Apodemus Division. Diagnosed by Niethammer (1978d) using skeletal and soft tissues, body size, and dental traits, and by Martín Suárez and Mein (1998) using dental characters only. Recognized species have been allocated among the subgenera Apodemus, Sylvaemus, Alsomys, and Karstomys (Corbet, 1978c; Zimmermann, 1962), but whether these names designate monophyletic clusters and should be retained as subgenera or instead raised to generic rank remains to be answered by critical systematic revision of the entire group, which is currently unavailable. Most taxonomists appreciate at least the subgeneric validity of Apodemus and Sylvaemus; some suggest Sylvaemus should be raised to generic rank because of its great morphological and genetic divergence from Apodemus (Britton-Davidian et al., 1991; Mezhzherin and Zykov, 1991); others either treat Sylvaemus as a separate genus or suggest it should be recognized at that rank (Bonhomme et al., 1985; Chelomina, 1998; Chelomina et al., 1998b; Filippucci et al., 1996; Mezhzherin, 1997a; Pavlinov and Rossolimo, 1998; Pavlinov et al., 1995a; Zagorodnyuk, 1992b, 1993; Zagorodnyuk et al., 1997); and even recognition of Apodemus, Alsomys, and Sylvaemus as genera (with Karstomys as a subgenus of the latter) has been advocated (Mezhzherin, 1997a). Pavlinov et al. (1995a) and Pavlinov and Rossolimo (1998) listed Alsomys as a subgenus of Apodemus, and Karstomys as a subgenus of genus Sylvaemus. Phylogenetic analysis of complete mtDNA cytochrome b sequences identified A. agrarius, A. mystacinus, A. uralensis, A. flavicollis, A. sylvaticus, and A. alpicola as a monophyletic clade (Martin et al., 2000); the latter four form a monophyletic subgroup (subgenus Sylvaemus), and A. agrarius (subgenus Apodemus) and A. mystacinus (subgenus Karstomys) fall outside that cluster, a pattern taxonomically interpreted by employing three subgenera (Martin et al., 2000). But Martin et al. also noted that the genetic distances of species in subgenus Sylvaemus compared to A. agrarius and A. mystacinus was of the same magnitude as the Sylvaemus cluster was to other murine genera sampled (Micromys, Mus, Rattus, and Acomys). Results from analyses of combined nuclear IRBP and mtDNA cytochrome b and 12S rRNA sequences indicated Apodemus to be monophyletic and consisting of Apodemus (A. semotus, A. peninsulae, and A. agrarius) and Sylvaemus (A. sylvaticus, A. alpicola, and A. flavicollis) groups, with A. mystacinus being related to Sylvaemus, although with weak support and possibly forming a third group, Karstomys (Michaux et al., 2002a). Using morphological and published genetic information, Musser et al. (1996) defined three groups into which the species discussed in the accounts that follow are allocated: Apodemus Group (A. agrarius, A. chevrieri, A. speciosus, A. peninsulae, A. latronum, A. draco, and A. semotus); Sylvaemus Group (A. sylvaticus, A. flavicollis, A. uralensis, A. mystacinus, A. epimelas, A. alpicola, A. witherbyi, A. hyrcanicus, A. ponticus, A. rusiges, and A. pallipes); and A. argenteus as the sole member of the third group. The partitioning has generally been corroborated by analyses of protein electrophoretic results (Filippucci et al., 2002) and mitochondrial and nuclear gene sequences (Chelomina, 1998; Chelomina et al., 1998b; Liu et al., 2004; Michaux et al., 2002a; Serizawa et al., 2000). Liu et al. (2004) and Serizawa et al. (2000) diverged from this phyletic trichotomy by isolating A. gurkha as the sole member of a fourth group, which we recognize here, along with the other three groups, and at the same time abandon subgeneric designations. These studies underline the cladistic diversity within what is now called Apodemus, identify at least four monophyletic groups, and undermine past efforts to cram all species into either Apodemus or Sylvaemus, which besides being a futile endeavor does not reflect phylogenetic reality.
Examples of morphological, chromosomal, and molecular inquires bearing on systematics of Apodemus are (also consult those references cited in species accounts): phallic comparisons among European (Williams et al., 1980) and Chinese (Liu et al., 2000; Yang and Fang, 1988) species; taxonomic differences in testes size among European species (Kratochvíl, 1971); comparative chromosomal studies among European (Bekasova et al., 1980; Soldatovic et al., 1975; Vujosevic et al., 1984; Zima et al., 1997a) and Japanese (Tsuchiya, 1981) species; presence or absence of B chromosomes and their possible biological and taxonomic significance (Zima and Macholán, 1995); electrophoretic variations of allozymes among a variety of species in systematic context (Darviche et al., 1979, and references therein; Filippucci, 1992; Filippucci et al., 2002; Fraguedakis-Tsolis et al., 1983; Gemmeke, 1980; Gill et al., 1987; Macholán et al., 2001b; Mezhzherin et al., 1992; Verimli et al., 2001); electrophoretic, karyological, and morphological distinctions among several species (Britton-Davidian et al., 1991; Mezhzherin, 1997a; Vorontsov et al., 1989); differences in restriction endonuclease of nuclear DNA used to assess relationships among three Austrian species (Csaikl et al., 1990) and determine genetic distances and reconstruct phylogenetic tree for six species (Chelomina, 1993a, b); evolution of microsatellite alleles in four species (Makova et al., 2000); DNA-DNA hybridization results from analyses of three species (Catzeflis, 1990; Catzeflis et al., 1987); evolutionary classification of European members of subgenus Sylvaemus based on allozymic and chromosomal data (Orlov et al., 1996a, b); relationships among EuroCaucasian and Asian species as assessed by mtDNA cytochrome b sequences (Chelomina et al., 1998b), restriction analysis of total nuclear DNA (Chelomina, 1998), and a combination of cranial morphometrics, genotypic and allele frequencies of allozymes, and typing of mtDNA cytochrome b restriction fragment patterns and sequences (Hille et al., 2002); analyses of mtDNA cytochrome b divergence among western European species (Reutter et al., 2003); isozymic, chromosomal, and molecular divergence among Caucasian species (Chelomina, 1998a); differentiation of ribosomal DNA restriction sites among seven species (Suzuki et al., 1990); reconstructing evolutionary history and phylogenetic relationships among eastern Asian species derived from mtDNA cytochrome b and nuclear IRBP sequences (Serizawa et al., 2000; Suzuki et al., 2003; also see accounts of species); phylogeny and taxonomy of 15 species emphasizing Chinese species (Liu et al., 2004); and phylogeny of Apodemus focusing on subgenus Sylvaemus using nuclear IRBP gene sequences and cytochrome b and 12S rRNA mitochondrial markers (Michaux et al., 2002a). The inconsistency of molecular evolutionary rates affecting allozyme divergence within Apodemus is illuminated in an intelligent discussion by Hartl et al. (1992).
Numerous taxonomic provincial studies described morphological and other distinctions among sympatric or allopatric species of Apodemus; examples are a study of five species from Bulgaria (Popov, 1981); four species from Germany (Feiler and Tegegn, 1998), lower Danube River region (Fedorchenko and Zagorodniuk, 1994), Serbia and Montenegro (Todorovi
et al., 1974), W Turkey (Filippucci et al. 1996), E Turkey, W Armenia, and N Iran (Frynta et al., 2001); three species from Poland (Ruprecht, 1979), the Baltic region (Zagorodnyuk and Mezhzherin, 1992), and Daghestan (Lavrenchenko and Likhnova, 1995); and two species from Korea (Koh, 1988; Park et al., 1990). Vorontsov et al. (1989) described the presence of at least five species in the Caucasus, some of which at the time were distinguished only by biochemical traits; Vorontsov et al. (1992) recently identified and defined four species from there, and Mezhzherin et al. (1992) compared the Trancaucasian assemblage with European species using morphology and allozymic data. Tchernov (1979, 1994) reported on polymorphism, size trends and Pleistocene paleoclimatic responses of three Israeli species in an evolutionary context. Geographic ranges of various Palaearctic species mapped by Panteleyev (1998). Palaearctic species reviewed by Corbet (1978c, 1984) and Kobayashi (1985), N Eurasian by Mezhzherin (1997a), and Chinese by Xia (l984, 1985). Liu et al. (2002) provided review of taxonomy and phylogenetic relationships among the Chinese species they recognize.
Evolutionary history of Apodemus is documented back to late Pliocene in NW India (Kotlia, 1992), early Pliocene in S and N China (Cai and Qiu, 1993; Qiu and Storch, 2000; Wu and Flynn, 1992; Zheng, 1993) and middle Miocene (early Vallesian, 11.1-9.7 million years ago) of S and C Europe (Freudenthal and Martín Suárez, 1999; Martín Suŕrez and Mein, 1998). The genus was formerly regarded as no older than late Turolian (8.7-7.5 million years ago) and to have been derived from Progonomys (then thought to be the oldest European murine) through Parapodemus. Recent studies indicate that European fossil species of Apodemus exhibit their diagnostic derived dentition when first encountered in the Miocene, the genus is older than European Progonomys (Freudenthal and Martín Suárez, 1999) or any other European murine, and represents an independent lineage "of unknown origin" (Martín Suárez and Mein, 1998; but see de Bruijn et al., 1999, for another interpretation). Only Progonomys from the Siwaliks of Pakistan is older, the earliest record of fossils identified as that genus coming from strata dated at about 11.8 million years ago (Jacobs and Downs, 1994; Jacobs et al., 1990), or "Progonomys like species, or indeed true Progonomys" at 12.3 million years ago (Pilbeam et al., 1996:103). Whether Apodemus diverged from this earlier Asian Progonomys stock (in which case its presence in Europe represents westward dispersal during the Miocene) or evolved independently from an ancient Asian murine progenitor is unknown. Certainly Apodemus (and perhaps Mus; see that generic account) has the longest range in time of any known murine (Martín Suárez and Mein, 1998).