A new species of forest hedgehog (Mesechinus, Erinaceidae, Eulipotyphla, Mammalia) from eastern China

Abstract

The hedgehog genus Mesechinus (Erinaceidae, Eulipotyphla) is currently comprised of four species, M.dauuricus, M.hughi, M.miodon, and M.wangi. Except for M.wangi, which is found in southwestern China, the other three species are mainly distributed in northern China and adjacent Mongolia and Russia. From 2018 to 2023, we collected seven Mesechinus specimens from Anhui and Zhejiang provinces, eastern China. Here, we evaluate the taxonomic and phylogenetic status of these specimens by integrating molecular, morphometric, and karyotypic approaches. Our results indicate that the Anhui and Zhejiang specimens are distinct from the four previously recognized species and are a new species. We formally described it here as Mesechinusorientalissp. nov. It is the only Mesechinus species occurring in eastern China and is geographically distant from all known congeners. Morphologically, the new species is most similar to M.hughi, but it is distinguishable from that species by the combination of its smaller size, shorter spines, and several cranial characteristics. Mesechinusorientalis sp. nov. is a sister to the lineage composed of M.hughi and M.wangi from which it diverged approximately 1.10 Ma.

Introduction

In recent years, interest in the faunal inventory of insectivorous mammals in different countries has increased (Kryštufek and Motokawa 2018; Andreychev and Kuznetsov 2020). The forest hedgehog genus Mesechinus Ognev, 1951 is one of five extant genera in the subfamily Erinaceinae. Mesechinus was previously regarded as a subgenus of Erinaceus Linnaeus, 1758 or Hemiechinus Fitzinger, 1866 (Pavlinov and Rossolimo 1987; Corbet 1988; Bannikova et al. 2002). Frost et al. (1991) promoted it to full-genus status, a conclusion supported by analysis of morphological characters (Gould 1995) and chromosomal data (Korablev et al. 1996). The most distinctive morphological character distinguishing Mesechinus from Erinaceus and Hemiechinus is the unique shape of the suprameatal fossa; the lateral border of this fossa is somewhat U-shaped in Mesechinus but C-shaped in the other erinaceine genera (Frost et al. 1991).

Currently, four species are recognized in the genus, including M.dauuricus (Sundevall, 1842), M.hughi (Thomas, 1908), M.miodon (Thomas, 1908), and M.wangi He, Jiang & Ai, 2018 (Wilson and Mittermeier 2018). Mesechinus species mainly occur in northern China and adjacent Mongolia and Russia, with an isolated species (M.wangi) on Mount Gaoligong, Yunnan, southwestern China (Frost et al. 1991; Ai et al. 2018). Mesechinusdauuricus and M.hughi mainly inhabit semidesert habitats, including cold-temperate deciduous and temperate deserts, warm-temperate deserts, grasslands and deciduous broad-leaf forests, M.miodon mainly inhabits semiarid and dry steppe habitats and subalpine and low-elevation coniferous forests, and only M.wangi inhabits tropical or subtropical rainforest (Ai et al. 2018; Wilson and Mittermeier 2018).

Hugh’s Hedgehog (M.hughi) is the smallest species of Mesechinus and is mainly distributed in southern Shaanxi, southern Shanxi, and northern Sichuan in China (Ai et al. 2018) (Fig. 1). This dark-coloured hedgehog with no all-white spines was first described by Thomas (1908) based on specimens from Paochi (= Baoji), Shaanxi, China. Chen et al. (2020) reported the first record of M.hughi in eastern China based on a specimen collected from Xuancheng, Anhui Province. They pointed out that Anhui Mesechinus specimen was genetically distant from Shaanxi specimens (4.9–5.3% of CYT B gene) and may have undergone isolated differentiation (Chen et al. 2020). Recently, we obtained a CYT B sequence of M.wangi. Our preliminary phylogenetic analysis revealed that the Anhui specimens form a lineage sister to the lineage composed of M.hughi from Shaanxi and M.wangi, which suggests that additional studies with more specimens were necessary to confirm the taxonomic status of the Anhui Mesechinus.

Figure 1.

Distribution of the genus Mesechinus.

From 2022 to 2023, we collected six specimens of Mesechinus from Anhui and Zhejiang provinces, eastern China (Fig. 2). Our morphological and molecular results reveal the eastern China samples differ from M.hughi and other known Mesechinus species. We recognize it as a new species, Mesechinusorientalis sp. nov., which we describe here.

Figure 2.

A living Mesechinusorientalis sp. nov. (XC 2205003) from Xuancheng, Anhui.

Materials and methods

Sampling

Seven Mesechinusorientalis sp. nov. specimens, including a specimen collected by Chen et al. (2020), were obtained from Anhui and Zhejiang provinces in eastern China (Suppl. material 1). Our specimens were euthanized, and liver or muscle tissues were extracted and preserved in pure ethanol. Skulls were extracted and cleaned. All the specimens and tissues were deposited at Anhui Normal University (AHNU). Animals were handled consistent with the Guidelines of the American Society of Mammalogists (Sikes 2016).

Morphological measurement and analysis

Five external measurements, including weight (W), head–body length (HB), tail length (TL), hind-foot length (HF), and ear length (EL) of M.orientalis sp. nov. were measured in the field to the nearest 1 g or 1 mm. Twelve craniodental variables were measured using digital calipers graduated to 0.01 mm following Pan et al. (2007) and Ai et al. (2018): greatest length of the skull (GLS), condylobasal length (CBL), basal length (BL), cranial height (CH), palatal length (PL), zygomatic breadth (ZMB), interorbital breadth (IOB), mastoid width (MTW), greatest width of nasal (GWN), breadth of first upper molar (BM1), length of upper tooth row (LUTR), and length of below tooth row (LBTR). Comparative morphological data of other Mesechinus species were obtained from Ai et al. (2018), which included 4 M.wangi, 18 M.miodon, 31 M.hughi, and 13 M.dauuricus.

Thirty-seven complete skulls were used for PCA, including specimens of 3 M.wangi, 20 M.hughi, 6 M.miodon, 1 M.dauuricus, and 7 M.orientalis sp. nov. Morphometric variation was analyzed using a principal component analysis (PCA) in SPSS 19.0 based on 12 log₁₀-transformed cranial measurements. To further confirm the validity of the potential new species, we coded the characters of Mesechinus species according to Gould (1995). In this procedure, we systematically compared the morphological characteristics of the new species with other Mesechinus species, especially the most morphologically similar species, M.hughi.

Mitogenome sequencing, assembly, and annotation

We used next-generation sequencing (NGS) to obtain the complete mitochondrial genome of M.orientalis sp. nov. Illumina high-throughput sequencing platform was employed for sequencing with a strategy of 150 paired-ends, and the quality was checked using FastQC (de Sena Brandine and Smith 2021). The mitochondrial genome assembly was performed using NOVOPlasty (Dierckxsens et al. 2017).

The mitochondrial genome was annotated using MitoZ in the MITOS WebServer with analytical parameters set using the vertebrate genetic code (Bernt et al. 2013; Meng et al. 2019). Geneious v. 9.0.2 (Kearse et al. 2012) was used to examine all mitochondrial genes. The obtained sequences were edited and aligned with MEGA v. 11(Tamura et al. 2021). The newly obtained mitogenome has been deposited in GenBank (accession number OR774964).

Phylogenetic analysis

The mitochondrial genomes of four other Mesechinus species, and six erinaceid species, including representatives of Paraechinus Trouessart, 1879, Hemiechinus, and Atelerix Pomel, 1848, were downloaded from GenBank and included in our analyses. Mitochondrial genomes of Neotetracussinensis Trouessart, 1909 and Hylomyssuillus Müller, 1840, also obtained from GenBank, were used as the outgroup (Table 1). The phylogenetic analyses were conducted using the two rRNA and 12 coding genes on the heavy chain and excluded ND6 on the light chain. Each gene was aligned using MUSCLE and then checked manually.

Table 1.

Samples used for molecular phylogenetic analysis in this study.

Subfamily	Species	Museum code	Collection localities	GenBank no.
Galericinae	Hylomyssuillus		Java, Indonesia	AM905041
	Neotetracussinensis		Pingshan, Yibin, Sichuan, China	NC_019626
Erinaceinae	Paraechinusmicropus	USNM369316		OP654708
	Hemiechinusauritus			AB099481
	Atelerixalbiventris	USNM325883		OP654703
	Erinaceusamurensis		Gongwon, Korea	KX964606
	Mesechinusmiodon		Yulin, Shaanxi, China	KT824773
	M.dauuricus	KIZ200908002		OP654710
	M.wangi	GLGS0907001		OP654712
	M.hughi	KIZ200908004		OP654727
	M.orientalis sp. nov.	XC 2205003	Xuancheng, Anhui, China	OR774964

To reconstruct the phylogenetic relationships, maximum-likelihood (ML) and Bayesian-inference (BI) analyses were performed in IQ-TREE and MrBayes, respectively, in PhyloSuite (Zhang et al. 2020). The phylogenetic tree was visualized and annotated in tvBOT (Xie et al. 2023). The best-fit partitioning schemes were estimated based on the Bayesian Information Criterion (BIC) using PartitionFinder 2 (Lanfear et al. 2017).

Divergence time estimation

BEAST v. 2.6 (Bouckaert et al. 2014) was used to estimate divergence times in the CIPRES Science Gateway (Miller et al. 2015). The data were partitioned according to the results of PartitionFinder 2 (Suppl. material 2). We used the unlinked site model and linked clock model and time tree across partitions, and the relaxed lognormal clock model and a birth-death model for the tree prior. Two secondary calibrations were used: (1) the most recent common ancestor of the subfamilies Galericine and Erinaceinae, which was estimated at ca 28.3–48.8 Ma (Meredith et al. 2011) using a lognormal distribution prior (mean = 3.61, SD = 0.142, offset = 0); and (2) The most recent common ancestor of Erinaceinae, which was estimated at ca 6.97 Ma (He et al. 2021), with a normal distribution prior (mean = 6.97, sigma = 2.05, offset = 0). The analyses were conducted twice, each for 100 million generations, sampling every 10000 generations. The posterior distribution of each parameter in the log file was estimated using Tracer v. 1.7 (Rambaut et al. 2018) to ensure that the effective sampling size of all parameters was greater than 200. For all BEAST analyses, we compiled time trees with TreeAnnotator v. 2.6 (Bouckaert et al. 2014) and excluded 10% of each chain as burn-in. The generated tree was viewed in FigTree v. 1.4 (Rambaut 2017) and beautified in tvBOT (Xie et al. 2023).

Cell culture and karyotype analysis

A female individual of M.orientalis sp. nov. (XC 2205003) collected in May 2022 was used for cell cultures. Standard procedures were applied for fibroblast culture, chromosome preparation, and G-banding. Two fibroblast cell lines derived from M.orientalis sp. nov. (XC 2205003) were established and deposited in the Kunming Cell Bank, Yunnan, China. A CytoVision system (Applied Imaging Co., USA) with a CCD camera mounted on a Zeiss microscope (Germany) was used to karyotype analysis. Chromosomes of M.orientalis sp. nov. (XC 2205003) were numbered according to M.wangi (Ai et al. 2018).

Results

Morphological analyses

Summaries of external morphology and craniodental measurements are given in Table 2. According to the measurements, Mesechinusorientalis sp. nov. (HB = 188.83 mm ± 8.13; GLS = 49.95 mm ± 1.69) is similar in size to M.hughi (HB = 189.71 mm ± 23.80; GLS = 49.39 mm ± 1.54) but much smaller than M.wangi, M.dauuricus, and M.miodon (Table 2).

Table 2.

External and cranial measurements (mm) of Mesechinus specimens examined; mean ± S), range for each measurement, and number of specimens measured (n) are given.

	M.orientalis sp. nov.	M.hughi	M.dauuricus	M.miodon	M.wangi
	n = 7	n = 31	n = 13	n = 18	n = 4
W	339 ± 52.97	341 ± 125.75	562 ± 124.31	505 ± 154.03	401 ± 43.27
	299–414; 3	112–750; 31	423–840; 11	230–750; 6	336–449; 4
HB	188.83 ± 8.13	189.71 ± 23.80	373.91 ± 21.35	205 ± 23.53	208.75 ± 21.90
	176–198; 6	148–232; 31	175–261; 12	120–220; 17	180–140; 4
TL	23.50 ± 3.77	19.23 ± 3.26	24.08 ± 3.50	33.22 ± 5.07	17.08 ± 1.78
	16–27; 6	12–24; 27	17–30; 12	25–43; 17	14–18; 4
HF	36.75 ± 3.19	37.97 ± 4.29	34.74 ± 7.08	58.80 ± 82.43	47.00 ± 1.12
	31–40; 6	30–47; 31	18–41; 12	35–38; 16	45–48; 4
EL	26.00 ± 2.66	22.94 ± 3.93	31.19 ± 3.28	28.81 ± 3.03	30.00 ± 1.49
	23–30; 6	16–33; 31	22–34; 11	24–35; 17	28–31; 4
GLS	49.95 ± 1.69	49.39 ± 1.54	55.18 ± 3.07	54.10 ± 2.10	54.75 ± 0.70
	47.64–51.76; 7	45.10–52.40; 23	50.20–58.40; 12	49.30–57.20; 14	53.70–55.60; 4
CBL	49.49 ± 1.64	48.46 ± 1.58	54.72 ± 2.83	53.18 ± 2.35	54.55 ± 0.59
	47.27–51.42; 7	44.40–51.20; 23	49.40–57.40; 13	48.50–56.30; 11	53.60–55.20; 4
CH	15.42 ± 0.54	16.14 ± 0.95	17.76 ± 2.00	18.67 ± 0.66	17.13 ± 0.60
	14.46–16.39; 7	14.90–18.20; 21	17.20–19.10; 9	17.80–19.70; 6	16.10–17.60; 4
BL	46.66 ± 1.45	45.55 ± 1.29	51.83 ± 1.94	49.64 ± 2.04	50.00 ± 1.37
	44.47–48.28; 7	43.20–48.80; 21	48.10–54.50; 13	44.70–52.30; 14	47.70–51.30; 4
PL	27.46 ± 0.77	26.58 ± 0.62	28.60; 1	28.82 ± 1.41	30.25 ± 0.50
	26.17–28.52; 7	25.70–28.40; 21		27.00–32.18; 14	29.50–30.80; 4
ZMB	29.62 ± 1.51	28.90 ± 1.68	32.62 ± 2.82	32.77 ± 2.09	33.97 ± 0.19
	27.78–31.41; 7	25.70–32.00; 22	28.40–36.40; 13	28.70–37.08; 14	33.70–34.10; 3
IOB	12.29 ± 0.43	12.51 ± 0.50	13.86 ± 0.68	13.87 ± 0.76	14.68 ± 0.33
	11.51–12.95; 7	11.70–13.60; 23	13.00–15.10; 9	12.90–15.10; 6	14.20–15.10; 4
MTW	24.68 ± 1.00	21.67 ± 1.57	25.58; 1	25.93 ± 1.18	25.60 ± 0.64
	23.66–26.38; 7	19.50–24.50; 21		24.30–28.30; 14	24.70–26.20; 4
GWN	3.07 ± 0.29	2.97 ± 0.29	2.96; 1	2.70 ± 0.21	4.30 ± 0.00
	2.70–3.51; 7	2.60–3.60; 23		2.37–2.94; 6	4.30–4.30; 3
BM1	19.54 ± 0.64	17.38 ± 0.75	20.20; 1	21.08 ± 0.66	21.43 ± 0.25
	19.20–20.27; 7	16.50–19.50; 21		20.30–22.30; 14	21.10–21.70; 3
LUTR	25.27 ± 0.51	24.65 ± 1.12	27.85 ± 1.25	27.25 ± 0.99	27.90 ± 1.02
	24.45–25.89; 7	21.40–26.10; 23	25.00–29; 13	25.70–29.02;14	26.70–29.10; 4
LBTR	22.32 ± 1.02	21.19 ± 0.78	24.30; 1	24.91 ± 0.70	24.85 ± 0.44
	21.31–24.16; 7	20.20–23.70; 21		23.40–25.70;14	24.20–25.30; 4

The first two PCA axes had eigenvalues exceeding 1.0 (Table 3). The first principal component (PC1) accounted for 69.32% of the total variance and was positively correlated with all variables (loading > 0.69), reflecting a size effect. The second principal component (PC2) accounted for 10.01% of the variance and was strongly positively correlated with GWN, MTW, and BM1 (loading > 0.53). The PC1 vs PC2 scatter plot (Fig. 3) showed M.orientalis sp. nov. slightly overlapping with M.hughi but well separated from M.wangi, M.dauuricus, and M.miodon. Specimens of M.orientalis sp. nov. and M.hughi mainly occupy the negative regions of PC1, reflecting their smaller size compared to M.wangi and M.miodon. Mesechinusorientalis sp. nov. plotted on the positive regions of PC2, whereas most M.hughi specimens plotted on the negative regions, suggesting the new species has wider nasal, mastoid, and M¹ than M.hughi.

Table 3.

Factor loading eigenvalues and percentage of variance explained for PC1 and PC2 from principal component analysis.

Variables	Component
	1	2
BL	0.964	−0.107
CBL	0.956	−0.143
GLS	0.944	−0.208
LUTR	0.922	−0.090
PL	0.910	0.165
LBTR	0.895	0.100
BM1	0.874	0.251
ZMB	0.832	0.020
IOB	0.791	−0.022
MTW	0.745	0.306
GWN	0.308	0.715
CH	0.591	−0.642
Eigenvalues	8.294	1.210
Variance explained (%)	69.116	10.081

Figure 3.

Plot of Mesechinusorientalis sp. nov. for PCA 1 and PCA 2.

Morphological characteristic matrix

The morphological characteristics matrix is summarized in Suppl. material 3, and the specific characters represented by each number are interpreted in Suppl. material 4. Based on the matrix, M.orientalis sp. nov. differs from the most morphologically similar species, M.hughi in several characteristics: (1) the parietal is relatively higher than the frontals (frontals more elevated than parietals in M.hughi; character 32 in Suppl. material 3); (2) the posterior palatal spine is vestigial (vs spine is well developed in M.hughi; character 25); (3) suprameatal fossa is moderately developed (vs shallow in M.hughi; character 30); (4) P² is two-rooted and not completely fused (Fig. 4) (the single root or two roots of P² are well fused in M.hughi; character 73); (5) P³ is small because of a vestigial protocone (vs larger and with protocone well developed in M.hughi; character 82).

Figure 4.

Ventral and lateral views of left upper toothrow of M.orientalis sp. nov. A, C, B, DM.hughi. The arrows point at the root of P² (A, B), and the protocone of P³ (C, D).

Phylogenetic relationships

The ML and BI trees showed identical topologies, and only the BI tree is shown (Fig. 5). Relationships among all Mesechinus species were strongly supported (PP = 1.00). Mesechinusorientalis sp. nov. is strongly supported as embedded within the Mesechinus clade (PP = 1.00). Among the Mesechinus species, the new species forms a sister relationship (PP = 1.00) to the M.hughi + M.wangi clade, whose sister relationship was also strongly supported (PP = 1.00).

Figure 5.

Mitochondrial gene tree of the genus Mesechinus and other genera of Erinaceinae and the outgroup. Branch lengths represent substitutions per site. Numbers above branches are posterior probability supporting the relationship.

Divergence times

Divergence time estimates show that the most recent common ancestor of Mesechinus occurred in the early Pleistocene, ca 1.71 Ma (95% CI = 1.23–2.24 Ma) (Fig. 6). Mesechinusorientalis sp. nov. diverged from the M.hughi + M.wangi ancestor ca 1.10 Ma (95% CI = 0.78–1.47 Ma), with M.hughi and M.wangi diverging ca 0.74 Ma (95% CI = 0.50–1.02 Ma).

Figure 6.

Divergence times estimated using BEAST based on mitogenome data. Branch lengths represent time. Numbers above branches refer to divergence time in millions of years. Asterisks indicate the location of correction points.

Karyotypic characteristics of Mesechinusorientalis sp. nov.

The comparison of the G-banding chromosomes of M.orientalis sp. nov. and M.wangi is shown in Fig. 7. Since the specimen was a female individual, its X chromosome was determined by comparing its karyotype with that of M.wangi. The two species’ autosomes are almost identical; the diploid number (2n) and autosomal fundamental number (FNa) are 48 and 92, respectively. The only difference was the X chromosome, where that of M.wangi was meta centric, while that of M.orientalis sp. nov. was submetacentric.

Figure 7.

Comparison of karyotypes in two Mesechinus species AM.orientalis sp. nov. (XC 2205003) BM.wangi.

Based on the morphological, morphometric, and molecular evidence and the modern phylogenetic species concept (phylogenetic species concept based on both diagnosability and monophyly as operational criteria) (Mayden 1997; Gutiérrez and Garbino 2018), we recognize the Mesechinus population from Anhui as a new species and formally describe it below.

Taxonomic account

. Mesechinus orientalis sp. nov.

D0FB49BA-C64C-5E96-9B69-C0640308DDCA

https://zoobank.org/BB3A29EC-F0A8-4DFD-A954-D5AC8E03B4B2

Suggested common name.

Eastern Forest Hedgehog, 华东林猬 (Huadong Linwei).

Type materials.

Holotype: XC 23001, an adult male collected from Xikou Town (30°34'42"N, 118°41'47"E), Xuancheng City, southern Anhui, China, Zifan Shi leg., May 2023. The dried skin, cleaned skull, and tissue samples are deposited in AHNU. Paratypes: XC 18001, XC 2205001, XC 2205003, XC 2205005, XC 2205006, HZ 22001, six adult specimens collected from southeast Anhui and northwest Zhejiang, China, between 2018 and 2023. The specimens are deposited in AHNU.

Etymology.

The specific name orientalis is derived from the Latin oriens, “the east”, and suffix -alis, “pertaining to”, in reference to the new species’ eastern distribution in Anhui and Zhejiang provinces in eastern China.

Diagnosis.

This is a small-bodied hedgehog (GLS = 49.95 ± 1.69 mm), similar to M.hughi, but smaller than other Mesechinus species. It has the shortest spines in the genus (18–20 mm); the spines have four-colour rings, similar to the spines of M.dauuricus and M.hughi, but different from those of M.miodon and M.wangi (Fig. 8). The parietal is higher than the frontals, which differs from that of M.hughi and M.wangi (Fig. 9). The P² is two-rooted and not completely fused (Fig. 4). The protocone of P³ is vestigial, which differs from that of M.hughi, and smaller than P², which distinguishes it from M.dauuricus. The dental formula of M.orientalis sp. nov. [I 3/2, C1/1, P 3/2, M 3/3 (×2) = 36], which distinguishes it from M.wangi.

Figure 8.

External morphs and spines of Mesechinusorientalis sp. nov. A male specimen (XC 2205001) B female specimen (XC 2205005).

Figure 9.

Dorsal ventral and lateral views of skull and mandible of Mesechinus species AM.orientalis sp. nov. (holotype XC 23001) BM.hughi (KIZ 027029) CM.dauuricus (KIZ 027005) DM.miodon (holotype BM 9.1.1.9) EM.wangi (holotype KIZ 022028).

Description.

This is a small-bodied Mesechinus species (HB = 188.83 mm; GLS = 49.95 mm) (Table 2). The ears are small and nearly the same length as the surrounding spines (Fig. 8). The nose is brown, with black whiskers on the snout; these whiskers shorten towards the nose. The spines are the shortest (18–20 mm) among Mesechinus species. There are four colour rings on the spine from the base to the tip: two-thirds of the length is white at the base, followed by a 3–4 mm black ring, a narrow light ring, and a black tip (Fig. 8). This species appears to be sexually dimorphic; among the specimens we collected, the pelage of males was generally grey, while that of most of the females (2 of 3 specimens) was reddish brown. However, this is a relatively small sample size, and further investigation is required to establish sexual dimorphism with more certainty.

The skull is heavy and with a shortened rostrum, and the lambdoidal crest is evident. The parietal is relatively higher than the frontal (Fig. 9). On the ventral side of the skull, a posterior palatal shelf and vestigial posterior palatal spine (<1 mm) extend slightly posteriorly. The suprameatal fossa is moderately developed and has the anterior and posterior rim nearly parallel, giving the fossa a more angular or U-shaped appearance. The zygomatic arches are significantly expanded, and the temporal fossa is large and subrounded. The middle palatine foramen is relatively larger than in other Mesechinus species. The coronoid process of the mandible is tall, rising upward from the posterior of the toothrow; the tips are slightly curved to the posterior, and the posterior surface is concave (Fig. 9). The mandibular condyle sits posteriorly below the coronoid process at a nearly 45° angle. The angular process is enlarged, thick, long, and triangulate.

As with other Mesechinus species, except M.wangi which has an additional M⁴, the dental formula of the new species is I 3/2, C1/1, P 3/2, M 3/3 (×2) = 36. The I¹ is enlarged, I² is much smaller than I¹ and I³, and I³ has two roots. P² also has two roots which are not completely fused. P³ is small (smaller than P²) and has a vestigial protocone. M¹ is slightly larger than M², and M³ is reduced.

Comparison.

The hedgehogs from China’s Anhui and Zhejiang provinces can be easily classified as Mesechinus based on the following morphological characteristics: the absence of pure white spines; relatively small ears, almost similar in length to the surrounding spines; no bare part on the forehead nor at the top of the forehead which divides the spines on the head into two halves; and a U-shaped suprameatal fossa.

Among the Mesechinus species, M.orientalis sp. nov. is morphologically most similar to M.hughi. However, the new species can be distinguished by many characters. Mesechinusorientalis sp. nov. has the shortest spines in the genus (18–20 mm), shorter than those in M.hughi (22–24 mm). The parietal is relatively higher than the frontals in the new species, whereas the frontals are relatively higher than parietals in M.hughi. P² has two roots which are not completely fused in M.orientalis sp. nov., while in M.hughi P² the two roots are well fused. The P³ protocone is vestigial in the new species but well developed in M.hughi. The posterior palatal spine is vestigial, and the suprameatal fossa is moderately developed in the new species, which differs from the well-developed posterior palatal spine and shallow suprameatal fossa in M.hughi. In addition, the MTW and BM¹ of the new species are significantly greater than those of M.hughi (P < 0.01).

Mesechinusorientalis sp. nov. (HB = 188.83 mm ± 8.13; GLS = 49.95 mm ± 1.69) is distinguishable from M.dauuricus (HB = 373.91 mm ± 21.35; GLS = 55.18 mm ± 3.07), M.miodon (HB = 205 mm ± 23.53; GLS = 54.10 mm ± 2.10), and M.wangi (HB = 208.75 mm ± 21.90; GLS = 54.75 mm ± 0.70) by its smaller size. The spines of the new species are much shorter (18–20 mm) than those of M.dauuricus (21–24 mm), M.miodon (~26 mm), and M.wangi (21–24 mm). The spines of M.orientalis sp. nov. have four-colour rings similar to those of M.dauuricus and M.hughi, but they differ from M.miodon and M.wangi. P³ of the new species is much smaller than P², which differs from M.dauuricus, in which P³ is of equal size to P². The parietal is relatively higher than the frontals in M.orientalis sp. nov., which differs from M.wangi. Additionally, the presence of M ⁴ in M.wangi is unique in the genus, which easily distinguishes it from other species.

Distribution and habitat.

Mesechinusorientalis sp. nov. is currently known from southern Anhui (Xuancheng and Huangshan) and northwestern Zhejiang (Anji, Changxing, Deqing, Yuhang, Linan, Chunan), both in eastern China. Most specimens were collected in scrubland and subtropical broad-leaf evergreen forests at elevations from 30 to 700 m a.s.l.

Discussion

For a long time, the genus Mesechinus was thought to be restricted to northern China and adjacent Mongolia and Russia (Wilson and Reeder 2005) until Ai (2007) reported a small population of Mesechinus from Mount Gaoligong in Yunnan Province, southwestern China. This Mount Gaoligong population was subsequently described as a new species, M.wangi (Ai et al. 2018). In this study, we recognized a population of Mesechinus from eastern China as another isolated species, M.orientalis sp. nov. Morphologically, the new species is most similar to M.hughi, but it is distinguishable from all recognized Mesechinus species in having the shortest spines, an incompletely fused two-rooted P², and a smaller, vestigial P³. The new species is geographically isolated from its congeners by at least 1000 km, and it is currently the southeasternmost species of Mesechinus (Fig. 1). Except for M.orientalis sp. nov., only one species of hedgehog, Erinaceusamurensis Schrenk, 1859, occurs in eastern China. While these species are sympatric, at least in Xuancheng, Anhui Province, M.orientalis sp. nov. can easily be distinguished from E.amurensis by the absence of pure-white spines, and no bare part on the forehead nor at the top of the forehead which divides the spines on the head into two halves (Fig. 8).

The discovery of a new species of Mesechinus in eastern China has greatly expanded the known range of the genus and is vital in understanding the macroevolution of the genus. The oldest fossils of Mesechinus are from the Early Pleistocene near Taijiaping Village in Nangaoya Township, Tianzhen, Shanxi Province (Bai et al. 2022). Our molecular results reveal that the divergences among M.orientalis, M.wangi, and M.hughi occurred in the Middle Pleistocene 0.74–1.10 Ma (Fig. 6). Increased cooling and aridification during the middle Pleistocene (known as the middle Pleistocene transition at ca 1.2–0.5 Ma) appear to have been critically important in the split of the three species, which may also have facilitated the migration of the ancestors of M.wangi and M.orientalis sp. nov. to southwestern and southeastern China, respectively. The north–south trending Dabie Mountains, which are located between the Qinling Mountains and southern Anhui, may have provided a migration route for the ancestor of M.orientalis sp. nov. to reach southern Anhui. The mountainous area of southern Anhui and northwestern Zhejiang Province also likely acted as glacial refugia in the Pleistocene for the new species.

Citation

Shi Z, Yao H, He K, Bai W, Zhou J, Fan J, Su W, Nie W, Yang S, Onditi KO, Jiang X, Chen Z (2023) A new species of forest hedgehog (Mesechinus, Erinaceidae, Eulipotyphla, Mammalia) from eastern China. ZooKeys 1185: 143–161. https://doi.org/10.3897/zookeys.1185.111615

Funding Statement

The study was supported by the National Natural Science Foundation of China (No. 31900318, 32170452), the Science Foundation of Hebei Normal University (No. L2023B54), the Anhui Provincial Natural Science Foundation (2008085QC106), and the University Synergy Innovation Program of Anhui province (GXXT-2020-075).

Additional information

Conflict of interest

The authors have declared that no competing interests exist.

Ethical statement

Animals were handled complying with the animal care and use guidelines of the American Society of Mammologists , and following the guidelines and regulations approved by the internal review board of Anhui Normal University.

Funding

The study was supported by the National Natural Science Foundation of China (no. 31900318, 32170452), the Science Foundation of Hebei Normal University (No. L2023B54), the Anhui Provincial Natural Science Foundation (2008085QC106), and the University Synergy Innovation Program of Anhui province (GXXT-2020-075).

Author contributions

Conceptualization: ZC, KH. Data curation: ZS, HY, JF. Funding acquisition: ZC, KH, WB. Investigation: ZS, HY, JZ, WN, WS. Project administration: ZC. Resources: ZC, ZS, JZ, SY XJ. Supervision: ZC, KH, XJ. Visualization: ZS, WS, WN. Writing - original draft: ZS, ZC. Writing - review and editing: ZC, ZS, XJ, KOO, KH, WB.

Author ORCIDs

Zifan Shi https://orcid.org/0009-0008-5165-6604

Kai He https://orcid.org/0000-0002-6234-2589

Jiajun Zhou https://orcid.org/0000-0003-1038-1540

Kenneth O. Onditi https://orcid.org/0000-0003-4034-6818

Xuelong Jiang https://orcid.org/0000-0003-2052-2490

Zhongzheng Chen https://orcid.org/0000-0003-3821-0145

Data availability

All of the data that support the findings of this study are available in the main text or Supplementary Information.

Supplementary materials

Supplementary material 1

Collection information of Mesechinusorientalis sp. nov.

This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.

Zifan Shi, Hongfeng Yao, Kai He, Weipeng Bai, Jiajun Zhou, Jingyi Fan, Weiting Su, Wenhui Nie, Shuzhen Yang, Kenneth O. Onditi, Xuelong Jiang, Zhongzheng Chen

Data type

xlsx

Supplementary material 2

Partitioning schemes used in mitogenome RAxML analyses

This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.

Author: Zifan Shi

Data type

xlsx

Supplementary material 3

Morphological characteristic matrix

This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.

Zifan Shi, Hongfeng Yao, Kai He, Weipeng Bai, Jiajun Zhou, Jingyi Fan, Weiting Su, Wenhui Nie, Shuzhen Yang, Kenneth O. Onditi, Xuelong Jiang, Zhongzheng Chen

Data type

xlsx

Explanation note

Matrix of morphological characters of erinaceid species. The specific characters represented by each number are interpreted in Suppl. material 4.

Supplementary material 4

Morphological transformation series

This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.

Zifan Shi, Hongfeng Yao, Kai He, Weipeng Bai, Jiajun Zhou, Jingyi Fan, Weiting Su, Wenhui Nie, Shuzhen Yang, Kenneth O. Onditi, Xuelong Jiang, Zhongzheng Chen

Data type

doc