Skip to main content

What mechanism of niche segregation allows the coexistence of sympatric sibling rhinolophid bats?

Abstract

Introduction

Our purpose was to assess how pairs of sibling horseshoe bats coexists when their morphology and echolocation are almost identical. We collected data on echolocation, wing morphology, diet, and habitat use of sympatric Rhinolophus mehelyi and R. euryale. We compared our results with literature data collected in allopatry with similar protocols and at the same time of the year (breeding season).

Results

Echolocation frequencies recorded in sympatry for R. mehelyi (mean = 106.8 kHz) and R. euryale (105.1 kHz) were similar to those reported in allopatry (R. mehelyi 105–111 kHz; R. euryale 101–109 kHz). Wing parameters were larger in R. mehelyi than R. euryale for both sympatric and allopatric conditions. Moths constitute the bulk of the diet of both species in sympatry and allopatry, with minor variation in the amounts of other prey. There were no inter-specific differences in the use of foraging habitats in allopatry in terms of structural complexity, however we found inter-specific differences between sympatric populations: R. mehelyi foraged in less complex habitats. The subtle inter-specific differences in echolocation frequency seems to be unlikely to facilitate dietary niche partitioning; overall divergences observed in diet may be explained as a consequence of differential prey availability among foraging habitats. Inter-specific differences in the use of foraging habitats in sympatry seems to be the main dimension for niche partitioning between R. mehelyi and R. euryale, probably due to letter differences in wing morphology.

Conclusions

Coexistence between sympatric sibling horseshoe bats is likely allowed by a displacement in spatial niche dimension, presumably due to the wing morphology of each species, and shifts the niche domains that minimise competition. Effective measures for conservation of sibling/similar horseshoe bats should guarantee structural diversity of foraging habitats.

Introduction

According to the ecomorphological paradigm, species with similar morphology should exhibit similarities in behaviour and ecology[1]. This prediction, however, raises the possibility of competition between such species when they occur in sympatry. Inter-specific competition takes place when two (or more) species with similar ecological requirements consume resources that are limited in supply[2]. Nevertheless, the stable coexistence of competitors will be possible if their respective niches differ sufficiently[3].

Niche differentiation is easy to conceptualise as a consequence of inter-specific competition. However, concrete evidence in support of it is difficult to acquire, because the demonstration of niche differentiation per se does not necessarily indicate anything about the contribution of competition. Removal or demographic response experiments, if adequately designed, may demonstrate a cause-effect relationship between niche differentiation and inter-specific competition[4, 5]. However, these experiments are inappropriate for rare, elusive, K-selected, or endangered species, and currently, non-disruptive, more inductive approaches are the only practical alternatives. These alternative approaches usually compare morphology, behaviour, and ecology of two (or more) species which occur under allopatric and sympatric conditions and assume that any niche displacement is directly associated with competition[6].

Morphologically similar species are numerous among bats[7], and the number is increasing as molecular tools uncover cryptic species-complexes comprising genetically isolated taxa[8, 9]. An excellent illustration is the discovery that the most abundant and best-known European bat species, the pipistrelle, is in fact a cryptic complex of two species: the common pipistrelle (Pipistrellus pipistrellus) and the soprano pipistrelle (P. pygmaeus)[10, 11]. Despite being morphologically almost indistinguishable, ecological data show a marked divergence in ecological requirements[12, 13], contradicting ecomorphological predictions.

Bats depend on flight as their principal means of locomotion, so wing morphology greatly influences foraging behaviour[14]. Additionally, most bats use echolocation to obtain information about their environment, the precision of which depends on the structure of the echolocation signal[15]. Thus, characterisation of bats’ wing morphology and echolocation signals, facilitate inferences about ecological significance. For example, bats with narrow, pointed wings usually emit low-frequency calls and tend to forage in open spaces, whereas bats with broad, rounded wings usually emit high-frequency signals and tend to forage in cluttered space (i.e. forests)[16, 17]. Some species in the families Vespertilionidae and Molossidae, exhibit flexibility in the structure of search-phase echolocation signals[18, 19], enabling them to access a wide variety of habitats[20]. Other species, for example in the families Mormoopidae and Rhinolophidae, emit unique, stereotyped signal structures with little quantitative variation; these bats are more rigid in the use of foraging habitats[21, 22].

Rhinolophids have a highly specialised auditory system that discriminates the modulated echoes of the beating wings of insects (e.g. moths) from unmodulated background echoes[23]. Based on their auditory system and wing morphology, horseshoe bats are classified as narrow-space, flutter-detecting foragers: they are specialised to catch slow-flying insects very close to or within vegetation[15]. Horseshoe bats consist of a single genus, Rhinolophus, with 77 recognised species distributed exclusively in the Old World[7]. Although the group is diverse, its morphological uniformity is striking compared to other families. This uniformity is mirrored in the numerous, morphologically similar, and frequently confused pairs of sympatric species[2426]. Segregation in habitat use and diet have been proposed as major mechanisms for niche differentiation in sympatric bat species[2730]. However, it remains unclear what promotes niche differentiation in sympatric horseshoe bats, as there are few ecological or behavioural studies (but see[22]), and none of potential sibling species (sensu[31]).

The Mehely’s (Rhinolophus mehelyi Matschie, 1901) and the Mediterranean (R. euryale Blasius, 1853) horseshoe bats can be considered as sibling species (i.e. they are morphologically similar and share a recent common ancestor[32, 33]). R. mehelyi and R. euryale are two cave-dwelling species whose distributions overlap extensively in the Mediterranean basin[25]. Radio-tracking studies show that in allopatric conditions both forage in and along forest edges[34, 35]. A preliminary study suggests that in sympatric conditions they tend to segregate foraging habitats with R. euryale found in more dense woodlands[36]. However, the small sample size and the coarse resolution of locations attained by triangulation of that study[36], as well as the lack of information on diet, limit any firm conclusion about niche differentiation. R. mehelyi and R. euryale emit similar echolocation signal structures differing only in the frequency of maximum energy, 107 and 104 kHz, respectively[37, 38]. Although broad differences in frequency may facilitate dietary niche differentiation, frequency differences between both species may be too small to allow any dietary differentiation[22]. So far no diet data have been collected for sympatric populations, but moths (Lepidoptera) are both species’ main prey in allopatry[34, 39]. The question remains, how do sibling horseshoe bat species coexist when echolocation signal structure is unlikely to result in differences in diet and their morphology is almost identical?

To address this question we investigated echolocation, wing morphology, diet, and habitat use of R. mehelyi and R. euryale in sympatric conditions during breeding season in the Iberian Peninsula, and compared our results with previous literature in allopatric populations. To make results comparable we focused on data gathered in the Iberian Peninsula during breeding season. Our aim was to ascertain the mechanism(s) of coexistence. We predicted that dietary niche differentiation is not likely to occur but rather that coexistence stems from spatial niche differentiation, with slight differences in wing parameters as the prominent contributing factor. Lower wing loading and aspect ratio values in R. euryale should make it better adapted to forage in more cluttered space, facilitating habitat partitioning.

Results and discussion

Echolocation and wing parameters

We measured echolocation frequencies and wing parameters from 16 R. mehelyi and 24 R. euryale individuals in sympatry. We found significant differences in resting frequency (RF) and wing parameters between R. mehelyi and R. euryale (multivariate ANOVA: F = 15.9; d.f. = 7, 30; p < 0.001). The interaction between species and sex was not significant (multivariate ANOVA: F = 2.0; d.f. = 7, 30; p = 0.110). R. mehelyi emitted significantly higher RF than R. euryale, 106.8 ± 1.5 kHz (mean ± SD) versus 105.1 ± 1.2 kHz, respectively (Table 1). Mass, forearm length, wingspan, wing area, aspect ratio, and wing loading were greater for R. mehelyi (Table 1). The echolocation frequencies and wing parameters recorded in this study (sympatric condition) are in accordance to previous published data on allopatric conditions[17, 24, 25, 38, 4042].

Table 1 Echolocation and wing parameters in R. mehelyi and R. euryale in sympatry

Diet and prey size

In sympatry, we analyzed 221 faecal pellets collected from 16 R. mehelyi and 31 R. euryale individuals (mean = 4.7 pellets/bat). We identified eight prey categories belonging to five insect orders. The bulk of the diet of R. mehelyi and R. euryale were composed by moths (Lepidoptera), which represented 95% and 85% of average prey volume respectively. Neuroptera (Families Chrysopidae, Myrmeleontidae and Hemerobidae) were important prey for R. euryale, representing more than 12% of the average volume (Table 2). R. mehelyi showed a less diverse diet, consuming three prey categories, whereas R. euryale consumed up to seven categories (Table 2). There were slight but statistically significant differences between species (two-way ANOVA: F = 1.72; d.f. = 7; p = 0.04); R. mehelyi consumed significantly more moths and fewer green lacewings (Chrysopidae) than R. euryale (two-way ANOVA: F = 5.32; d.f. = 1; p = 0.03 and F = 15.70; d.f. = 1; p < 0.001, respectively). We measured 35 moth tarsus fragments from the faeces of 30 R. mehelyi individuals, and 24 fragments from 22 R. euryale individuals. Mean estimated body length of moths did not differ between the two species: R. mehelyi = 14.1 ± 2.1 mm; R. euryale = 12.7 ± 2.1 mm (Mann-Whitney: U = 294.5; d.f. = 57; p = 0.06).

Table 2 Diet of R. mehelyi and R. euryale in sympatry vs allopatry

In allopatry, Salsamendi et al. analysed 215 faecal pellets from 43 R. mehelyi individuals (mean = 5 pellets/bat)[39], and Goiti et al. analized 237 pellets from 43 R. euryale individuals (5.5 pellets/bat)[34]. The diets of R. mehelyi and R. euryale were also composed by moths, which represented 98% of average prey volume for R. mehelyi[39] and 93% of average prey volume for R. euryale[34]. Neuroptera (Families Chrysopidae and Myrmeleontidae) again showed to be important for R. euryale, representing more than 5% of the average volume (Table 2). R. mehelyi showed again a less diverse diet, consuming 4 prey categories[39], whereas R. euryale consumed 6 prey categories[34]. There were significant inter-specific differences in diet (two-way ANOVA: F = 2.37; d.f. = 6; p = 0.02); R. mehelyi consumed more moths and fewer green lacewings (Chrysopidae) than R. euryale (two-way ANOVA: F =9.68; d.f. = 1; p < 0.01 and F = 5.29; d.f. = 1; p = 0.03, respectively). We found no information on the prey sizes consumed in allopatric conditions.

R. mehelyi displaying higher echolocation frequencies, consumed more moths than R. euryale both in sympatric and allopatryc conditions, in accordance with the allotonic frequency hypothesis, which predicts that bats displaying higher-frequency echolocation signals should consume a higher proportion of moths[43]. However, we argue that differences in echolocation frequencies used by R. mehelyi and R. euryale are presumably too small to produce dietary resource partitioning and might not be responsible for the diet divergence[22, 37, 38]. Instead, we propose that observed differences in diet may stem from differences in availability of insect prey in the foraging habitats that bats use (please refer to habitat use subsection below)[44]. For instance, the diet and prey selection in R. euryale varies with seasonal prey availability[34]. Differences in diet of R. mehelyi and R. euryale observed elsewhere can also be explained by local variability of available prey[34, 37, 4447].

The average moth size of R. mehelyi was slightly larger than that for R. euryale (difference = 1.4 mm). This may be attributable to bat’s body size thresholds. Although prey size differences were not statistically significant, the larger overall size of R. mehelyi suggests that it may select larger prey as a response to differences in skull and jaw biometrics[22]. We also propose that differences in the bats’ energy budget relative to prey detection, handling, and consumption may be an alternative explanation. The slightly larger body size of R. mehelyi would reduce manoeuvrability, making larger prey worth hunting.

Habitat use

For the four environmental variables measured in sympatry we found strong associations between habitat type and canopy perimeter (ANOVA: F = 1513.9; d.f. = 1, 9; p < 0.001) and between habitat type and canopy cover (ANOVA: F = 1083.3; d.f. = 1, 9; p < 0.001). We also found a strong correlation between canopy perimeter and canopy cover (Pearson’ s rxy = 0.789; p < 0.001). Rice and corn fields, pastures, and scrublands were the habitat types with the lowest canopy perimeter and cover values, whereas eucalypt plantations had the highest values (Figure 1).

Figure 1
figure 1

Structural complexity of habitat types in sympatry. Mean values and 95% confidence intervals for canopy perimeter (left) and canopy cover (right) among habitat types in sympatric conditions. Habitat types are ranked from lowest to highest values of canopy perimeter and cover as surrogates for structural complexity. Different letters denote significant differences (Dunnett T3 post-hoc tests) between habitat types.

In sympatry, 12 R. mehelyi individuals and 13 R. euryale individuals provided data to feed the CART analysis (Additional file1: Appendix A). Overall, we obtained 542 foraging fixes, 214 for R. mehelyi and 328 for R. euryale. The CART model assigned habitat type as the first splitting variable to distinguish foraging site preferences between R. mehelyi and R. euryale, followed by distance to water, then canopy perimeter, and finally canopy cover (Figure 2). Rice and corn fields were exclusively used by R. mehelyi (foraging probability = 1.00), whereas R. euryale avoided this habitat type (foraging probability = 0.00). Scrublands, broadleaved woodlands, eucalypt plantations, coniferous plantations, and chestnut groves were used almost exclusively by R. euryale (foraging probability = 0.99), whereas both species used dehesas, riparian forests, and olive groves. Neither species used pastures. Generally, foraging probability for both species increased as distance to water decreased; distance to water bodies from foraging sites did not differ between species (ANOVA: F = 0.14; d.f. = 1,1; p = 0.71). The CART model showed an overall high predictive power, correctly classifying 88.3% of the validation fixes (c = 0.86; confidence interval = 0.81–0.91); specifically, it correctly classified 72.8% of R. mehelyi validation fixes and 99.2% for R. euryale.

Figure 2
figure 2

CART model for habitat use by R. mehelyi and R. euryale in sympatry. Classification and regression tree model for differential habitat use by R. mehelyi and R. euryale in sympatric conditions in Villuercas (Spain). The response variables are presence of R. mehelyi and R. euryale and the explanatory variables are habitat type, distance to water, canopy perimeter, and canopy cover. Top node represents training data set (60% of entire data set), non-terminal nodes represent data splits, and terminal nodes represent homogeneous classes. All nodes are labelled with their determining variable’s value/category and the number of foraging fixes for both species in each group (italicised and in brackets), as well as probability of finding a foraging R. mehelyi or R. euryale (in parentheses). An illustration of how to use the CART model: in any site where habitat type is dehesa, olive grove, or riparian forest (follow middle branch in the 1st node habitat type), if distance to water is more than 100 m (follow right branch in the 2nd node distance to water), the resulting probability of this site being used for foraging by R. mehelyi is 0.16, whereas the probability for R. euryale is 0.84.

In allopatry, Salsamendi et al. radio-tracked 25 R. mehelyi individuals[35] and Goiti et al. radio-tracked 15 R. euryale individuals[34]. Overall, 398 foraging fixes were obtained for R. mehelyi, whereas 373 were obtained for R. euryale[34, 35]. R. mehelyi showed preference for eucalypt plantations, riparian forests and broadleaved woodlands[35], whereas R. euryale showed preference for hedgerows, broadleaved woodlands and isolated trees[34]. None of the species used pastures for foraging.

In allopatric conditions, we found no differences in the use of canopy cover values between R. mehelyi and R. euryale (Tukey post-hoc comparison: p = 0.113; Figure 3), but there were differences in the use of canopy perimeter values between R. mehelyi and R. euryale (Tukey post-hoc comparison: p = 0.004). However, in sympatric conditions we found significant inter-specific differences in both canopy cover and canopy perimeter values (Tukey post-hoc comparisons: p < 0.001 and p < 0.001, respectively; Figure 3).

Figure 3
figure 3

Niche shift between R . mehelyi and R . euryale in allopatry vs sympatry. Distribution box-plots for canopy perimeter (left) and canopy cover (right) values used by R. mehelyi and R. euryale in allopatric and sympatric conditions; significance is also provided (*** = p<0.001; ** = p<0.01; ns = not significant). Bold lines indicate median and boxes encompass interquartile range (IQR, 25%–75%). T-bars encompass values within 1.5 IQR from median and the circles represent outliers.

Our results show that, in sympatry, both species exhibit similar but not completely identical foraging habitat preferences. We found that exclusive foraging habitats for R. euryale included broadleaved woodlands, scrublands, eucalypt and coniferous plantations, and chestnut groves, where R. mehelyi was rarely encountered; in contrast, exclusive foraging habitats for R. mehelyi included rice and corn fields. In a semiarid ecosystem as in the Mediterranean, where water availability is limited, habitats associated with water likely support high insect abundances[48, 49]. Crops requiring irrigation or flooding, such as rice and corn fields, may offer alternative foraging sites when other habitats dry up during summer[35]. We note that while both species foraged close to water bodies, where moths are readily found, foraging sites diverged in structural complexity.

Foraging behaviour

Stereotyped foraging behaviour of both species was characterised by continuous back-and-forth flights close to vegetation; R. mehelyi flew at 0.5–1.5 m from to tree canopies, linear structures, and isolated trees, whereas R. euryale flew within 0.5 m, repeatedly plunging in and even diving through the branches. We also observed R. mehelyi flying continuously back-and-forth in open environments, close (30-50 cm) to ground vegetation, whereas R. euryale was never recorded in open environments.

Mean straight-line distance from the roost to foraging sites was significantly larger in R. mehelyi than in R. euryale; 19.2 ± 7.0 km and 4.8 ± 2.4 km, respectively (Mann-Whitney: U = 12.0; d.f. = 23; p < 0.001). Maximum individual flight distance for R. mehelyi was 29.1 km and 10.1 km for R. euryale. There was no inter-specific difference in foraging home range (R. mehelyi = 242 ± 341 ha vs. R. euryale = 153 ± 316 ha; Man-Whitney: U = 53.5; d.f. = 22; p = 0.29). Mean and maximum commuting distances for R. euryale are similar to previously published data[36, 5053]. However, mean and maximum distances to foraging sites were nearly five-fold greater for R. mehelyi[35, 36]. This contrasts with previous studies of R. mehelyi and other congeners in which commuting distances were short, consistent with the slow, butterfly-like flight of rhinolophids[17, 35, 36, 54]. R. mehelyi is theoretically better-adapted to commute longer distances than R. euryale, as its wing parameters should make it a more efficient flyer for long distances[17, 55]. Indeed, Rainho and Palmeirim[56] reported commuting distances for R. mehelyi similar to those we found (up to 22 km), confirming that these bats are capable of flying long distances to foraging sites. The commuting distances we recorded are likely a consequence of landscape configuration, where favoured foraging areas are particularly distant from the roost.

Niche breadth and overlap

Habitat breadth did not differ between R. mehelyi (B = 2.81 ± 0.56) and R. euryale (B = 4.1 ± 0.77; t = -0.63; p = 0.55, randomisation test after 10000 iterations). However, the magnitude of intra-specific habitat overlap was significantly larger in R. mehelyi (FT = 0.47 ± 0.38) than in R. euryale (FT = 0.33 ± 0.35; t = -2.14; p = 0.01). The magnitude of intra-specific habitat overlap for each species was significantly larger than the inter-specific overlap (FT = 0.21 ± 0.27; t = 2.66; p = 0.009, and t = 5.64; p < 0.001, respectively). Trophic niche breadth was larger for R. mehelyi than R. euryale, B = 1.34 ± 0.59 and B = 1.12 ± 0.13, respectively (t = -2.2; p = 0.03). Magnitude of intra-specific trophic overlap was lower in R. euryale than in R. mehelyi, FT = 0.88 ± 1.4 versus FT = 0.93 ± 0.19, respectively (t = 3.39; p < 0.001). Inter-specific trophic overlap, FT = 0.89 ± 0.17, was significantly lower than the intra-specific overlap of R. mehelyi (t = 2.3; p = 0.03) but significantly higher than that of R. euryale (t = -0.9; p = 0.02).

Niche segregation

Although differences exist in the diet of R. mehelyi and R. eurale, the bulk of the diet of both species in sympatric and allopatric conditions consist of moths[34, 39, 46, 47]. Moreover, in sympatry we found that intra-specific trophic overlap of R. euryale was lower than inter-specific trophic overlap, making the species’ coexistence through a dietary niche dimension unstable[57]. Although molecular identification of moths eaten is required prior to discarding any functional difference of diet patterns e.g.[58], segregation in foraging habitats appears to be a more likely scenario for niche segregation of these bat species in sympatry. Subtle differences in echolocation frequencies between both species seem to be unlikely to facilitate dietary partitioning. Divergences observed in diet may be explained as a consequence of differential prey availability among foraging habitats.

As narrow-space, flutter-detecting foragers, R. mehelyi and R. euryale are both adapted to forage in structurally complex, highly cluttered environments[15]. Based on the species’ similarities in wing morphology and echolocation parameters, their respective niches should largely overlap in relation to habitat use. In allopatric populations of southern Iberian Peninsula, R. mehelyi forages not only in dehesas but also in riparian forests, broadleaved woodlands, and eucalypt plantations[35]. In areas with an Atlantic climate in the northern Iberian Peninsula, R. euryale forages not only in habitats with high structural complexity but also semi-open habitats, predominantly hedgerows, woodland edges and isolated trees in meadows[34]. However, foraging habitats reported here (i.e. under sympatric conditions) for R. mehelyi, seem to be more restricted to open spaces, whereas R. euryale uses more cluttered ones. This is best illustrated in Figure 3, where there is a significant niche displacement in the use of canopy cover and canopy perimeter values from allopatric to sympatric conditions. All of this suggests that the potential niches related to habitat use are likely broader than those we observed; i.e. under sympatric conditions, the two species occupy narrower habitat niches, most likely a consequence of inter-specific interactions that lead to a niche displacement. This niche displacement is conditioned by subtle differences in wing morphology and shifts the niche domains that may limit competition between both species.

Conclusions

This study shows that R. mehelyi and R. euryale exhibit similar ecologies in sympatry yet showing subtle but significant differences. The most pronounced ecological difference is their divergent use of foraging habitats that contrasts with the low divergence in diet. This is reflected by the low magnitude of inter-specific habitat overlap compared to inter-specific trophic overlap.

Differences in echolocation call frequencies between R. mehelyi and R. euryale are presumably too small to allow any dietary partitioning and, accordingly, there were no inter-specific differences in the size of moth consumed. Diet divergences between R. mehelyi and R. euryale in both sympatric and allopatric conditions may stem from differences in prey availability among habitats used for foraging. As a general pattern, there were virtually no differences in the use of foraging habitats between R. mehelyi and R. euryale in allopatric conditions. However, under sympatric conditions, both species display a shift in the use of foraging habitats: R. euryale forages in more complex environments characterized by higher values of canopy cover and canopy perimeter, whereas R. mehelyi forages in less cluttered/more open environments.

Differences in the use of foraging habitats, rather than differences in diet, seem to be the main mechanisms for resource partitioning between R. mehelyi and R. euryale, as a consequence of slight differences in wing morphology. Coexistence between sympatric sibling horseshoe bats is likely allowed by a displacement in spatial niche dimension, presumably due to morphological features of each species, and shifts the niche domain that minimise competition. Effective measures for conservation of sibling/similar horseshoe bats should guarantee structural diversity of foraging habitats so coexisting species have enough “space” to segregate their respective niches.

Materials and methods

Literature review

To make results comparable between sympatric (this study) and allopatric conditions, literature review for allopatric data followed the next criterions: 1) study period should be the same time of the year (i.e. breeding season); 2) study areas should be geographically as close as possible (preferably the Iberian Peninsula); 3) protocols used to measure echolocation, wing parameters, and diet should be similar and results comparable; and 4) radio-tracking protocols must be identical (e.g. homing-in) and sampling effort comparable. Following these criterions, our literature review outputted seven publications for echolocation and wing parameters[17, 24, 25, 38, 4042], whereas for diet and habitat use we only found three publications[34, 35, 39]. The following subsections describe the protocols used for data and sample collection in sympatric conditions. Protocols used to data and sample collection in allopatry will be added when needed.

Study area and bat capture

Fieldwork was conducted during breeding season (June–July) in 2007 at Sierra de Las Villuercas mountain range, in Extremadura, Spain (UTM 30S 2924 4359), south-western Iberian Peninsula. In the area, maternity colonies of R. mehelyi and R. euryale roost together in three adjacent, disused train tunnels. Emergence video-counts revealed that in 2007 at least 190 R. mehelyi/R. euryale individuals roosted in the three tunnels altogether; the small inter-specific differences in echolocation frequency precluded species identification from video-recordings (G. Schreur and O. De Paz, unpublished data). Common movements between tunnels (maximum distance of 6.8 km) by radio-tracked individuals led us to consider the area’s bats as a single population.

Bats were caught in harp-traps as they entered the roost after foraging. Once identified, sexed, and aged, we placed them individually in cloth bags until sampling was carried out. Capture and handling protocols met the guidelines for treatment of animals in research and teaching[59], were approved by the Regional Council (license number: 0532041 PC 120), and met Spanish legal requirements. Bats were released at the roost after sampling. To minimise stress, retention time never exceeded 90 minutes. We discriminated between the two species based on the shape of the lancet and connecting process of the noseleaf. The lancet in R. euryale is essentially triangular, whereas in R. mehelyi it is notably concave laterally and very narrow distally. The connecting process is narrower and more pointed in R. euryale[25]. We restricted our analyses to adults, which were distinguished from juveniles by trans-illumination of the cartilaginous epiphyseal plates in the phalanges[60].

Echolocation and wing parameters

Echolocation calls were recorded from bats held 30 cm from the microphone of a Pettersson D-980 ultrasonic detector (Pettersson Elektronik AB, Sweden). Signals were recorded into a laptop with ultrasound analysis software (BatSound 4.0; Pettersson Elektronik AB, Sweden). Because bats were held motionless, calls were not affected by Doppler shift; thus, the emitted frequency corresponded to the resting frequency (RF)[61]. Three-second samples were recorded at a rate of 448 kHz and time-expanded (10×). The resulting sequences of 30 seconds were then analysed using a sampling frequency of 44.1 kHz and a 1024-pt FFT[38]. We chose a random sample of ten echolocation signals from each individual, and measured the frequency of maximum energy from the power spectrum of a 10-ms portion of the constant frequency component of each call. Because horseshoe bats start emitting at a lower frequency before reaching the final RF level, we rejected starting signals for the analyses[41]. Starting signals were rejected manually in a spectrogram window after visual inspection. We calculated the mean frequency for the ten signals of each individual and used this mean for comparisons.

For each bat, we measured forearm length and body mass to the nearest 0.05 mm and 0.1 g, respectively. Wing morphology measurements included wingspan, wing area, aspect ratio, and wing loading. Wing area included the combined area of both wings, the entire tail membrane, and the body areas between the wings, excluding the head[17]. Aspect ratio is the square or the wingspan divided by the wing area and is related to energy efficiency. A higher aspect ratio corresponds to lower energy loss during flight. Wing loading is the weight of the bat divided by the area of its flight membrane and is correlated with flight speed. As wing loading increases, so too does the speed required to fly[17]. These variables were determined from wing tracings (to the nearest 1mm) on graph paper of the extended left wing. Tracings were scanned at 600 dpi and incorporated into a GIS (Arc View 3.2, ESRI, California, USA) for surface calculations.

To test for inter-specific differences in size and echolocation frequencies in sympatry we used multivariate ANOVA on arcosine-transformed measurements. Sex was introduced as a main factor in the analyses. We were unable to perform any statistical analysis with data on allopatry, since literature mostly provided mean values[17, 24, 25, 38, 4042].

Diet

We collected faecal samples, which were air-dried prior to analysis. Where possible, we randomly selected a minimum of five pellets from each individual; however, some bats did not excrete this many while handled (90 min). Pellets were soaked in water for 20 min and then teased apart with dissecting needles under a microscope. Based on arthropod fragments, we identified prey remains to the lowest taxonomic level possible using identification keys[6264] and a reference collection. We estimated the percentage volume of each identifiable prey category visually to the nearest 5% for each pellet and then calculated a % for each individual[64]. We also estimated the relative sizes of moths consumed by R. mehelyi and R. euryale in sympatry. We extracted fragments of moth tarsus from the samples and measured the length of the first segment (LFS, from the joint with the second segment to the base of the claw) under a compound microscope to the nearest μm (magnification 32x). For specimens in our reference collection (n = 97), we measured moth body lengths (BL, excluding antennae and cerci) and LFS to obtain a regression equation (BL = 30.6LFS + 5.6; r = 0.95; F = 915.1; d.f. = 96; p < 0.001). Using this regression equation, we estimated the BL of moths consumed by bats based on LFS lengths in faecal samples.

To test for inter-specific differences in diet in sympatry and allopatry, we used a permutation test for two-way ANOVA between the proportions of prey consumed by both species. The significance of the F-statistic was obtained comparing the observed F-value to a reference distribution of F-values generated by 10000 permutations of the observed values between species[65]. Differences in estimated BL of consumed moths were tested using the Mann-Whitney U test.

Radio-tracking

After clipping the fur from the mid sagittal dorsal surface, radio-transmitters (0.45 g; Pip II, Biotrack Ltd. Dorset, UK) were attached using surgical cement (Skinbond, Smith and Nephew, Largo, Florida, USA). We tagged 15 R. mehelyi and 16 R. euryale; from these we obtained radio-tracking data on 12 R. mehelyi and 13 R. euryale (Additional file1: Appendix A). Radio-transmitter mass never exceeded 5% of the bat’s body mass[66]: percentage for the smallest R. mehelyi was 3.3% and for the smallest R. euryale was 4.0%. Before release at the roost, we checked whether bats were able to remove the transmitter and whether its position interfered with flight. The transmitter eventually fell off after 11-23 days (personal observation).

Bats were radio-tracked simultaneously by up to three mobile teams equipped with radio-receivers (1000-XRS, Wildlife Materials Inc., Carbondale, Illinois, USA, and FT290Rii, Andreas Wagener Telemetrienalegen, Köln, Germany) and hand-held three-element Yagi antennas. Mobile teams tracked bats from vehicles and on foot. Each night up to three observers equipped with radio-receivers located at different stationary vantage points guided the mobile teams to bats’ activity areas. The searching scope of the stationary observers encompassed almost all the study area, and they were able to detect the transmitter at a maximum distance of 10 km. Mobile and stationary trackers were coordinated using transceivers (VX-110, Yaesu Musen Co. Ltd., Japan) and cellular phones. Two or three bats were simultaneously tracked each night. Whenever possible, tracking was conducted continuously during the entire night. Bats were tracked by the “homing-in” technique, which involved following the bats to their activity areas as closely as possible and identifying their commuting routes and foraging areas in situ[67]. Bat locations were taken every 10 minutes to minimise spatiotemporal autocorrelation, because bats were able to commute to distant areas and had access to all habitat types with this time interval[67]. Subsequently, locations were transferred into a GIS database. Locations were recorded when bats were active and resting, but only active locations (hereafter referred to as foraging fixes) were used in our analyses. Individual foraging home ranges were determined by minimum convex polygons (MCPs) using individual foraging fixes. The colonial foraging range was determined as the MCP encompassing all foraging fixes of all individuals and was used to define the study area. This protocol is identical to that used by Salsamendi et al.[35] and Goiti et al.[34] to obtain radio-tracking locations in allopatric populations of R. mehelyi and R. euryale.

Habitat use

For habitat use analyses, we included four environmental variables following Salsamendi et al.[35]: habitat type, canopy perimeter, canopy cover, and distance to water. Habitat types were mapped from local forest inventories and through photo-interpretation of ortho-photographs at a resolution of 4 pixels/m2. The resulting habitat-map was ground-truthed in the field. We identified ten habitat types: 1) pastures; 2) rice and corn fields; 3) scrublands, low vegetation dominated by Mediterranean maquis, interspersed with evergreen oaks; 4) dehesas, semi-natural savannah-like oak woodlands dominated by Quercus rotundifolia and Q. suber; 5) olive groves; 6) chestnut groves; 7) eucalypt plantations; 8) coniferous plantations; 9) riparian forests, woods along river banks dominated by Populus sp. and Alnus glutinosaPopulus sp. plantations, tree lines, and isolated trees near water courses were also included; and 10) broadleaved woodlands. Each foraging fix was assigned to a habitat type.

Canopy perimeter and canopy cover were quantified using GIS tools and used as an indication of habitat structure. We followed the next protocol: i) we transformed the original 4-pixels/m2-resolution ortho-photographs into 1-pixel/m2-resolution grey-scale geo-referenced images, in which each pixel had a value ranging from 0 to 255; ii) pixels representing trees (values from 1 to 100) were identified and selected at the attribute table, and a Boolean layer was generated from this selection (cut of levels were set after checking the range of values by visual inspection); iii) simultaneously, we modelled the study area as a randomly generated grid-layer (1 hectare per grid) covering the colonial MCP (41555 grids in total); and iv) we computed a geometric intersection between the tree (Boolean) layer and the grid layer to obtain a third layer, from which we obtained values of canopy perimeter and cover for each grid-cell. Thus, canopy cover was based on the sum of the areas of the pixels representing trees and indicated the percentage of tree cover for each 1-hectare grid-cell. Tree pixels connected to each other were aggregated into blocks and the perimeter of each block was measured. Canopy perimeter was based on the sum of the perimeters of the blocks and indicated the amount of edge habitat for each 1-hectare grid-cell. Each foraging fix in this study (sympatric condition) as well as those obtained by Salsamendi et al.[35] for R. mehelyi and by Goiti et al.[34] for R. euryale (both in allopatry), were assigned to a single grid cell canopy cover and perimeter values. Differences in canopy cover and canopy perimeter values between R. mehelyi and R. euryale in sympatric and allopatric conditions were assessed on square root-transformed values using one-way ANOVAs.

Onto a separate layer, we identified natural and artificial water bodies (rivers, wetlands, canals, and ponds for irrigation and livestock) within the colonial MCP from photo-interpretation of ortho-photographs and observations in the field. Minimum width for rivers was set to 1 m and minimum area for water bodies to 4 m2 to be included in the analyses. The corresponding distance from foraging fixes to the nearest water source was included in the analyses and was used as an indication of water availability. We used one-way ANOVA to determine if there was an association between canopy perimeter, canopy cover and habitat type. Repeated-measures ANOVA was used to test for inter-specific differences in mean canopy perimeter, canopy cover, and distance to water.

We used Classification and Regression Tree (CART) analysis to model differences in foraging habitat preferences between R. mehelyi and R. euryale in sympatry. CART is a statistical method to construct binary trees promoted as a strong tool for predictive modelling[68]. Splits are based on cut-off levels of the predictors, which produce maximum separation between two data subgroups and minimum variability within them with respect to the outcome. These splits are determined from a search of all possible cut-off points, and only significant splits (p< 0.05) were used in the final model[69]. Foraging fixes of both species were randomly divided, with 60% used as training data to select the best model and 40% retained for model validation. The response variable was R. mehelyi/R. euryale foraging fix, whereas the habitat type, canopy perimeter, canopy cover, and distance to water were explanatory variables. To avoid over-fitting the data, we used a pruning methodology that determined a nested sequence of sub-trees by recursively removing the least-important splits from the full tree. We selected the optimal tree based on cross-validation[70]. We evaluated the predictive ability of the CART model by calculating the c statistic and 95% confidence interval from a logistic regression model in the validating model. By cross-tabulation, we calculated the percentage of foraging fixes correctly predicted for each species in the validation sample.

Niche breadth and overlap

We computed habitat and trophic niche breadth by means of Levins’ index B[71]: Bi = 1/(Σpir2), where Bi is the Levins’ index of individual i. For habitat niche breadth, pir is the proportion of r habitat type used by individual i, whereas for trophic niche breadth, pir is the proportion of r prey category consumed by individual i. Levins’ index ranges from 1 (one habitat/prey category used) to n (total number of habitat/prey categories). We also computed intra-specific and inter-specific habitat and dietary niche overlap using the Freeman-Tukey index[72]: FTij = Σ (pir · pjr)1/2, where FTij is the Freeman-Tukey measure of niche overlap between individuals i and j. For habitat niche overlap, pir and pjr are the proportions of r habitat types used by individuals i and j, whereas for trophic niche overlap, pir and pjr are the proportions of r prey category consumed by individuals i and j. This measure of overlap ranges from 0 (no habitat/prey used in common) to 1 (complete overlap in habitat/prey used). For intra-specific niche breadth and overlap, we calculated the measure for every possible pair of individuals within one species. For inter-specific niche overlap, we calculated the measure for every possible pair of individuals of both species. Differences in both niche breadths and overlaps relied on randomisation procedures to avoid the statistical pseudo-replication inherent in pairwise comparisons[73]. To test for differences between niche breadths and overlap means, we carried out two-sample randomisation tests. The significance of the t-statistic was obtained by comparing the observed t-value to a reference distribution of t-values generated by 10000 permutations of the observed values between the two species. We computed niche and breadth overlap measures for sympatric data on habitat use and diet.

Statistical analyses were performed using SAS v.9.2 (SAS Institute Inc., Cary, USA), except CART analysis was conducted using R software v.2.14 (R Foundation for Statistical Computing, Vienna, Austria). For all tests we assumed a significant result when alpha < 0.05.

Abbreviations

RF:

Resting frequency: echolocation call frequency of maximum energy emitted by horseshoe bats when resting

BL:

Body length measure of a moth’s body length excluding antennae and cerci

B:

Levins’ index (B) computed to calculate habitat and trophic niche breadth

FT:

Freeman-tukey index (FT) computed to calculate habitat and trophic overlap

GIS:

Geographic information system

MCP:

Minimum convex polygon

CART analysis:

Classification and regression tree analysis.

References

  1. Wainwright PC, Reilly SM: Ecological Morphology: Integrative Organismal Biology. 1994, Chicago: University of Chicago Press

    Google Scholar 

  2. Leibold MA, McPeek MA: Coexistence of the niche and neutral perspectives in community ecology. Ecology. 2006, 87: 1399-1410. 10.1890/0012-9658(2006)87[1399:COTNAN]2.0.CO;2.

    Article  PubMed  Google Scholar 

  3. Levine JM, HilleRisLambers J: The importance of niches for the maintenance of species diversity. Nature. 2009, 461: 254-257. 10.1038/nature08251.

    Article  CAS  PubMed  Google Scholar 

  4. Suding KN: The effects of gap creation on competitive interactions: separating changes in overall intensity from relative rankings. Oikos. 2001, 94: 219-227. 10.1034/j.1600-0706.2001.940202.x.

    Article  Google Scholar 

  5. Diaz S, Symstad AJ, Chapin FS, Wardle DA, Huenneke LF: Functional diversity revealed by removal experiments. Trends Ecol Evol. 2003, 18: 140-146. 10.1016/S0169-5347(03)00007-7.

    Article  Google Scholar 

  6. Hutchinson GE: Concluding remarks. Cold Spring Harb Sym. 1957, 22: 415-427. 10.1101/SQB.1957.022.01.039.

    Article  Google Scholar 

  7. Simmons NB: Order Chiroptera. Mammal Species of the World: A Taxonomic and Geographic Reference. Edited by: Wilson DE, Reeder DM. 2005, Baltimore: Johns Hopkins University Press, 312-529. Volume 1, 3

    Google Scholar 

  8. Mayer F, Dietz C, Kiefer A: Molecular species identification boosts bat diversity. Front Zool. 2007, 4: 4-10.1186/1742-9994-4-4.

    Article  PubMed Central  PubMed  Google Scholar 

  9. Ruedi M, McCracken GF: Phylogeographic analysis of bats. Ecological and Behavioral Methods for the Study of Bats. Edited by: Kunz TH, Parsons S. 2009, Baltimore: Johns Hopkins University Press, 739-756. 2

    Google Scholar 

  10. Jones G, van Parijs SM: Bimodal echolocation in pipistrelle bats—are cryptic species present?. P Roy Soc Lond B Bio. 1993, 251: 119-125. 10.1098/rspb.1993.0017.

    Article  CAS  Google Scholar 

  11. Barrat EM, Deaville R, Burland TM, Bruford MW, Jones G, Racey PA, Wayne RK: DNA answers the call of pipistrelle bat species. Nature. 1997, 387: 138-139.

    Article  Google Scholar 

  12. Davidson-Watts I, Walls S, Jones G: Differential habitat selection by Pipistrellus pipistrellus and Pipistrellus pygmaeus identifies distinct conservation needs for cryptic species of echolocating bats. Biol Conserv. 2006, 133: 118-127. 10.1016/j.biocon.2006.05.027.

    Article  Google Scholar 

  13. Sattler T, Bontadina F, Hirzel AH, Arlettaz R: Ecological niche modelling of two cryptic bat species calls for a reassessment of their conservation status. J Appl Ecol. 2007, 44: 1188-1199. 10.1111/j.1365-2664.2007.01328.x.

    Article  Google Scholar 

  14. Swartz SM, Freeman PW, Stockwell EF: Ecomorphology of bats: comparative and experimental approaches relating structural design to ecology. Bat Ecology. Edited by: Kunz TH, Fenton MB. 2003, Chicago: University of Chicago Press, 257-300.

    Google Scholar 

  15. Schnitzler H-U, Moss CF, Denzinger A: From spatial orientation to food acquisition in echolocating bats. Trends Ecol Evol. 2003, 18: 366-394. 10.1016/S0169-5347(03)00096-X.

    Article  Google Scholar 

  16. Aldridge HDJN, Rautenbach IL: Morphology, echolocation and resource partitioning in insectivorous bats. J Anim Ecol. 1987, 56: 763-778. 10.2307/4947.

    Article  Google Scholar 

  17. Norberg UM, Rayner JMV: Ecological morphology and flight in bats (Mammalia; Chiroptera): wing adaptations, flight performance, foraging strategy and echolocation. Philos T Roy Soc B. 1987, 316: 335-427. 10.1098/rstb.1987.0030.

    Article  Google Scholar 

  18. Surlikke A, Moss CF: Echolocation behaviour of big brown bats, Eptesicus fuscus, in the field and the laboratory. J Acoust Soc Am. 2000, 108: 2419-2429. 10.1121/1.1315295.

    Article  Google Scholar 

  19. Guillén-Servent A, Ibañez C: Unusual echolocation behavior in a small molossid bat, Molossops temminckii, that forages near background clutter. Behav Ecol Sociobiol. 2007, 61: 1599-1613. 10.1007/s00265-007-0392-4.

    Article  Google Scholar 

  20. Kalko EKV, Schnitzler H-U: Plasticity in echolocation signals of European pipistrelle bats in search flight: implications for habitat use and prey detection. Behav Ecol Sociobiol. 1993, 33: 415-428.

    Article  Google Scholar 

  21. Macías S, Mora EC, García A: Acoustic identification of mormoopid bats: a survey during the evening exodus. J Mammal. 2006, 87: 324-330. 10.1644/05-MAMM-A-124R1.1.

    Article  Google Scholar 

  22. Jacobs DS, Barclay RMR, Walker MH: The allometry of echolocation call frequencies of insectivorous bats: why do some species deviate from the pattern?. Oecologia. 2007, 152: 583-594. 10.1007/s00442-007-0679-1.

    Article  PubMed  Google Scholar 

  23. von der Emde G, Schnitzler H-U: Classification of insects by echolocating greater horseshoe bats. J Comp Physiol A. 1990, 167: 423-430.

    Article  Google Scholar 

  24. Bogdanowicz W: Phenetic relationships among bats of the family Rhinolophidae. Acta Theriol. 1992, 37: 213-240.

    Article  Google Scholar 

  25. Csorba G, Ujhelyi P, Thomas N: Horseshoe Bats of the World. 2003, Shropshire: Alana Books

    Google Scholar 

  26. Stoffberg S, Jacobs DS, Mackie IJ, Matthee CA: Molecular phylogenetics and historical biogeography of Rhinolophus bats. Mol Phylogenet Evol. 2010, 54: 1-9. 10.1016/j.ympev.2009.09.021.

    Article  PubMed  Google Scholar 

  27. Arlettaz R: Habitat selection as a major resource partitioning mechanism between two sympatric sibling bat species Myotis myotis and Myotis blythii. J Anim Ecol. 1999, 68: 460-471. 10.1046/j.1365-2656.1999.00293.x.

    Article  Google Scholar 

  28. Siemers BM, Schnitzler H-U: Echolocation signals reflect niche differentiation in five sympatric congeneric bat species. Nature. 2004, 429: 657-661. 10.1038/nature02547.

    Article  CAS  PubMed  Google Scholar 

  29. Zhang L, Liang B, Parsons S, Wei L, Zhang S: Morphology, echolocation and foraging behaviour in two sympatric sibling bats (Tylonycteris pachypus and Tylonycteris robustula) (Chiroptera: Vespertilionidae). J Zool. 2007, 271: 344-351.

    Google Scholar 

  30. Jacobs DS, Barclay RMR: Niche differentiation in two sympatric sibling bat species, Scotophilus dinganii and Scotophilus mhlanganii. J Mammal. 2009, 90: 879-887. 10.1644/08-MAMM-A-235.1.

    Article  Google Scholar 

  31. Mayr E: Evolution and the Diversity of Life. 1976, Cambridge: Harvard University Press

    Google Scholar 

  32. Guillén A, Francis CM, Ricklefs RE: Phylogeny and biogeography of the horseshoe bats. Horseshoe Bats of the World. Edited by: Csorba G, Ujhelyi P, Thomas N. 2003, Shropshire: Alana Books, xii-xxiv.

    Google Scholar 

  33. Zhou Z-M, Guillent-Servent A, Lim BK, Eger JL, Wang Y-X, Jiang X-L: A new species from southwestern China in the Afro-Paleartic lineage of the horseshoe bats (Rhinolophus). J Mammal. 2009, 90: 57-73. 10.1644/08-MAMM-A-048.1.

    Article  Google Scholar 

  34. Goiti U, Garin I, Almenar D, Salsamendi E, Aihartza J: Foraging by Mediterranean horseshoe bats (Rhinolophus euryale) in relation to prey distribution and edge habitat. J Mammal. 2008, 89: 493-502. 10.1644/07-MAMM-A-054R2.1.

    Article  Google Scholar 

  35. Salsamendi E, Arostegui I, Aihartza J, Almenar D, Goiti U, Garin I: Foraging ecology in Mehely’s horseshoe bats: influence of habitat structure and water availability. Acta Chiropterol. 2012, 14: 121-132. 10.3161/150811012X654330.

    Article  Google Scholar 

  36. Russo D, Almenar D, Aihartza J, Goiti U, Salsamendi E, Garin I: Habitat selection in sympatric Rhinolophus mehelyi and R. euryale (Mammalia: Chiroptera). J Zool. 2005, 266: 327-332. 10.1017/S0952836905006990.

    Article  Google Scholar 

  37. Salsamendi E, Aihartza J, Goiti U, Almenar D, Garin I: Echolocation calls and morphology in the Mehely’s (Rhinolophus mehelyi) and Mediterranean (R. euryale) horseshoe bats: implications for resource partitioning. Hystrix. 2005, 16: 149-158.

    Google Scholar 

  38. Russo D, Mucedda M, Bello M, Biscardi S, Pidinchelda E, Jones G: Divergent echolocation call frequencies in insular rhinolophids (Chiroptera): a case of character displacement?. J Biogeogr. 2007, 34: 2129-2138. 10.1111/j.1365-2699.2007.01762.x.

    Article  Google Scholar 

  39. Salsamendi E, Garin I, Almenar D, Goiti U, Napal M, Aihartza J: Diet and prey selection in Mehely’s horseshoe bat Rhinolophus mehelyi (Chiroptera, Rhinolophidae) in the south-western Iberian Peninsula. Acta Chiropterol. 2008, 10: 279-286. 10.3161/150811008X414854.

    Article  Google Scholar 

  40. Russo D, Jones G, Mucceda M: Influence of age, sex and body size on echolocation calls of the Mediterranean and Mehely’s horseshoe bats, Rhinolophus euryale and R. mehelyi (Chiroptera: Rhinolophidae). Mammalia. 2001, 65: 429-436.

    Article  Google Scholar 

  41. Siemers BM, Beedholm K, Dietz C, Dietz I, Ivanova T: Is species identity, sex, age or individual quality conveyed by echolocation call frequency in European horseshoe bats?. Acta Chiropterol. 2005, 7: 259-274. 10.3161/1733-5329(2005)7[259:ISISAO]2.0.CO;2.

    Article  Google Scholar 

  42. Dietz C, Dietz I, Siemers BM: Wing measurement variation in the five European horseshoe bat species (Chiroptera: Rhinolophidae). J Mammal. 2006, 87: 1241-1251. 10.1644/05-MAMM-A-299R2.1.

    Article  Google Scholar 

  43. Rydell J, Jones G, Waters D: Echolocating bats and hearing moths: who are the winners?. Oikos. 1995, 73: 419-424. 10.2307/3545970.

    Article  Google Scholar 

  44. Clare EL, Barber BR, Sweeney BW, Hebert PDN, Fenton BM: Eating local: influences of habitat on diet of little brown bats (Myotis lucifugus). Mol Ecol. 2011, 20: 1772-1780. 10.1111/j.1365-294X.2011.05040.x.

    Article  CAS  PubMed  Google Scholar 

  45. Goiti U, Aihartza J, Garin I: Diet and prey selection in the Mediterranean horseshoe bat Rhinolophus euryale (Chiroptera, Rhinolophidae) during the pre-breeding season. Mammalia. 2004, 68: 397-402.

    Article  Google Scholar 

  46. Sharifi M, Hemmati Z: Food of Mehely’s horseshoe bat Rhinolophus mehelyi in a maternity colony in western Iran. Myotis. 2001, 39: 17-20.

    Google Scholar 

  47. Sharifi M, Hemmati Z: Variation in the diet of Mehely’s horseshoe bat, Rhinolophus mehelyi, in three contrasting environments in western Iran. Zool Midd East. 2004, 33: 65-72.

    Article  Google Scholar 

  48. Yela JL, Herrera CM: Seasonality and life cycles of woody plant-feeding noctuid moths (Lepidoptera: Noctuidae) in Mediterranean habitats. Ecol Entomol. 1993, 18: 259-269. 10.1111/j.1365-2311.1993.tb01099.x.

    Article  Google Scholar 

  49. Russo D, Jones G: Use of foraging habitats by bats in a Mediterranean area determined by acoustic surveys: conservation implications. Ecography. 2003, 26: 197-209. 10.1034/j.1600-0587.2003.03422.x.

    Article  Google Scholar 

  50. Russo D, Jones G, Migliozzi A: Habitat selection by the Mediterranean horseshoe bat, Rhinolophus euryale (Chiroptera: Rhinolophidae), in a rural area of southern Italy and implications for conservation. Biol Conserv. 2002, 107: 71-81. 10.1016/S0006-3207(02)00047-2.

    Article  Google Scholar 

  51. Aihartza J, Garin I, Goiti U, Zabala J, Zuberogoitia I: Spring habitat selection by the Mediterranean horseshoe bat (Rhinolophus euryale) in the Urdaibai Biosphere Reserve (Basque Country). Mammalia. 2003, 67: 25-32.

    Article  Google Scholar 

  52. Goiti U, Aihartza J, Garin I, Zabala J: Influence of habitat on the foraging behaviour of the Mediterranean horseshoe bat, Rhinolophus euryale. Acta Chiropterol. 2003, 5: 75-84.

    Article  Google Scholar 

  53. Goiti U, Aihartza J, Almenar D, Salsamendi E, Garin I: Seasonal foraging by Rhinolophus euryale (Rhinolophidae) in an Atlantic rural landscape in northern Iberian Peninsula. Acta Chiropterol. 2006, 8: 141-156. 10.3161/1733-5329(2006)8[141:SFBRER]2.0.CO;2.

    Article  Google Scholar 

  54. Bontadina F, Schofield H, Naef-Daenzer B: Radio-tracking reveals that lesser horseshoe bats (Rhinolophus hipposideros) forage in woodland. J Zool. 2002, 258: 281-290. 10.1017/S0952836902001401.

    Article  Google Scholar 

  55. Voigt CC, Schuller B-M, Grief S, Siemers BM: Perch-hunting in insectivorous Rhinolophus bats is related to the high energy costs of manoeuvring in flight. J Comp Physiol B. 2010, 180: 1079-1088. 10.1007/s00360-010-0466-x.

    Article  PubMed  Google Scholar 

  56. Rainho A, Palmeirim JM: The importance of distance to resources in the spatial modelling of bat foraging habitat. PLoS One. 2011, 6: e19227-10.1371/journal.pone.0019227.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  57. Pianka ER: Competition and niche theory. Theoretical Ecology. Edited by: May RM. 1981, Oxford: Blackwell, 167-196.

    Google Scholar 

  58. Alberdi A, Garin I, Aizpurua O, Aihartza J: The Foraging Ecology of the Mountain Long-Eared Bat Plecotus macrobullaris Revealed with DNA Mini-Barcodes. PLoS One. 2012, 7: e35692-10.1371/journal.pone.0035692.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  59. Animal Behaviour Society: Guidelines for the treatment of animals in behavioural research and teaching. Anim Behav. 2006, 71: 245-253.

    Article  Google Scholar 

  60. Anthony EL: Age determination in bats. Ecological and Behavioural Methods for the Study of Bats. Edited by: Kunz TH. 1988, Washington: Smithsonian Institute Press, 47-58.

    Google Scholar 

  61. Fenton MB: Aerial-feeding bats: getting the most out of echolocation. Echolocation in Bats and Dolphins. Edited by: Thomas JA, Moss CF, Vater M. 2004, Chicago: University of Chicago Press, 350-355.

    Google Scholar 

  62. Chinery M: A Field Guide to the Insects of Britain and Northern Europe. 1977, London: Collins

    Google Scholar 

  63. Barrientos JA: Bases para un Curso Práctico de Entomología. 1988, Barcelona: Asociación Española de Entomología

    Google Scholar 

  64. McAney M, Shiel C, Sullivan C, Farley JS: The Analysis of Bat Droppings. 1991, London: Mammal Society

    Google Scholar 

  65. Anderson M, Ter Braak C: Permutation test for multi-factorial analysis of variance. J Stat Compt Sim. 2003, 73: 85-113. 10.1080/00949650215733.

    Article  Google Scholar 

  66. Aldridge HDJN, Brigham RM: Load carrying and manoeuvrability in an insectivorous bat: a test of the 5% "rule" of radio-telemetry. J Mammal. 1988, 69: 379-382. 10.2307/1381393.

    Article  Google Scholar 

  67. White GC, Garrot RA: Analysis of Wildlife Radio-Tracking Data. 1990, London: Academic Press

    Google Scholar 

  68. Steyerberg EW: Clinical Prediction Models: A Practical Approach to Development, Validating, and Updating Series. 2009, New York: Springer

    Book  Google Scholar 

  69. Breiman L, Friedman J, Olshen R, Stone C: Classification and Regression Trees. 1984, Belmont: Wadsworth International Group

    Google Scholar 

  70. Venables WN, Ripley BD: Modern Applied Statistics with S-Plus. 1994, New York: Springer-Verlag

    Book  Google Scholar 

  71. Feinsinger P, Spears EE, Poole RW: A simple measure of niche breadth. Ecology. 1981, 62: 27-32. 10.2307/1936664.

    Article  Google Scholar 

  72. Krebs CJ: Ecological Methodology. 1989, New York: Harper & Collins

    Google Scholar 

  73. Manly BJF: Randomization, Bootstrap and Monte Carlo Methods in Biology. 1997, London: Chapman & Hall

    Google Scholar 

Download references

Acknowledgements

Many thanks go to John O’Brien and Kate Johnson, who kindly edited and proofread the manuscript, and especially to Mark Brigham for providing valuable suggestions on a later version of it. Borja Ruiz advised with GIS analyses and Aitor Larrañaga with the statistics. Special thanks to Godfried Schreur and Ana Cordero for sharing their knowledge on local bat fauna. We are grateful to David Almenar, María Napal, Toni Castelló, Sergio Couto, Aritz Aranzabal, and Maider Guiu for help with fieldwork. We appreciate all the help of Oscar de Paz (SECEMU). This work was supported by the Consejería de Agricultura y Medio Ambiente of The Regional Council of Extremadura and the European Commission as part of project LIFE04/NAT/E/000043. The Basque Government and the University of the Basque Country UPV/EHU supported UG and ES.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Joxerra Aihartza.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

ES, IG and JA conceived of, designed, and coordinated the study. IA participated in the design of the study and performed the statistical analyses. UG participated in the design of the study and in fieldwork. All authors read and approved the final manuscript.

Electronic supplementary material

12983_2012_208_MOESM1_ESM.docx

Additional file 1: Appendix A. Radio-tracking details of R. mehelyi and R. euryale individuals followed in sympatry; Sex, reproductive status (pregnant females, lactating females), tracking effort (tracking period and active foraging fixes [AFF]), individual foraging home ranges (MCP 100%), and travelled distances (maximum and mean) of radio-tracked R. mehelyi and R. euryale individuals in sympatric conditions. (DOCX 33 KB)

Authors’ original submitted files for images

Below are the links to the authors’ original submitted files for images.

Authors’ original file for figure 1

Authors’ original file for figure 2

Authors’ original file for figure 3

Rights and permissions

Open Access This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License ( https://creativecommons.org/licenses/by/2.0 ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Salsamendi, E., Garin, I., Arostegui, I. et al. What mechanism of niche segregation allows the coexistence of sympatric sibling rhinolophid bats?. Front Zool 9, 30 (2012). https://0-doi-org.brum.beds.ac.uk/10.1186/1742-9994-9-30

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://0-doi-org.brum.beds.ac.uk/10.1186/1742-9994-9-30

Keywords