Keen's long-eared bat (Myotis keenii) COSEWIC assessment and status report: chapter 3

Species Information

Name and classification

Keen’s long-eared bat, Myotis keenii (Merriam, 1895), belongs to the Order Chiroptera, Family Vespertilionidae. Findley (1972) assigned the North American long-eared Myotis species M. keenii, northern long-eared Myotis (Myotis septentrionalis), western long-eared bat (Myotis evotis), and the southwestern bat (Myotis auriculus) to the evotis group within the subgenus Myotis. The evotis group may also include Miller's Myotis (Myotis milleri) and the fringed bat (Myotis thysanodes). Although they share a number of similar morphological and behavioural traits, Manning (1993) speculated that North American long-eared Myotis were not a monophyletic group. Reducker et al. (1983), however, suggested that, based on chromosomal, electrophoretic, and mtDNA sequence data this group is monophyletic. The most recent mtDNA analysis by Dewey (unpublished data) indicates that only M. evotis, M. keenii, M. thysanodes and M. milleri form a monophyletic group.

Taxonomy of M. keenii and related species of long-eared Myotis is complex with a confusing nomenclatural history. This taxon was first described and named by Merriam (1895) who treated it as a subspecies of Verspertilio subulatus. The type locality for the type specimen (USNM 729220) is Masset, Graham Island, Queen Charlotte Islands, British Columbia. Miller (1897) subsequently classified this species as a member of the genus Myotis (Myotis subulatus keenii). In their taxonomic revision of American Myotis, Miller and Allen (1928) treated M. keenii as a distinct species, Myotis keenii with two allopatric subspecies: a coastal form M. keenii keenii and a central-eastern form M. keenii septentrionalis. Although a third subspecies (M. keenii auriculus) was recognized by Findley (1960) when he assigned long-eared Myotis from the southwestern United States to this species, Genoways and Jones (1969) treated the auriculus group as a distinct species M. auriculus, an arrangement recognized by most taxonomists. Based on a multivariate study of morphological traits, van Zyll de Jong (1979) concluded that M. k. keenii and M. k. septentrionalis were separate species distinguished by cranial, dental, and pelage characters. The two taxa were formally recognized as distinct species by van Zyll de Jong (1985) and Jones et al. (1986).

Taxonomists (e.g., Manning, 1993) have generally considered M. septentrionalis to be the closest relative among the long-eared Myotis species to M. keenii. However, from morphological and biogeographic evidence, van Zyll de Jong and Nagorsen (1994) concluded that M. keenii was most closely related to M. evotis. Using discriminant function analysis, they found that M. keenii from Haida Gwaii (=Queen Charlotte Islands)Footnote1 and M. evotis from areas east of the Coast Mountains showed no morphological overlap, and museum specimens from coastal areas of British Columbia could be confidently assigned to either species. Specimens from western Washington, however, showed some overlap in their morphology. They suggested that the two species were parapatric, with M. keenii being found in coastal regions, while M. evotis was found primarily east of the coastal mountain ranges. They speculated that M. keenii may have evolved in a coastal refugium.  Because they had few samples from the lower Fraser River valley, southeastern Vancouver Island, and western Washington State where the ranges of the two species overlap, they were unable to assess introgression.

Results from a recent molecular study using mtDNA (cytochrome b gene) by Tanya Dewey (unpublished data) support the close association of M. keenii to M. evotis and their distant relationship with M. septentrionalis (Figure 1). Sequence divergence between M. keenii and M. septentrionaliswas greater than 10% consistent with differentiation at the species level (Bradley and Baker 2001). Dewey also identified 3 relatively distinct lineages within M. keenii/ evotis – a clade corresponding to the present range of M. keenii, an evotis clade found in southern BC, Washington and Oregon, and an evotis clade found primarily in Alberta through to South Dakota. The M. evotis and M. keenii clades are largely allopatric, but individuals from both clades were found at the same location in the Skeena Mountains, and all 3 clades are represented in the Skagit valley. The sequence divergences among these lineages range between 0.8% and 3.3%, which, according to Bradley and Baker (2001) are well within the range of intraspecific divergence. Dewey’s results thus suggest that M. evotis and M. keenii are conspecific, with M. keenii representing a coastal subspecies. Until a broad systematic study is done integrating molecular (nuclear and mitochondrial DNA) and morphological data, however, we recommend that M. keenii be treated as a distinct taxonomic unit for conservation and management.

Other common names used for this species include: Keen's bat, Keen’s Myotis, and Keen's long-eared Myotis. The French common name is Vespertilio de Keen. We know of no Aboriginal name specific to this bat species. The Haida, for example, collectively referred to bats as “GudGadu Gamhlgaal” (pronounced “Goot gaado gum hl gaal”), which translates into “the animal that hangs upside down” (B. Wilson, pers. comm.). Consultation with other Aboriginal groups indicates that there appears to be no cultural memory of bats (N. Crookes, pers. comm.; H. Morven, pers. comm).

Description

M. keenii is a small bat (Figure 2) with dark brown dorsal fur and indistinct shoulder patches (van Zyll de Jong, 1985; Nagorsen and Brigham, 1993). The ventral fur tends to be buffy. A tiny fringe of hairs is evident on the outer edge of the tail membrane. The ears are long (usually >16 mm); the calcar (a cartilaginous spur on the heel bone) lacks a distinct keel. The skull has a relatively narrow rostrum and a steep sloping forehead region. The dental formula is: incisors 2/3, canines 1/1, premolars 3/3, and molars 3/3. Representative body measurements for adults (range in parentheses) are: wingspan 241 mm (224-262), ear length 17 mm (13-20), tragus length 9 mm (6-12), forearm length 36.2 mm (33.8-39.5), body mass 5.1 g (3.8-6.7).

Figure 1. Phylogenetic relationships among long-eared Myotis species from the Pacific Northwest. Consensus tree of parsimony derived from an analysis of 765 base pairs of cytochrome b, 225 parsimony-informative characters for 187 terminal taxa. Duplicate haplotypes merged and tree pruned to display samples of interest. From T. Dewey (unpublished data).

Figure 1. Phylogenetic relationships among long-eared Myotis species from the Pacific Northwest. Consensus tree of parsimony derived from an analysis of 765 base pairs of cytochrome b, 225 parsimony-informative characters for 187 terminal taxa. Duplicate haplotypes merged and tree pruned to display samples of interest. From T. Dewey (unpublished data).

Figure 2. Photograph of Keen’s long-eared bat (Myotis keenii) captured at Gwaii Haanas National Park Reserve, Haida Gwaii, British Columbia. Photo by D.W. Burles.

Figure 2. Photograph of Keen’s long-eared bat (Myotis keenii) captured at Gwaii Haanas National Park Reserve, Haida Gwaii, British Columbia. Photo by D.W. Burles.

M. keenii closely resembles M. evotis. Identification of live animals is problematic and a major impediment to field studies. M. keenii has on average a shorter forearm, metacarpal bones, and ears, but these measurements overlap among individuals of the two species (van Zyll de Jong and Nagorsen, 1994). Moreover, Burles (2001) demonstrated that ear length was a variable trait subject to significant measuring error. The identification keys in van Zyll de Jong (1985) and Nagorsen and Brigham (1993) used the distance that the ear extends beyond the nose (< 5 mm for M. keenii; > 5 mm for M. evotis) as a diagnostic trait for discriminating the species, but this trait appears to be highly variable among individuals. Burles (unpublished data) measured the extension of the ears beyond the nose in a large sample of M. keenii from Haida Gwaii and found that this measurement ranged from 2 to 9 mm. He concluded that it was not a valid taxonomic character. Ear colour, another trait used in the keys of van Zyll de Jong (1985) and Nagorsen and Brigham (1993), also appears to be variable. Nagorsen (2002) concluded that M. keenii and M. evotis cannot be reliably discriminated using external traits. Skull and dental measurements taken on museum specimens are more reliable, toothrow length being the best univariate discriminator (van Zyll de Jong and Nagorsen, 1994). Nevertheless, these measurements show some overlap between the two taxa, and positive identification requires multivariate analysis of cranial and dental measurements .

Nationally significant populations

No subspecies are recognized in M. keenii and there are no nationally significant populations.

Page details

Date modified: