Merriam’s (1894) definition of the subfamily included North American woodrats and certain South American fossils with high-crowned molars (Ptyssophorus and Tretomys, now considered synonyms of Reithrodon by Pardiñas, 2000a). Usage as a formal subfamily was observed (e.g., Miller and Rehn, 1901; Miller, 1912b) until Miller and Gidley (1918) considered the genera to be members of a diverse Cricetinae, as did Ellerman (1940) and Simpson (1945). A broadened family-group concept reemerged in an informal way as the "neotomine-peromyscines" (Hooper, 1960; Hooper and Musser, 1964a; Carleton, 1980) and was eventually nomenclaturally recognized as distinct from sigmodontines, whether as a tribe (Hershkovitz, 1966b, as Peromyscini) or subfamily (Reig, 1980, 1981, as Neotominae). Phylogenetic diagnosis and cladistic demonstration of neotomine monophyly remain ambiguous based on taxonomically broad surveys of morphological traits (Carleton, 1973, 1980; Steppan, 1995; Voss and Linzey, 1981) or cytochrome b data (D’Elía et al., 2003), but not other mitochondrial genes (Engel et al., 1998) or mitochondrial and nuclear genes in combination (D’Elía, 2003). Coupled with this uncertainty is that of the sister-group relationship between neotomines and sigmodontines (see D’Elía, 2000, for review), as assumed in evolutionary narratives (e.g., Hershkovitz, 1966b; Marshall, 1979; Patterson and Pascual, 1972) and early interpretations of phylogeny (Hooper and Musser, 1964a). Other cognate possibilities, such as arvicolines and Old World cricetines, are variously implicated in phylogenetic studies of morphology (Carleton, 1980), of DNA-DNA hybridization (Catzeflis et al., 1993), and of mitochondrial and nuclear DNA sequences (D’Elía, 2003; D’Elía et al., 2003; Engel et al., 1998; Michaux et al., 2001b).
The ancestry of extant neotomines has been loosely connected to Copemys (Jacobs and Lindsay, 1984; Slaughter and Ubelaker, 1984), a middle Miocene-early Pliocene North American cricetid that has been variously invoked as progenitor of Peromyscus (Lindsay, 1972), Onychomys (Jacobs, 1977b), and Bensonomys (Baskin, 1978). The morphological limits and specific contents of Copemys are poorly understood, however, and its evolutionary relationships and biogeographic origin have been subject to several, sometimes contradictory, interpretations (see commentary and references in Carleton and Musser [1984:306-308], Baskin [1986:296], and Korth [1994:231-232]). Examples of extant genera are known from the late Miocene (Neotoma and Peromyscus), and others appear in early Pliocene strata (Baiomys, Onychomys, Reithrodontomys) (Carleton and Eshelman, 1979; Korth, 1994; Packard, 1960; Zakrewski, 1993).
Periodic synopses of systematic understanding and distributions provided by Miller and Rehn (1901), Miller (1924), Hall and Kelson (1959), and Hall (1981). Recent faunal treatises update and summarize, in varying detail, aspects of natural history, distribution, biogeography, and specific and-or infraspecific taxonomy: North America (Baker et al., 2003b; Jones et al., 1997; Wilson and Ruff, 1999) and the E USA (Whitaker and Hamilton, 1998); Honduras (Marineros and Gallegos, 1998) and Central America (Reid, 1997); México (Ceballos et al., 2002a; Ramírez-Pulido et al., 1996), NW México (Alvarez-Castañeda and Cortés-Calva, 1999), and the states of Baja California (Hafner and Riddle, 1997), Chiapas (Espinoza M. et al., 1999a, b), Jalisco (Guerrero Vázquez et al., 1995), México (Ramírez-Pulido et al., 1995), Morelos (Alvarez-Castañeda, 1996; Alvarez et al. 1998), Sonora (Caire, 1997), and Quintana Roo (Pozo de la Tijera and Escobedo Cabrera, 1999).