In the early systematic literature, red-backed voles were commonly listed under Evotomys (e.g., Hinton, 1926a; Miller, 1896, 1924; Miller and Rehn, 1901; Trouessart, 1880-1881), until Palmer (1928) argued the priority of Clethrionomys Tilesius, 1850. Most subsequent systematic authorities and faunal works throughout the 1900s have adopted Palmer’s correction in recognizing Clethionomys as the valid name (Corbet, 1978c; Corbet and Hill, 1991; Ellerman, 1941; Gromov and Erbajeva, 1995; Hall, 1981; McKenna and Bell, 1997; Miller and Kellogg, 1955; Mitchell-Jones et al., 1999; Musser and Carleton, 1993; Wilson and Ruff, 1999). In his deliberate search of the early European literature, Palmer (1928) had anticipated that he would discover an older synonym for Evotomys.
Regrettably, Palmer overlooked Pallas’ (1811) Myodes and Lataste’s (1883b) subsequent designation of Mus rutilus as its type species, the same one that he selected for Clethrionomys. Pallas (1811) included 10 species in his Myodes but did not specify a type species, a common omission for the era. In his studies of the voles of France, Lataste (1883a:324) had discussed the origin and contents of Pallas’ Myodes, noted (1883b:348-349) the junior status of North American Evotomys as then newly named by Coues (1874), and finally (1883b:349) clearly indicated rutilus as type species of Myodes to preserve the genus-group name. Lataste reiterated his restricted usage of Myodes, as subgenus, and the type status of rutilus in his 1887 classification of New and Old World Microtus. Some modern workers have accordingly employed Myodes as the oldest name for red-backed voles (Heller, 1968; Kretzoi, 1964, 1969; Kretzoi and Vertes, 1965a, 1965b; Naglov, 1998; Zagorodnyuk, 1990), and others have acknowledged it as probably valid while using Clethrionomys for reasons of familiarity (Gromov and Polyakov, 1977; Pavlinov and Rossolimo, 1987, 1998; Pavlinov et al., 1995a).
How Myodes became so long misassociated as a junior synonym of Lemmus is pertinent to consideration of its status. Whereas some early European workers narrowed the definition of Pallas’ Myodes to identify red-backed voles (Sélys Longchamps, 1839), North American taxonomists had used the genus to embrace brown lemmings (Baird, 1857; Coues, 1874). Lataste (1883a, b; 1887) frequently mentioned Sélys Longchamps as the prior authority for applying Myodes to red-backed voles (glareolus and rutilus), but took the critical additional step in designating Mus rutilus as its type species. Baird (1857) expressly followed Keyserling and Blasius (1840) for precedence in applying Myodes to North American brown lemmings (all as M. obensis). The latter’s classificatory listing of European birds and mammals arranged four species within Myodes (lemmus, obensis, lagurus, and torquatus), but they did not note a type species. The latter meaning was the one emphasized by Miller (1896:15), who stated "Since Myodes contained species of exactly the same modern genera as Lemmus Link and no groups not included in the latter, the name is a synonym of Lemmus." In other words, he based their synonymy on the equivalence of contents, as he understood generic limits in 1896, not on the principle of typification (Article 61; ICZN, 1999). Miller (1896:19-24) elsewhere summarized Lataste’s 1883 papers and even presented his composition of Myodes (as a subgenus of Microtus) that included only rutilus and glareolus. Palmer (1904:439) also had knowledge of Lataste’s nomenclatural action with regard to Myodes ("type said to be M. rutilus!"), yet confoundingly omitted any mention of it in his (1928) subsequent fixation of Mus rutilus in order to employ Clethrionomys instead of Evotomys. Miller’s interpretation was formalized by Hinton (1926a:210), who considered Mus lemmus to be the type species of Myodes and thereby reinforced its junior synonymy under Lemmus. In light of Lataste’s (1883b) legitimate indication of Mus rutilus as type species for Myodes (Article 69.1), Hinton’s passing listing of Mus lemmus as its type is invalid to establish the equivalence of Myodes with Lemmus (Article 70.2). Unfortunately, Hinton’s incorrect synonymy was perpetuated in the influential classifications of Ellerman (1941) and Simpson (1945) and became broadly adopted thereafter, especially by North American authors but not universally by European workers.
To our knowledge, three genus-group taxa are founded on Mus rutilus Pallas as type species: Myodes Pallas (1811), as subsequently designated by Lataste (1883b); Clethrionomys Tilesius (1850), as subsequently designated by Palmer (1928); and Evotomys Coues (1874), as originally designated by Coues. Where two or more genus-group taxa have the same type species, their names are considered to be objective synonyms (Article 61.3.3; ICZN, 4th Ed., 1999) and the senior among them assumes priority. Pavlinov et al. (1995a) questionably indicated Myodes as a nomen oblitum in the synonymy of Clethrionomys. The ICZN (1999), however, explains that both conditions of Article 23.9.1 must be jointly satisfied to reverse the principle of priority: the senior synonym cannot have been used as a valid name after 1899 (nomen oblitum), and the junior synonym must have been used extensively as the prevailing name (also see, 1999:111). Although the second condition supports retention of Clethrionomys, the first inescapably demands recognition of Myodes. In the following species comments, and elsewhere, readers should understand that wherever we use Myodes the supporting literature generally refers to Clethrionomys or Evotomys.
Usage of the family-group name is not so clearcut. Hooper and Hart (1962) named Clethrionomyini for those arvicolines with rooted molars and simple occlusal patterns. Later, Kretzoi (1969) coined Myodini for this tribe, given his recognition of Lataste’s (1883b) type species fixation and apparently unaware of Hooper and Hart’s action. Both Clethrionomyini (Gromov and Polyakov 1977; McKenna and Bell, 1997) and Myodini (Pavlinov and Rossolimo, 1998; Pavlinov et al., 1995a; Zagorodnyuk, 1990) have been recognized in recent classifications that employ arvicoline tribal ranks. The ICZN does not require replacement of an older family-group name when its type genus is reallocated as junior synonym of some other genus, upon which another family-group name is based (Article 40.1); however, maintenance of the junior family-group name may be defended in instances where a valid subsequent fixation had been overlooked and instability or confusion could likely ensue (Articles 41, 65.2.2). In view of the non-universal employment of Clethrionomyini versus Myodini and to lessen such potential confusion, we recommend usage of the junior name Myodini. The case must be ultimately decided by petition to the Commission.
We much regret that Miller, Hinton, and Palmer failed to appreciate Lataste’s (1883b) nomenclatural action affecting the status of Myodes. Over 50 years passed between description of Evotomys (Coues, 1874) and its replacement by Clethrionomys (Palmer, 1928). Just 36 years later, Kretzoi (1964) reminded the systematic community of Lataste’s type-species selection, reestablished the priority of Myodes, and listed Clethrionomys and Evotomys among its junior synonyms. With the advent of the global internet and greatly enhanced communication among scientists, we doubt that such oversights, issuing from the inertia of provincially familiar taxonomies, can be so long perpetuated.
Myodini. Myodes is closely related to Eothenomys (Koenigswald, 1980; Kretzoi, 1969), if not congeneric with it as suggested by some authors (Corbet, 1978c, 1984; Hooper and Hart, 1962; and see discussion under Eothenomys). Their close phylogenetic alliance is also supported by results of chromosomal analyses (Iwasa et al., 1999b) and DNA-DNA hybridization (Din et al., 1993).
Broad taxonomic reviews are provided by Aimi (1980), Corbet (1978c, 1984), and Gromov and Polyakov (1977). Aimi also retraced the complex taxonomic history of species, especially Japanese forms, variously allocated to Evotomys, Clethrionomys, Anteliomys, Phaulomys, or Eothenomys. Gromov and Erbajeva (1995) mapped the generic range and reviewed morphological and geographic contrasts among species within Russia and adjacent regions. All species of Myodes so far examined have 2n = 56, with 26 pairs of telocentric and one pair of metacentric chromosomes (Gamperl, 1982; Iwasa, 1998; Iwasa et al., 1999; Kaneko et al., 1998; Kartavtseva et al., 1998; Ma and Jiang, 1996; Modi and Gamperl, 1989; Nadler et al., 1976; Obara et al., 1995; Sokolov et al., 1990; Vorontsov et al., 1978; Yoshida et al., 1989). Electrophoretic data presented by Nadler et al. (1978), Chaline and Graf (1988), and Mezhzherin and Serbenyuk (1992); variation in mitochondrial DNA control region examined by Matson and Baker (2001). Fossil Myodes date to the late Pliocene of China (Zheng, 1993), early Pleistocene of Europe, late Pliocene of Ukraine (Topachevskii et al., 1998), and middle Pleistocene of North America (McKenna and Bell, 1997). Kawamura (1988) interpreted relationships among European and Japanese species based on living and Pleistocene samples.