphimorphia: Didelphidae) related to trophic and spatial niches?
  9. Gone before it’s known? Camera-trapping shows alarming levels of putative hybrids in the wildcat (Felis silvestris) population of the Gargano National Park (Southern Italy)
  10. A note on bat faunal assessment in Karuna bat cave, Baglung, Nepal
  11. Long-distance dispersal of two species of shrews (Sorex caecutiens Laxmann, 1788 and Sorex minutus Linnaeus, 1766)
  12. Conservation
  13. A new threat for the rarest neotropical carnivore: the vulnerable Colombian weasel, Neogale felipei
  14. Evolutionary Biology
  15. Assessment of craniometric sexual dimorphism and ontogenetic variation in invasive Rattus norvegicus and R. rattus from urban and peri-urban areas of Gauteng Province, South Africa
  16. A geometric morphometric analysis of geographic variation in the Cape Short-eared gerbil, Desmodillus auricularis (Rodentia: Gerbillinae)
  17. Taxonomy/Phylogeny
  18. Cephalic shield morphology as species diagnostic trait and individual natural mark in three-banded armadillos (Tolypeutes; Cingulata: Chlamyphoridae)
  19. Annotations on the taxonomy of the opossums (Didelphimorphia: Didelphidae) of Honduras
  20. Clarifying the taxonomic status of Crocidura cantabra Cabrera, 1908 (Eulipotyphla: Soricidae: Crocidurinae)
Sign up now to receive a 20% welcome discount
Institutional Access
How does access work?
Have an idea on how to improve our website?
Please write us.
© 2025 De Gruyter Brill
Downloaded on 22.9.2025 from https://www.degruyterbrill.com/document/doi/10.1515/mammalia-2021-0191/html
Scroll to top button