Printer Friendly

A blind-test comparison of the reliability of using external morphology and echolocation-call structure to differentiate between the Little Brown Bat (Myotis lucifugus) and Yuma Myotis (Myotis yumanensis).

ABSTRACT--Species that overlap in their morphologies are sometimes difficult to distinguish from one another, which can complicate species' conservation and management. The Little Brown Bat (Myotis lucifugus) and Yuma Myotis (Myotis yumanensis) are sympatric in parts of their range in western North America, and they overlap in morphology, making them difficult to tell apart in the hand in some areas, such as the Pacific Northwest. We compared various methods of distinguishing between M. lucifugus and M. yumanensis to genetic results, using a blind test approach to remove observer bias. Using multiple independent observers, we used external morphology and echolocation-call structure to classify bats from a maternity colony consisting of both species. Genetic analysis confirmed 13 M. lucifugus and 40 M. yumanensis. Minimum echolocation-call frequency separated 100% of M. lucifugus from M. yumanensis using a cut-off of 43 kHz. All M. lucifugus had a minimum echolocation-call frequency [greater than or equal to] 42.8 kHz, whereas M. yumanensis had a minimum frequency [greater than or equal to] 44.55 kHz. There was some overlap in forearm length; a cut-off of 36 mm would have correctly identified 77% of M. lucifugus and 100% of M. yumanensis to species. Criteria based on subjective assessment of fur sheen and length as well as ear color were moderately successful (90.5 and 77% success by 2 separate observers) in distinguishing between the 2 species. The use of Munsell soil color charts and multivariate statistics to classify fur and membrane color and confirm species identification was not successful. Our results suggest that mean minimum call frequency alone is sufficient for distinguishing between M. lucifugus and M. yumanensis. Use of quantitative rather than qualitative criteria eliminates observer bias and appears to be better for identifying these 2 species.

Key words: bats, blind test, Chiroptera, echolocation, Little Brown Bat, Myotis lucifugus, Myotis yumanensis, species identification, Yuma Myotis

**********

The inability to accurately identify species that overlap wholly or partially in morphology can have scientific, management, conservation, and even economic, implications (Arlettaz and others 1997; Davidson-Watts and Jones 2006; Bickford and others 2007; Sattler and others 2007; Boston and others 2010). There are multiple examples of morphological overlap across various mammalian taxa, including rodents (Russo and others 2006; Lalis and others 2009), tenrecs (Olson and others 2004), and bats (Pipistrellus pipistrellus and P. pygmaeus: Barlow and others 1997; Haussler and others 2000; Myotis myotis and M. blythii: Arlettaz 1999; Corynorhinus townsendii, C. rafinesquii, and C. mexicanus: Piaggio and Perkins 2005; Myotis evotis and M. keenii: Nagorsen and Brigham 1993; and Myotis lucifugus and M. yumanensis: Fenton and Barclay 1980; Nagorsen and Brigham 1993). The prevalence of examples among bats is likely due to the fact that flight constrains morphology; therefore, bats of the same size tend to be similar in their morphology (Barclay and Brigham 1991).

The Little Brown Bat (Myotis lucifugus) and the Yuma Myotis (Myotis yumanensis) are sympatric in parts of their range in western North America, and they overlap in size and morphology, making them difficult to tell apart in the hand in some areas (Barbour and Davis 1969; Parkinson 1979; Herd and Fenton 1983; Nagorsen and Brigham 1993; Olson and others 2014). The 2 species were even once thought to hybridize in some areas where their ranges overlap (Barbour and Davis 1969; Parkinson 1979), although more recent research does not support this (Herd and Fenton 1983; Piaggio and others 2002). Despite their similarities in morphology, M. yumanensis is more maneuverable than M. lucifugus (Aldridge 1986). The 2 species also have different foraging preferences and diet (Herd and Fenton 1983; Saunders and Barclay 1992; Nagorsen and Brigham 1993), as well as different conservation issues. Myotis lucifugus was recently assessed as Endangered in Canada due to the effects and continued threat of white-nose syndrome on populations (COSEWIC 2013), and subsequently listed on Schedule 1 of the Species at Risk Act (SC 2002) in December 2014 under the emergency listing provisions of that Act. These differences in ecology and conservation status make it especially important to be able to distinguish between M. lucifugus and M. yumanensis.

External morphology has been used to distinguish between M. lucifugus and M. yumanensis in the field, with varying success (Fenton and Barclay 1980; van Zyll de Jong 1985; Nagorsen and Brigham 1993; Rodhouse and others 2008). Myotis lucifugus is described as having long, sleek, and glossy dorsal fur, dark basal fur on the shoulders, dark brown to blackish ears, and a forearm length usually >36 mm (Herd and Fenton 1983; van Zyll de Jong 1985; Nagorsen and Brigham 1993), although in the Okanagan Valley, British Columbia, non-reproductive females had a dull pelage (Herd and Fenton 1983). Myotis yumanensis, the smaller of the 2 species, is characterized as having short and dull dorsal fur, no dark basal fur on the shoulders, paler ears, and a forearm length usually [less than or equal to] 36 mm (Nagorsen and Brigham 1993). In 1 study, pelage sheen separated the 2 species with 96% success, whereas ear color was 82% successful (Rodhouse and others 2008). For some other species, pelage color is the key character used to distinguish between them in the field (for example, Piaggio and Perkins 2005). However, pelage color can vary with geography and reproductive status even within a species (Herd and Fenton 1983; Warner and Czaplewski 1984; Manning and Jones 1989; Solick and Barclay 2006). Likewise, in a study of 101 bats, only 18% of M. yumanensis and 17% of M. lucifugus were correctly identified to species based on forearm length due to considerable interspecies overlap in these measurements (Rodhouse and others 2008).

Differences in echolocation-call structure have been used to distinguish between species of bats with morphological overlap. For example, the cryptic species Pipistrellus pipistrellus (Common Pipistrelle) and Pipistrellus pygmaeus (Soprano Pipistrelle) are almost impossible to distinguish morphologically (Barlow and others 1997; Haussler and others 2000), but can be reliably distinguished by a 10-kHz difference in the mean frequency of maximum energy in their echolocation calls 0ones and van Parijs 1993; Barlow and Jones 1997). Herd and Fenton (1983) found significant differences in the search phase echolocation calls of M. lucifugus and M. yumanensis, but did not think that they could be reliably identified in this way due to overlap in measurements. Within species such as M. lucifugus that forage in a variety of habitats, echolocation-call structure varies with the amount of clutter (Broders and others 2004; Wund 2006), which further complicates the use of echolocation-call measurements to distinguish between species (Barclay and Brigham 2004). Clutter, which has been adapted from radar theory, refers to any obstacle (to a flying bat) that produces echoes that may interfere with detection of echoes from a prey target, or presents an obstacle to flight (Fenton 1990). Despite the interspecific overlap, the minimum frequency of echolocation-calls of M. lucifugus ranges from about 35 to 43 kHz (Saunders and Barclay 1992; O'Farrell 1999; Murray and others 2001; Broders and others 2004; Wund 2006), while that of M. yumanensis ranges from about 46 to 50 kHz (O'Farrell and others 1999), suggesting that these ranges may be a reliable measure to distinguish between species. Differences in characteristic frequency, the end point of the flattest (in terms of frequency) portion of an echolocation call (Corben and O'Farrell 1999), have also been used to distinguish between the 2 species (Weller and others 2007). In fact, characteristic frequency has been suggested as a better measure than minimum frequency because it is less likely to be affected by decreasing amplitude (such as when the distance between the bat and the microphone increases as the bat flies away; Chris Corben, pers. comm.; Szewczak 2000). Depending on the type of ultrasonic detector used, there may be broad overlap in minimum frequency and characteristic frequency (Weller and others 2007).

Due to observed overlap in morphological and acoustic characteristics, none of the aforementioned criteria are reported to provide 100% success in distinguishing M. lucifugus from M. yumanensis (Fenton and Barclay 1980; Nagorsen and Brigham 1993; Rodhouse and others 2008). Using a combination of characteristics has provided varying degrees of success (Herd and Fenton 1983; Weller and others 2007; Rodhouse and others 2008). For instance, Weller and others (2007) successfully identified 92% of M. lucifugus and 91% of M. yumanensis in the Pacific Northwest using a combination of forearm length and characteristic frequency of echolocation calls. Likewise, others found that a combination of forearm length and pelage sheen correctly identified 92% of M. yumanensis and 66% of M. lucifugus (Rodhouse and others 2008).

Classifying a species using qualitative characteristics such as fur length or sheen is subjective because it depends on the experience of the observer and environmental conditions (such as lighting), and, if relied on exclusively, could result in different identifications from 1 observer to the next (for example, Lobert and others 2001). The ideal way to determine the reliability of an identification technique is to use blind testing with multiple observers (for example, Scheuer 2002). Despite the difficulties in distinguishing between M. lucifugus and M. yumanensis in some areas of geographic overlap, there are no published studies that have compared external morphology, echolocation-call structure, and genetics in a blind test. The objective of our study was to determine the validity and reliability of using external morphology, echolocation-call structure, or both to distinguish between M. lucifugus and M. yumanensis using a blind-test approach and genetic confirmation of identity. A maternity colony of both species roosting in the same building provided the ideal opportunity for such a study.

METHODS

Our study site was a maternity colony for both M. lucifugus and M. yumanensis in a maintenance building at the Hozomeen Ranger Station in North Cascades National Park, USA (UTM: Zone 10, 2377506.85E, 6248600.55N, NAD83; elevation = 500 m) in the Skagit Valley of the Cascade Range. Both species roosted in the attic of this large building, although we do not know if they were spatially separated from each other. As a part of a larger study (Luszcz 2004), we conducted fieldwork for this project over 2 d (17 and 18 May 2002). All capture and handling of bats followed guidelines set by the University of Calgary Animal Care Committee, and the Canadian Council on Animal Care. At dusk each night, we captured bats exiting the building using mist nets erected along nearby flyways. We held captured bats in cloth bags until mist nets were closed and measuring could begin.

Each bat was assigned a number, which was used throughout the study to link all measurements. Each observer was responsible for 1 identification task throughout the entire measurement process, and tentative identifications were not disclosed. Only 1 observer assessed sex, age, mass, and forearm length because these measures are less subject to observer variation. Two observers independently assessed fur and membrane (ear or wing) color using Munsell[R] Soil Color Charts (Munsell Color Laboratory 2000) as outlined in Solick (2004). Munsell Soil Color Charts consist of color chips organized by their hue, value, and chroma, the 3 color attributes recognized by the Munsell notation system (Miller 1958). Hue describes a color's relation to red, yellow, green, blue, and purple, with higher hue numbers being more yellow and less red. Value refers to the brightness of a color, and ranges from 0 (pure black) to 10 (pure white). Chroma denotes the degree of saturation, on a scale from 0 to 10. High chroma indicates a rich, vibrant color, whereas low chroma colors are pale (Munsell Color Laboratory 2000). Each observer classified fur and membrane color by placing the Munsell Soil Color Chart over the fur on the back below the scapula and on a membrane, using the same light fixed at a standard distance from the bat.

Two other observers, with similar bat identification experience, independently identified bats as M. lucifugus or M. yumanensis based on morphological criteria from the literature. Fur length was classified as either short or long, fur sheen as either dull or glossy, and ear color as either light to medium brown, or dark brown. A bat required a minimum of 2 out of 3 criteria (fur length, fur sheen, and ear color tone) to be classified to a species.

To assess whether wing loading or aspect ratio (2 measures of wing size and shape; Norberg and Rayner 1987) were useful in identification, 2 observers extended the right wing of each bat to its fullest and traced its outline onto paper. We later scanned the wing tracings and imported the digital images into BatWing.exe 1.0 software (Harley and Miller-Butterworth 2000). Using the software, we obtained measures of wing area and length and then calculated wing loading and aspect ratio following methods outlined in Solick and Barclay (2006). Wing loading and aspect ratio provide information about a bat's ability to exploit cluttered habitats (Norberg and Rayner 1987).

Another observer took a 3-mm diameter wing biopsy punch from each bat for genetic analysis (Worthington and Barratt 1996), conducted at Portland State University (JZ). DNA was extracted from the biopsy samples using the Qiagen[TM] DNEasy Tissue Extraction Kit. A 190 base-pair mitochondrial DNA fragment from the 12S ribosomal subunit gene was amplified under standard PCR conditions (Zinck and others 2004). The resulting PCR product was sequenced using an ABI 3100 automatic sequencer. Sequences were aligned using Seqed (ABI), and analyzed in PAUP 4.0. Sequences were initially compared to a library of over 300 additional Myotis sequences. We used the genetic analysis to identify individuals to species and then used those identifications to compare to the identifications based on the characteristics measured in this study.

We released each bat in the same location along a laneway in a forested campsite less than 500 m from the roost, and we recorded echolocation calls using Anabat II ultrasonic detectors (Titley Electronics, Ballina, NSW, Australia) set to a division ratio of 16. Detectors were turned on only once the bats were free-flying, and we recorded calls until bats flew out of range. We used Analook software to purge call sequences of extraneous noise (pixels) and to measure call morphology. We adapted rules from Patriquin (2001) to clean echolocation calls. The final rules were: (a) pixels had to follow the same general trajectory of a call or they were deleted; (b) along a trajectory of points in a call, a lone pixel near the beginning or end of the call had to be separated from the rest of the call by no more than 5 kHz or it was deleted; and (c) along a trajectory of points in a call, groups of points near the beginning or end of the call had to be separated from the rest of the call by no more than 10 kHz or they were deleted. Minimum call-frequency values and characteristic-frequency values were each averaged to obtain 1 data point per bat. Number of usable calls recorded per bat ranged from 7 to 172, except in 1 case where only 1 usable call was recorded.

For all statistical analyses, we used the online statistical tool VassarStats (http://vassarstats. net/) and Microsoft Excel. We compared minimum echolocation call-frequency, characteristic call-frequency, mass, forearm length, wingspan, wing loading, and aspect ratio within and between the 2 species using 2 sample t-tests ([alpha] = 0.05). Means and standard deviations are reported unless otherwise indicated. For the qualitative ear and wing variables, we examined agreement between observers, and between each observer and the genetic results, by calculating Cohen's Kappa statistic, a measure of inter-observer agreement whereby agreement due to chance is factored out (Cohen 1960). Kappa values can help to assess the reliability of a method of categorization. Kappa values range from 1 to-1. A Kappa value of 1 indicates perfect agreement between observers, 0 means that any agreement is totally due to chance, and-1 means there is perfect disagreement. Common criteria for Cohen's Kappa are as follows: <0 = no agreement; 0 to 0.20 = poor agreement; 0.2 to 0.4 = fair agreement; 0.4 to 0.6 = moderate agreement; 0.6 to 0.8 = good agreement; and [greater than or equal to] 0.8 = very good agreement (Landis and Koch 1977). We also used Cohen's Kappa to examine interobserver agreement of the Munsell Soil Color Chart variables (hue, value, and chroma).

RESULTS

Over 2 nights of mist netting, we captured 55 bats during their dusk exit, including 1 Myotis evotis (Long-eared Myotis) female. We obtained complete information, including usable echolocation calls, for 52 bats (43 females and 9 males); we did not obtain echolocation calls for 1 additional bat, but we included it in the morphological analyses. Genetic analysis confirmed 13 M. lucifugus (all females) and 40 M. yumanensis (of which 31 were females). Descriptive statistics for measurements of echolocation-call frequency and external morphology are summarized in Table 1.

There was no significant difference in mean minimum echolocation-call frequency between male and female M. yumanensis, or in mean characteristic frequency (Table 2); thus, we combined male and female M. yumanensis in the interspecies analysis of call parameters. Mean minimum call frequency was significantly different between M. lucifugus (39.82 [+ or-] 1.84 kHz) and M. yumanensis (47.48 [+ or-] 1.35 kHz; t = 15.77, df = 50, P < 0.0001; Table 2, Fig. 1). The highest minimum frequency for M. lucifugus was 42.8 kHz and the lowest minimum frequency for M. yumanensis was 44.5 kHz. Thus, in our study, a minimum frequency cut-off of 43 kHz separated 100% of M. lucifugus from M. yumanensis. Characteristic frequency also differed significantly between M. lucifugus and M. yumanensis (t = 11.85, df = 48, P < 0.0001; Table 2), although there appeared to be some overlap between species (ranges: MYLU = 38.99-50.42, MYYU = 45.85-57.26; Table 1). However, we considered the highest characteristic frequency for M. lucifugus (50.42 kHz) as an outlier. Thus, if we consider the second-highest characteristic frequency (45.55 kHz) for M. lucifugus as the correct highest characteristic frequency for the species in our study, and compare that to the lowest characteristic frequency for M. yumanensis (45.85 kHz), then the characteristic frequencies for the 2 species do not overlap.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Forearm lengths and wingspans did not differ significantly between male and female M. yumanensis, nor did wing loading or aspect ratio (Table 2). Although female M. yumanensis had a slightly greater wingspan than males (Table 1), this difference was not significant (Table 2). Female M. yumanensis, however, were significantly heavier than males (t = 4.3, df = 9.7, P = 0.002; Table 2). Therefore, given that all measurements of external morphology (except mass) for M. yumanensis were relatively similar, we combined males and females of this species when we compared M. lucifugus and M. yumanensis. Myotis lucifugus were significantly heavier, had longer forearms, and greater wingspans than M. yumanensis (Table 2, Fig. 2). Ten of the 13 M. lucifugus had a forearm length >36 mm, and all M. yumanensis (n = 40) had a forearm length [less than or equal to] 36 mm. Therefore, 77% of M. lucifugus and 100% of M. yumanensis in our study could have been correctly identified to species on the basis of forearm length alone if a cut off of 36 mm was used. Despite the differences in wingspan, the 2 species did not differ in their wing loading or aspect ratio (Table 2). When we compared females only, the results were the same. Female M. yumanensis weighed significantly less and had shorter forearms and smaller wingspans than female M. lucifugus (Table 2). They did not, however, differ in their wing loading or aspect ratio (Table 2).

Compared to the 100% success in identifying individuals to species using echolocation-call characteristics, criteria based on subjective assessment of fur sheen and length, and ear color were less successful in distinguishing between M. lucifugus and M. yumanensis. Observer 1 correctly identified 48 of 53 bats (90.5% success) to species (Cohen's Kappa statistic [kappa] = 0.76; good agreement), and Observer 2 correctly identified 41 of 53 bats (77% success) to species ([kappa] = 0.49; moderate agreement). There was moderate inter-observer agreement in the species identifications made by each of the 2 observers ([kappa] = 0.56). In the cases of misidentification: Observer 1 misidentified M. yumanensis as M. lucifugus in 4 cases, and M. lucifugus as M. yumanensis in 1 case; Observer 2 misidentified M. yumanensis as M. lucifugus in 10 cases, and M. lucifugus as M. yumanensis in 2 cases. In 8 of 14 cases of the misidentification of M. yumanensis as M. lucifugus, M. yumanensis was classified as having all 3 M. lucifugus traits: long glossy fur and dark ears. In the remaining 6 cases, M. yumanensis was classified as having 2 out of 3 M. lucifugus traits. In the cases of M. lucifugus being misidentified as M. yumanensis, M. lucifugus was classified in 2 of 3 cases as having all 3 M. yumanensis traits: short dull fur and light ears. Finally, the 3 M. lucifugus with forearm lengths <36 mm, which placed them in the area of overlap with the range of M. yumanensis forearm lengths (Fig. 2), were successfully identified to species using external morphology as well as minimum echolocation-call frequency.

Attempts to identify bats to species using Munsell Soil Color Charts were unsuccessful. Hue, value, and chroma variables for fur color resulted in poor (or no) agreement between observers ([kappa] = -0.24, 0, and 0.10, respectively). Based on membrane color, there was also poor agreement for hue ([kappa] = 0.08) and moderate agreement for value and chroma ([kappa] = 0.49 and 0.49, respectively) between observers. In the 2 instances of moderate agreement, there was little variation in value or chroma raw data values; thus, no further analysis was undertaken with these data.

DISCUSSION

Our blind test approach using multiple observers to compare methods for distinguishing between M. lucifugus and M. yumanensis provided valuable insight into the reliability of each method as a species identification tool. In our study, minimum echolocation-call frequency was the best variable for species confirmation. Based on the genetic results, the separation of minimum frequency at 43 kHz was distinct, with no overlap between species. In another study that examined the same parameter between M. lucifugus and M. yumanensis, there was no overlap in mean lowest frequency from Anabat data sets (and little overlap for Anabat single calls recorded from Anabat zero-crossing detectors; Figure 3 in Weller and others 2007), lending support to our methods and results. Weller and others (2007) did, however, observe broad overlap in lowest frequency between the 2 species in data recorded using time-expansion detectors and analyzed using SonoBat call analysis software. These differences in results may relate to differences in the 2 detection systems (Fenton 2000).

Minimum frequency is less consistent than characteristic frequency, because if there is a significant droop, or tail, after the characteristic frequency point, the extent to which this is detected becomes highly dependent on the distance to the bat. This is 1 reason why characteristic frequency is considered a better measure for acoustically distinguishing between species. As a bat flies further away, less of the tail will be recorded as its amplitude decreases as the frequency drops (Chris Corben, pers. comm.). On the other hand, as clutter increases, a bat's echolocation call becomes steeper. The characteristic frequency increases and requires more magnification to see changes in the slope in order to define it (Broders and others 2004; Wund 2006), compared to minimum frequency. Using our rules for cleaning Anabat calls, we did not find minimum frequency to be misleading. However, we did find what appeared to be overlap in characteristic frequency between the 2 species, although it was due to what we consider to be an outlier frequency for 1 M. lucifugus.

More recently, field researchers in California and British Columbia have successfully distinguished between captured M. lucifugus and M. yumanensis in the field using a "bag test" (Dave Johnston, HT Harvey and Associates, Los Gatos, CA, pers. comm.; Cori Lausen, Birchdale Ecological Ltd., Kaslo, BC, pers. comm.). They recorded echolocation calls of bats while they were still in their cloth holding bags, differentiating the 2 species by their minimum frequency. In British Columbia, they used an Echo Meter EM3 detector (Wildlife Acoustics, Massachusetts, USA), and observed emitted pulses in the field: minimum frequencies of 30 to 37 kHz for M. lucifugus and >42 kHz forM. yumanensis (Cori Lausen, pers. comm.). Their minimum frequency cut-off was similar to that observed in our study. This "bag test" was 100% accurate when compared to genetic results (n = 13; Cori Lausen, pers. comm.). A bat echolocating in a bag is responding to a high clutter environment. Broders and others (2004) similarly found that rates of misclassification were lowest in high clutter. Determining minimum echolocation-call frequency in situ would make identification of M. lucifugus and M. yumanensis much easier than our methods that measure the mean minimum frequency of all calls in a sequence after manual call cleaning, or even methods that apply custom filters in Analook or auto-identification software (Corben and O'Farrell 1999; Weller and others 2007). Application of the "bag test" in other geographic areas would be an ideal confirmation of its strength in distinguishing between M. lucifugus and M. yumanensis in the hand. Furthermore, comparing "bag test" minimum frequency to that of calls collected from a free-flying bat after it has been released would be an excellent test of the strength of both methods.

Measurements of external morphology resulted in variable and sometimes inconclusive species identifications. Although body mass and wingspan significantly differed between M. lucifugus and M. yumanensis, the interspecific overlap in these measurements reaffirmed that they alone are not adequate for species identification. Measurements of wing loading and aspect ratio were too similar to be of any value in separating the 2 species. Contrary to Rodhouse and others (2008), we found little overlap between species in measurements of forearm length; however, our sample size was relatively small and only from 1 geographic location. As has been found in other Myotis species, female M. yumanensis were significantly heavier than males (for example see Kalcounis and Brigham 1995). Had we captured male M. lucifugus in our study, we might have observed even more overlap in body mass and forearm length. The 3 M. lucifugus that overlapped with M. yumanensis in mass and forearm length (Fig. 2) were identified to species correctly by their fur length and sheen, and ear color, as well as minimum echolocation call frequency. It should also be noted that bats misidentified by their fur length and sheen, and ear color would have been correctly identified by their forearm length and by their minimum echolocation-call frequency, although we did not have an echolocation call recording for 1 individual.

Distinguishing between M. lucifugus and M. yumanensis based on fur characteristics and ear color (van Zyll de Jong 1985; Nagorsen and Brigham 1993) was only moderately successful and is quite subjective. Both observers had a similar amount of bat identification experience, yet showed only moderate agreement in their species identification. In our study, only 2 M. lucifugus were misidentified as M. yumanensis (due to short dull fur), whereas 8 misidentifications in the opposite direction occurred (due to long glossy fur and dark ears). The Skagit Valley, where this study occurred, is a coastal-interior transition zone in the Cascade Range. Thus, some of the M. yumanesis in our study may have been the coastal subspecies, M. yumanensis saturatus Miller 1897, which is a dark-brown to chestnut form with black ears and membranes (van Zyll de Jong 1985). Species identification using these characteristics could have been further complicated if we had conducted this study later in the season when juveniles were flying, because M. lucifugus juveniles have been reported to have darker fur than adults (Fenton and Barclay 1980), whereas yearlings tend to have duller fur than adults (Herd and Fenton 1983). Our attempts to use Munsell Soil Color Charts to distinguish differences in fur and membrane color between M. lucifugus and M. yumanensis were unsuccessful because the results were irreproducible, as evidenced by poor inter-observer agreement (Cohen 1960), suggesting that this method is likely not worthy of further testing.

In conclusion, our results demonstrate that mean minimum frequency can be a strong predictor of species on its own when distinguishing between M. lucifugus and M. yumanensis. However, due to variation in echolocation in different geographic locations and habitats across the range of these 2 sympatric species (for example, Weller and others 2007), a 43 kHz minimum frequency cut-off may not be applicable in all areas. Furthermore, we caution that this method should not be applied to free-flying bats across various habitat types, but should be limited to identification of captured bats released into the same conditions (for example, capture site). Further study in different geographic areas and habitats could determine whether our methods and specific cut-off value are generally applicable.

Acknowledgements

Funding for this work came from the Skagit Environmental Endowment Commission and the Natural Sciences and Engineering Research Council of Canada. Permission to study the colony was granted by the US National Park Service and Seattle City Light. We thank R Moody for volunteering his time on the field project. P Ormsbee provided advice on study design. Thanks also to D Solick for assistance with Munsell Soil Charts, L Ainsworth and R Joy for statistical assistance, C Corben for providing feedback on portions of the manuscript, and C Lausen and D Johnston for providing feedback on portions of the manuscript and for sharing unpublished data and information.

LITERATURE CITED

Aldridge H. 1986. Manoeuvrability and ecological segregation in the Little Brown (Myotis lucifugus) and Yuma (M. yumanensis) Bats (Chiroptera: Vespertilionidae). Canadian Journal of Zoology 64:1878-1882.

Arlettaz R. 1999. Habitat selection as a major resource partitioning mechanism between the two symptric sibling bat species Myotis myotis and Myotis blythii. Journal of Animal Ecology 68: 460-471.

Arlettaz R, Perrin N, Hausser J. 1997. Trophic resource partitioning and competition between the two sibling bat species, Myotis myotis and Myotis blythii. Journal of Animal Ecology 66:897-911.

Barbour R, Davis W. 1969. Bats of America. Lexington, KY: The University Press of Kentucky. 286 p.

Barclay RMR, Brigham RM. 1991. Prey detection, dietary niche breadth, and body size in bats: Why are aerial insectivorous bats so small? American Naturalist 137:693-703.

Barclay RMR, Brigham RM. 2004. Geographic variation in the echolocation calls of bats: A complication for identifying species by their calls. In: Brigham RM, Kalko EKV, Jones G, Parsons S, Limpens HJGA, editors. Bat echolocation research: Tools, techniques, and analysis. Austin, TX: Bat Conservation International. p144-149.

Barlow KE, Jones G. 1997. Differences in songflight calls and social calls between two phonic types of the vespertilionid bat Pipistrellus pipistrellus. Journal of Zoology London 241:315-324.

Barlow KE, Jones G, Barratt EM. 1997. Can skull morphology be used to predict ecological relationships between bat species? A test using two cryptic species of pipistrelle. Proceedings of the Royal Society of London B 265:1695-1700.

Bickford D, Lohman DJ, Sodhi NS, Ng PKL, Neier R, Winker K, Ingram K, Das I. 2007. Cryptic species as a window on diversity and conservation. Trends in Ecology and Evolution 22:148-155.

Boston ESM, Buckley DJ, Bekaert M, Gager MY, Lundy MG, Scott DD, Prodohl PA, Montgomery WI, Marnell F, Teeling EC. 2010. The status of the cryptic bat species, Myotis mystacinus and Myotis brandtii in Ireland. Acta Chiropterologica 12:457-4161.

Broders H, Findlay C, Zheng L. 2004. Effects of clutter on echolocation call structure of Myotis septentrionalis and M. lucifugus. Journal of Mammalogy 85:273-281.

Cohen J. 1960. A coefficient of agreement for nominal scales. Educational and Psychological Measurement 20:37-46.

Corben C, O'Farrell MJ. 1999. Techniques for the effective use of Anabat in identifying free-flying bat species. Anabat System Manual.

COSEWIC. 2013. COSEWIC assessment and status report on the Little Brown Myotis Myotis lucifugus, Northern Myotis Myotis septentrionalis and Tri-colored Bat Perimyotis subflavus in Canada. Ottawa, ON: Committee on the Status of Endangered Wildlife in Canada, xxiv + 93 p.

Davidson-Watts I, Jones G. 2006. Differences in foraging behaviour between Pipistrellus pipistrellus (Schreber, 1774) and Pipistrellus pygmaeus (Leach, 1825). Journal of Zoology 268:55-62.

Fenton, MB. 2000. Choosing the 'correct' bat detector. Acta Chiropterologica 2:215-224.

Fenton MB, Barclay RMR. 1980. Myotis lucifugus. Mammalian Species 142:1-8.

Harley EH, Miller-Butterworth CM. 2000. A software assistant for measuring bat wings. Bat Research News 41:99-102.

Haussler U, Nagel A, Braun M, Arnold A. 2000. External characters discriminating sibling species of European pipistrelles, Pipistrellus pipistrellus (Schreber, 1774) and P. pygmaeus (Leach, 1825). Myotis 37:27-40.

Herd RM, Fenton MB. 1983. An electrophoretic, morphological, and ecological investigation of a putative hybrid zone between Myotis lucifugus and Myotis yumanensis (Chiroptera: Vespertilionidae). Canadian Journal of Zoology 61: 2029-2050.

Jones G, Van Parijs SM. 1993. Bimodal echolocation in pipistrelle bats: Are cryptic species present? Proceedings of the Royal Society of London 251: 119-125.

Kalcounis MC, Brigham RM. 1995. Intraspecific variation in wing loading affects habitat use by Little Brown Bats (Myotis lucifugus). Canadian Journal of Zoology 73:89-95.

Lalis A, Evin A, Denys C. 2009. Morphological identification of sibling species: The case of West African Mastomys (Rodentia: Muridae) in sympatry. Comptes Rendus Biologies 332:480-488.

Landis JR, Koch GG. 1977. The measurement of observer agreement for categorical data. Biometrics 33:159-174.

Lobert B, Lumsden L, Brunner H, Triggs B. 2001. An assessment of the accuracy and reliability of hair identification of south-east Australian mammals. Wildlife Research 28:637-641.

Luszcz TMJ. 2004. Community structure and habitat use by forest-dwelling bats in southwestern British Columbia, [thesis]. Calgary, AB: University of Calgary. 171 p.

Manning RW, Jones JK. 1989. Myotis evotis. Mammalian Species 329:1-5.

Miller RS. 1958. The Munsell system of color notation. Journal of Mammalogy 39: 278-286.

Munsell Color Laboratory. 2000. Munsell soil color charts: Revised washable edition. New Windsor, NY: GretagMacbeth.

Murray KL, Brttzke ER, Robbins LW. 2001. Variation in search-phase calls of bats. Journal of Mammalogy 82:728-737.

Nagorsen DW, Brigham RM. 1993. Bats of British Columbia. Royal British Columbia Museum Handbook. Vancouver, BC: UBC Press. 164 p.

Norberg UM, Rayner JMV. 1987. Ecological morphology and flight in bats (Mammalia: Chiroptera): wing adaptations, flight performance, foraging strategy, and echolocation. Philosophical Transactions of the Royal Society of London B 316:335^27.

O'Farrell MJ. 1999. Blind test for ability to discriminate vocal signatures of the Little Brown Bat Myotis lucifugus and the Indiana Bat Myotis sodalis. Bat Research News 40:44-48.

O'Farrell MJ, miller BW, Gannon WL. 1999. Qualitative identification of free-flying bats using the Anabat detector. Journal of Mammalogy 80: 11-23.

Olson LE, Goodman SM, Yoder AD. 2004. Illumination of cryptic species boundaries in long-tailed shrew tenrecs (Mammalia: Tenrecidae; Microgale), with new insights into geographic variation and distributional constraints. Biological Journal of the Linnean Society 83:1-22.

Olson LE, Gunderson AM, MacDonald SO, Blejwas KM. 2014. First records of Yuma Myotis (Myotis yumanensis) in Alaska. Northwestern Naturalist 95:228-235.

Parkinson A. 1979. Morphologic variation and hybridization in Myotis yumanensis sociabilis and Myotis lucifugus carissima. Journal of Mammalogy 60:489-504.

Patriquin KJ. 2001. Ecology of a bat community in harvested boreal forest in northwestern Alberta, [thesis]. Calgary, AB: University of Calgary. 104 p.

Piaggio AJ, Perkins SL. 2005. Molecular phylogeny of North American long-eared bats (Verpertilionidae: Corynorhinus); inter- and intraspecific relationships inferred from mitochondrial and nuclear DNA sequences. Molecular Phylogenetics and Evolution 37:762-775.

Piaggio AJ, Valdez EW, Bogan MA, Spicer GS. 2002. Systematics of Myotis occultus (Chiroptera: Vespertilionidae) inferred from sequences of two mitochondrial genes. Journal of Mammalogy 83: 386-395.

Rodhouse TJ, Scott SA, Ormsbee PC, Zinck JM. 2008. Field identification of Myotis yumanensis and Myotis lucifugus: A morphological evaluation. Western North American Naturalist 68:437-143.

Russo I-R, Chimimba CT, Bloomer P. 2006. Mitochondrial DNA differentiation between two species of Aethomys (Rodentia: Muridae) from Southern Africa. Journal of Mammalogy 87:545-553.

Sattler T, Bontadina F, Hirzel AH, Arlettaz R. 2007. Ecological niche modelling of two cryptic bat species calls for a reassessment of their conservation status. Journal of Applied Ecology 44: 1188-1199.

Saunders MB, Barclay RMR. 1992. Ecomorphology of insectivorous bats: A test of predictions using two morphologically similar species. Ecology 73: 1335-1345.

Scheuer L. 2002. Brief communication: A blind test of mandibular morphology for sexing mandibles in the first few years of life. American Journal of Physical Anthropology 119:189-191.

Solick DI. 2004. Differences in the morphology and behaviour of western long-eared bats (Myotis evotis) within and between environments [thesis]. Calgary, AB: University of Calgary. 132 p.

Solick DI, Barclay RMR. 2006. Morphological differences among western long-eared myotis (Myotis evotis) populations in different environments. Journal of Mammalogy 87:1020-1026.

Species at Risk Act, SC 2002, c 29, ss 5, 7-12. http:// laws-lois.justice.gc.ca/PDF/S-15.3.pdf.

Szewczak JM. 2000. A consistent acoustic feature to discriminate Myotis species. Bat Research News 41:141.

Van Zyll de jong CG. 1985. Handbook of Canadian mammals, 2. Bats. Ottawa, ON: National Museum of Canada. 212 p.

Warner RM, Czaplewski NJ. 1984. Myotis volans. Mammalian Species 224:1-4.

Weller TJ, Scott SA, Rodhouse TJ, Ormsbee PC, Zinck JM. 2007. Field identification of the cryptic vespertilionid bats Myotis lucifugus and Myotis yumanensis in the Pacific Northwest. Acta Chiropterologica 9:133-147.

Worthington WJ, Barratt E. 1996. A non-lethal method of tissue sampling for genetic studies of chiropterans. Bat Research News 37:1-3.

Wund MA. 2006. Variation in the echolocation calls of Little Brown Bats (Myotis lucifugus) in response to different habitats. The American Midland Naturalist 156:99-108.

Zinck JM, Duffield DA, Ormsbee PC. 2004. Primers for identification and polymorphism assessment of vespertilionid bats in the Pacific Northwest. Molecular Ecology Notes 4:239-242.

Submitted 14 February 2015, accepted 19 August 2015. Corresponding Editor: Robert Hoffman.

Tanya MJ Luszcz (1), Jason MK Rip, Krista J Patriquin, Lydia M Hollis, Joanna M Wilson, and Heather DM Clarke

Department of Biological Sciences, University of Calgary, Calgary, AB T2N 1N4 Canada

Jan Zinck

Department of Biology, Portland State University, Portland, OR 97201 USA

Robert MR Barclay

Department of Biological Sciences, University of Calgary, Calgary, AB T2N 1N4 Canada

(1) Current address: Canadian Wildlife Service, Environment Canada, c/o 102 Industrial Place, Penticton, BC VIA 7C8 Canada.
TABLE 1. Descriptive statistics for measurements of
echolocation-call frequency and external morphology of
Myotis lucifugus (MYLU) and M. yumanensis (MYYU) from the
Skagit Valley, Washington, 2002. Mean [+ or -] standard
deviation, range in parentheses, and sample size are
provided.

                            MYLU                  MYYU
                        (all female)        males and females


ECHOLOCATION
Minimum frequency    39.82 [+ or -] 1.84   47.48 [+ or -] 1.35
(kHz)                   (36.32-42.80)        (44.55 -50.29)
                           n = 12                n = 40
Characteristic       42.31 [+ or -] 3.08   51.86 [+ or -] 2.21
frequency (kHz)         (38.99-50.42)         (45.85-57.26)
                           n = 12                n = 38
Morphology
Mass (g)             5.51 [+ or -] 0.49    4.90 [+ or -] 0.36
                          (4.7-6.2)             (4.0-5.4)
                           n = 13                n = 40
Forearm              36.47 [+ or -] 1.14   34.43 [+ or -] 0.70
length (mm)             (34.40-38.30)         (32.30-35.58)
                           n = 13                n = 40
Wingspan (cm)        23.10 [+ or -] 0.81   22.00 [+ or -] 1.03
                        (21.91-24.76)         (19.16-23.77)
                           n = 13                n = 39
Wing loading         6.04 [+ or -] 0.38    6.18 [+ or -] 0.55
(N/[m.sup.2])            (5.28-6.50)           (5.06-7.99)
                           n = 13                n = 39
Aspect ratio         5.98 [+ or -] 0.30    6.19 [+ or -] 0.36
                         (5.39-6.49)           (5.29-6.86)
                           n = 13                n = 39

                            MYYU                  MYYU
                           females                males

ECHOLOCATION
Minimum frequency    47.37 [+ or -] 1.31   47.87 [+ or -] 1.52
(kHz)                   (44.55-50.29)         (46.56-50.24)
                           n = 31                 n = 9
Characteristic       51.66 [+ or -] 1.95   52.60 [+ or -] 3.04
frequency (kHz)         (45.85-54.33)         (47.85-57.26)
                           m = 30                 n = 8
Morphology
Mass (g)             5.03 [+ or -] 0.23    4.46 [+ or -] 0.38
                          (4.5-5.4)             (4.0-4.9)
                           n = 31                 n = 9
Forearm                 34.50 + 0.67       34.18 [+ or -] 0.78
length (mm)             (32.30-35.58)         (32.95-35.50)
                           n = 31                 n = 9
Wingspan (cm)        22.16 [+ or -] 0.95   21.38 [+ or -] 1.14
                        (19.16-23.77)         (19.29-23.02)
                           n = 31                 n = 8
Wing loading         6.19 [+ or -] 0.50    6.15 [+ or -] 0.76
(N/[m.sup.2])            (5.06-7.11)           (5.65-7.99)
                           n = 31                 n = 8
Aspect ratio         6.15 [+ or -] 0.39    6.32 [+ or -] 0.19
                         (5.29-6.86)           (6.01-6.59)
                           n = 31                 n = 8

TABLE 2. Results of t-tests to determine statistical
differences in measurements of echolocation-call frequency
and external morphology between Myotis lucifugus (MYLU) and
M. yumanetisis (MYYU) from the Skagit Valley, Washington,
2002. We report t-value with degrees of freedom, P-value,
and estimate [+ or -] 95% confidence intervals. Significant
P-values are indicated in bold. Asterisk (*) indicates when
t-tests for samples with unequal variances were performed.

                                               Females only
                     MYLU versus MYYU        MYLU versus MYYU

ECHOLOCATION
Minimum             [t.sub.50] = 15.77      [t.sub.41] = 15.11
frequency (kHz)         P < 0.0001#             P < 0.0001#
                    7.66 [+ or -] 0.98      7.54 [+ or -] 1.01

Characteristic      [t.sub.48] = 11.85     [t*.sub.14.65] = 9.77
frequency (kHz)         P < 0.0001#             P < 0.0001#
                    9.55 [+ or -] 1.62      9.36 [+ or -] 2.04

Morphology           [t.sub.51] = 4.83     [t*.sub.14.17] = 3.35
Mass (g)                P < 0.0001#              P = 0.005
                    0.61 [+ or -] 0.25      0.48 [+ or -] 0.31

Forearm            [t*.sub.15.03] = 6.06   [t*.sub.15.6] = 5.78
length (mm)             P < 0.0001#             P < 0.0001#
                    2.04 [+ or -] 0.72      1.96 [+ or -] 0.72

Wingspan (cm)        [t.sub.50] = 3.50       [t.sub.42] = 3.11
                        P = 0.001#              P = 0.003#
                    1.10 [+ or -] 0.63      0.94 [+ or -] 0.61

Wing loading         [t.sub.50] = 0.86       [t.sub.42] = 0.97
(N/[m.sup.2])            P = 0.39                P = 0.34
                    0.14 [+ or -] 0.33      0.15 [+ or -] 0.31

Aspect ratio         [t.sub.50] = 1.86       [t.sub.42] = 1.43
                         P = 0.07                P = 0.16
                    0.21 [+ or -] 0.22      0.17 [+ or -] 0.25

                           MYYU
                   females versus males

ECHOLOCATION
Minimum              [t.sub.38] = 0.98
frequency (kHz)          P = 0.33
                    0.50 [+ or -] 1.04

Characteristic       [t.sub.36] = 107
frequency (kHz)          P = 0.29
                    0.94 [+ or -] 1.78

Morphology          [t*.sub.9.7] = 4.3
Mass (g)                P = 0.002#
                    0.57 [+ or -] 0.30

Forearm              [t.sub.38] = 1.21
length (mm)              P = 0.23
                    0.32 [+ or -] 0.53

Wingspan (cm)        [t.sub.37] = 1.99
                         P = 0.054
                    0.78 [+ or -] 0.80

Wing loading         [t.sub.37] = 0.17
(N/[m.sup.2])            P = 0.86
                    0.04 [+ or -] 0.45

Aspect ratio       [t*.sub.24.33] = 1.72
                         P = 0.10
                    0.17 [+ or -] 0.20
COPYRIGHT 2016 Society for Northwestern Vertebrate Biology
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Luszcz, Tanya M.J.; Rip, Jason M.K.; Patriquin, Krista J.; Hollis, Lydia M.; Wilson, Joanna M.; Clar
Publication:Northwestern Naturalist: A Journal of Vertebrate Biology
Article Type:Report
Date:Mar 22, 2016
Words:7127
Previous Article:First records of Finescale Triggerfish (Balistes polylepis) and Louvar (Luvarus imperialis) in British Columbia, Canada.
Next Article:Decline of the Pelagic Cormorant in western Queen Charlotte Strait, British Columbia, 1975-2014.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters