SEARCH

SEARCH BY CITATION

Keywords:

  • biodiversity;
  • cenotes;
  • echolocation;
  • forest;
  • insectivorous bats;
  • Mexico;
  • pastureland;
  • time-expansion

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • 1
    Microchiropteran bats have the potential to be important biodiversity indicator species as they are distributed globally and are important in ecosystem functioning. Survey and monitoring protocols for bats are often ineffective as sampling techniques vary in their efficacy depending on the species involved and habitats surveyed. Acoustic sampling using bat detectors may prove an alternative or complementary technique to capture methods but is largely untested in the tropics.
  • 2
    To compare the efficacy of bat detectors and capture methods in surveys, we used ground mist nets, sub-canopy mist nets and harp traps to sample bats while simultaneously recording the echolocation calls of insectivorous bats in a diversity of habitats in the Yucatan, Mexico. We described echolocation calls, analysed call characteristics to identify species, and compared species inventories derived from traditional capture methods with those derived from acoustic sampling.
  • 3
    A total of 2819 bats representing 26 species and six families were captured; 83% were captured in ground nets, 13% in sub-canopy nets and 4% in harp traps. Fourteen species and five phonic types were identified based on five echolocation call characteristics. Discriminant function analysis showed a high level of correct classification of the calls (84·1%), indicating that identification of species by their echolocation calls is feasible.
  • 4
    In all habitats, acoustic sampling and capture methods sampled significantly more species each night than capture methods alone. Capture methods failed to sample 30% of the bat fauna, and aerial insectivores were sampled only by bat detectors.
  • 5
    Synthesis and applications. Given the importance of bats in ecosystem functioning, and their potential as indicator species, developing effective methodologies to survey and monitor bats is important for sustainable forest management and biodiversity conservation. Acoustic sampling should be used with capture methods to increase inventory completeness in bat assemblage studies, and could form part of a single standardized monitoring protocol that can be used globally in tropical forests, as this method detects aerial insectivores not sampled by capture methods.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Bats are the second largest order of mammals with 1116 recorded species (Simmons 2005). Neotropical bats exhibit the greatest diversity of feeding habits and their key ecological roles as seed dispersers, pollinators and predators of insects and small vertebrates is well established (e.g. Heithaus, Fleming & Opler 1975; Lang et al. 2006). The different ecological requirements that bats exhibit, the relatively ease of identification of most species, their abundance in most habitats, and the ability of biologists to apply standard methods to sample them without jeopardizing their conservation, favour the use of bats as an indicator group (Favila & Halffter 1997). The species richness and abundance of gleaning animalivores, for example, are negatively affected in deforested, fragmented or selectively logged landscapes (e.g. Fenton et al. 1992; Gorresen & Willig 2004), whereas some frugivores such as Sturnira and Carollia species which feed largely on the fruits of pioneer plants, increase their populations in habitats dominated by such species (e.g. Estrada, Coates-Estrada & Meritt 1993; Clarke, Pio, & Racey 2005). Despite their value as bioindicators, at present there is a lack of long-term bat monitoring programmes globally and no common set of protocols developed to monitor bats. Determining effective and comprehensive methods for surveying and monitoring bats is an essential first step in developing monitoring protocols and conducting complete species inventories of habitats.

For more than 100 years, attempts have been made to inventory Neotropical bat faunas mainly by setting mist nets. However, mist nets set at ground level and in the forest canopy are heavily biased towards the capture of phyllostomid bats which are more easily captured than species in other families (Barnett et al. 2006). Thus, traditional sampling methods do not give a full representation of bat assemblages. Although studies of Neotropical bats assemblages using multiple capture techniques such as ground mist nets, canopy nets and harp traps (an array of stretched nylon lines that run vertically along the length of four metal frames, above an escape-proof collecting bag for bats; Francis 1989) have documented high species richness in some habitats, aerial insectivores are still under-represented among captures (Simmons & Voss 1998; Clarke, Rostant & Racey 2005; MacSwiney G. et al. 2007). Aerial insectivores often commute and forage above the forest canopy, and even when they are within the sampling range of nets and harp traps, they are able to easily detect and avoid such devices (Berry et al. 2004).

Bat detectors allow researchers to hear and subsequently visualize the ultrasonic echolocation calls of most species and, in Europe and North America, are regularly used to identify some species (e.g. Fenton & Bell 1979; Russ 1999). Acoustic sampling has rarely been conducted in the Neotropics (e.g. Rydell et al. 2002; Jung, Kalko & von Helversen 2007), and detailed descriptions of the echolocation calls of most of the Neotropical bat species are lacking. Bats are abundant and important components of vertebrate communities, and in particular, frugivores and insectivores play key roles in forest dynamics and regeneration. Their inclusion in management plans to conserve biodiversity and manage forests is hampered by incomplete species lists. The development of new sampling methods such as bat detectors, and the testing of their efficacy, is therefore essential and urgently required. Studies in a range of habitat types and environmental conditions are needed to test and compare the efficacy of acoustical sampling against traditional capture methods, in order to devise sampling methodologies and protocols which will increase inventory completeness (Moreno & Halffter 2000).

Ours is the first study to simultaneously compare acoustical sampling with the most commonly employed capture methods in a variety of natural and anthropogenic-modified habitats. We thus aimed to (i) describe the echolocation calls of free-flying insectivorous bats in the Yucatan Peninsula, (ii) test the efficacy of bat detectors at identifying species, and (iii) compare the effectiveness of the most commonly used sampling methods simultaneously with acoustical sampling in constructing species inventories in a range of habitat types.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

capture methods

A total of 96 nights of fieldwork were conducted in 2003, 2004 and 2005 at eight localities in the north of the Yucatan Peninsula: two cenotes (water-filled sink-holes) in pastureland, two cenotes in rainforest and two sites without cenotes in each of these two vegetation types (for description of sites, see Supplementary Material Appendix S1). Each locality was sampled for six nights during both the dry and wet season. Sampling on consecutive nights at the same location was avoided. Each night, five 12·6 m mist nets of 36 mm mesh size, two-ply, and with four shelves (AFO Mist Nets, Manomet, Inc., Manomet, MA, USA.) were erected at ground level (0–3 m). A 12·3-m mist net modified to become an 8·3-m sub-canopy mist net (Avinet Inc., Dryden, NY, USA) was set following the method described by Clarke et al. (2005). A four-frame harp trap (Francis 1989) with an area of 1·42 m2 was set in trails. Nets and the harp trap were set at sunset, left open for an average of 7 h and monitored every 30 min. We estimated capture rate for each catching method by dividing the number of bats captured by the number of mist net hours (bats/mnh) or harp-trap hours (bats/hth). Mist net hours were calculated as the length of all nets set multiplied by the number of hours they were left open (Medellín 1993). We standardized catching methods to achieve an equal sampling effort among sites.

acoustic sampling

Each night, echolocation calls were recorded with a Pettersson D980 bat detector (Pettersson Elektronik AB, Uppsala, Sweden) and stored using a Sony Walkman Professional WM-D6C recorder with TDK Type IV metal tapes. For a continuous period of 10 min, we set the bat detector to repeatedly record calls for 3 s of real time and time-expand these calls (10×). These time-expanded recordings were stored on Channel 1 of the Walkman. At the same time, the bat detector was also set to record in frequency division, and the output recorded on Channel 2 of the Walkman. Call characteristics were measured only from the time expansion recordings. This method of recording allowed only 9% of the real time to be recorded in each 10-min period (Jones, Vaughan & Parsons 2000). However, this protocol was standardized in all habitats which permitted comparison. The first recording period was carried out 30 min after sunset followed by the two subsequent sampling periods at hourly intervals. To minimize call variability and changes in detection due to habitat structure, we recorded the calls in open spaces. We are aware that recording the calls in open spaces reduces the detection of bats such as phyllostomids, which forage in cluttered habitats (Schnitzler & Kalko 1998). However, it is difficult to use bat detectors to sample phyllostomid bats because they produce low-intensity calls. In addition, their calls show low inter-specific variation (Vaughan et al. 2004) resulting in low classification rates in discriminant function analyses (D.V. Pio, F.M. Clarke and P.A. Racey, unpublished data). Consequently, we only used capture methods and not acoustic sampling to sample phyllostomid bats. In habitats with cenotes, the microphone was orientated towards the water and at the forested sites towards natural vegetation gaps. A tripod was used to maintain the bat detector at an angle of 45° from the horizontal at 1·3 m above-ground. To aid species identification, calls were analyzed with BatSound Pro 3·10 (Pettersson Elektronik AB, Uppsala, Sweden). Recordings from free-flying bats were identified by comparison with our echolocation call library constructed with captured individuals (identified to species). Detailed description of the methods applied and species identification are summarized in Supplementary Material Appendix S1.

statistical analyses

To assign confidence to identification of species, a discriminant function analysis (DFA) was performed. To construct the model, all call variables employed to visually separate and identify the species were used: frequency of maximum energy (FME), low frequency (LF), high frequency (HF), duration (DUR) and inter-pulse interval (IPI). We did not use data obtained from calls recorded from hand-released bats in the statistical analyses. Examination of covariance matrices with a Box's M test demonstrated that these were heterogeneous (F = 10·4, P < 0·001), therefore a quadratic discriminant analysis was performed. Cross-validation was used in the DFA. A multivariate analysis of variance (manova) was conducted to detect statistical significance of DFA analysis. Wilk's λ statistic and naive ranking were used to explain the importance of each variable in classification. Species accumulation curves were constructed to assess completeness of sampling effort with capture methods only and with capture plus acoustic methods at each habitat and at the landscape scale. To eliminate the influence of the order in which nights were added to the total, the sample was randomized 100 times using Species Richness and Diversity version 2·6 (Pisces Conservation Ltd., UK). We used a non-parametric Wilcoxon test to compare species richness per night determined by capture methods only and with these methods plus echolocation call analysis. Statistical analyses were performed with Minitab 14 (Minitab Inc., State College, PA, USA) and Brodgar version 2·5·1 (Highland Statistics Ltd, Newburgh, UK).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

capture methods

A total of 2819 bats of 26 species and six families was captured (Table 1). The Phyllostomidae was the best represented family with 17 species and members of this family accounting for 95% of the captures, whereas the Emballonuridae, Natalidae and Molossidae were each represented by only one species and relatively few captures. Artibeus jamaicensis Leach, 1821 was the most abundant species in all sites with 61–65% of captures. Eighty-three per cent of the bats were captured in ground mist nets, 13% in sub-canopy nets and 4% in harp traps (Table 2). We captured 16 species with more than one method, but five species were exclusively recorded in the ground mist nets, three species in the sub-canopy nets and two species in the harp traps (Table 1). Fifteen species of our captured species (57% of the assemblage) can be considered as rare (where rare species are those contributing < 0·5% of the total captures), with four species being represented by a single individual (Table 1).

Table 1.  Checklist of species recorded in the north of the Yucatan Peninsula, Mexico, by capture methods (C) and acoustic methods (A). Species in boldface were recorded exclusively by acoustic methods. CENPAST, cenotes in pastureland; PAST, pastureland; CENFOR, cenotes in forest; FOR, forest
TaxonPasturelandForest
CenotesNo cenotesCenotesNo cenotes
  • *

    rare species (contributing < 0·5% of the total captures). Exclusively captured in:

  • †ground level mist nets,

  • ‡sub-canopy mist net, and

  • §

    §harp trap.

Emballonuridae
 Peropteryx macrotisAAAA
 Saccopteryx bilineata*† AAC, A
 PHYLLOSTOMIDAE    
 Desmodontinae    
 Desmodus rotundusCC C
 Diphylla ecaudata*‡   C
 Glossophaginae    
 Glossophaga soricinaCCCC
 Phyllostominae    
 Chrotopterus auritus*†C   
 Lampronycteris brachyotis*†  CC
 Lonchorhina aurita  CC
 Micronycteris microtisCCCC
 Micronycteris schmidtorumC   
 Mimon cozumelae*C CC
 Carolliinae    
 Carollia sowelli  CC
 Carollia perspicillata* CCC
 Stenodermatinae    
 Sturnira liliumCC C
 Artibeus jamaicensisCCCC
 Artibeus lituratusCCCC
 Artibeus phaeotisCCCC
 Centurio senex*†C CC
 Chiroderma villosum*CCCC
Mormoopidae
 Mormoops megalophyllaAAAA
 Pteronotus parnelliiC, AC, AC, AC, A
 Pteronotus personatusA A 
 Pteronotus davyi*C, AAAC, A
Natalidae
 Natalus stramineusCCCC
Molossidae
 Molossus rufus*‡AC, A A
 Molossus sinaloaeA AA
 Nyctinomops laticaudatusAAAA
 Molossid 1A A 
 Molossid 2AAAA
 Molossid 3AAA 
 Molossid 4AAA 
 Molossid 5AA  
Vespertilionidae
 Eptesicus furinalis*†AAC, A 
 Lasiurus ega*‡AC, AA 
 Lasiurus intermediusAAA 
 Rhogeessa aeneus*†C, AC, AC, A 
 Myotis keaysiC, AAC, AC, A
CENPASTPASTCENFORFOR
No. of species captured17141720
Species recorded acoustically18161710
Total no. of species31263026
Table 2.  Sampling effort of the catching methods per habitat. n, number of bats captured; S, number of species; mnh, mist net hour; Bats/mnh, capture rate at number of bats captured divided by mist net hour; Bats/hth, capture rate at number of bats captured divided by harp-trap hour. CENPAST, cenotes in pastureland; PAST, pastureland; CENFOR, cenotes in forest; FOR, forest
HabitatGround mist netSub-canopy mist netHarp trap
nSmnh*Bats/mnhnSmnhBats/mnhnShthBats/hth
  • *

    mnh is the product of length of all nets in metres multiplied by the number of hours nets were set.

CENPAST43713239 7600·001856 532 6400·0017 2381590·14
PAST158 9241 9200·0006120 931 6800·0036  831640·04
CENFOR87316240 0480·0036113 931 7760·0035 4591610·27
FOR85915246 2400·0034881032 4480·0027 3991630·23
Total232721  37716  11513  

general description of calls

A total of 2280 recorded minutes of calls from free-flying bats was analyzed. Fourteen species belonging to the families Emballonuridae, Mormoopidae, Molossidae and Vespertilionidae, and five phonic types belonging to the family Molossidae were identified (Supplementary Material Table S1; Fig. 1). Most species consistently produced one type of call but some, such as Saccopteryx bilineata Temminck, 1838, Molossus rufus E. Geoffroy, 1805 and Molossus sinaloae J. A. Allen, 1906, displayed alternate high and low-frequency search phase calls. Molossid 5 produced two types of low frequency calls: one of short duration (Fig. 1c, M5c) and one of long duration (Fig. 1c, M5a), and a less common higher frequency call was also evident (Fig. 1c, M5b). These different types of calls are described but were not treated as different prior to the discriminant function analysis (see Supplementary Material Table S1).

image

Figure 1. Spectrogram with typical search calls of the four families identified in the north of the Yucatan Peninsula. (a) Emballonuridae, (b) Mormoopidae, (c) Molossidae: Ms, Molossus sinaloae; Mr, Molossus rufus; Nl, Nyctinomops laticaudatus; M1, phonic type molossid 1; M2, phonic type molossid 2; M3, phonic type molossid 3; M4, phonic type molossid 4; M5, phonic type molossid 5; and (d) Vespertilionidae.

Download figure to PowerPoint

Considerable variation in time and spectral characteristics was evident among species. The calls of Pteronotus personatus Wagner, 1843 had the highest FME with a mean of 80·1 kHz, whereas the call of Molossid 1 had the lowest FME with 15·7 kHz. Mean call duration of the species varied from 3·6 milliseconds (ms) for Myotis keaysi J. A. Allen, 1914 up to 65·6 ms for Molossid 5. Inter-pulse interval showed the greatest variation of all call characteristics measured, with coefficients of variation of up to 70%. Nyctinomops laticaudatus E. Geoffroy, 1805 showed the highest intraspecific variation in FME with a minimum of 23·3 kHz and a maximum of 31·1 kHz (Supplementary Material Table S1). For detailed description of species’ calls, see Supplementary Material Appendix S2.

discriminant function analysis

Quadratic discriminant function analysis resulted in a 92·1% correct classification of the original data and 84·1% when cross-validated (Supplementary Material Table S2). manova showed that the model was significantly different from random (Wilk's λ = 0·001, F = 93·7, P < 0·001). The calls of Pteronotus parnellii Gray, 1843, Pteronotus davyi Gray, 1838 and Myotis keaysi were 100% correctly classified. In contrast, for Lasiurus intermedius H. Allen, 1862, only 35% of the calls were correctly classified. The first three discriminant functions explain 98·6% of the total variance. To determine the importance of variables for discriminating between groups, a backward selection was performed. Wilk's λ statistic and naive ranking showed that the duration of the call and frequency of most energy were the most important variables discriminating between groups, whereas inter-pulse interval was the least important. In a further DFA analysis, removal of variables with lower discrimination power reduced the classification to 81·3% for the original data and 78·8% for the cross-validated data; thus, the best model for classification of all species relied on all five variables.

inventory completeness

The analysis of echolocation calls of free-flying bats revealed the presence of 11 insectivorous species not sampled by any of the capture techniques used to sample habitats (Table 1). Thus, harp traps and mist nets at ground level and in the canopy failed to sample 30% of the bat fauna, and this component was only sampled by bat detectors. The average number of species sampled per night with capture methods and capture methods plus acoustic sampling was 4·5 and 10·3, 3·0 and 5·6, 5·5 and 9·7 and 5·3 and 6·12 for pastureland with cenotes, pastureland without cenotes, forest with cenotes and forest without cenotes, respectively. The number of species recorded per night with acoustic sampling in combination with capture methods was significantly higher than that recorded with capture methods alone. This was true for all habitats sampled (all comparisons P < 0·01) (Fig. 2). Acoustic sampling demonstrated that locally rare insectivorous species (< 0·5% of total captures) captured in only one habitat were present in most habitats and much more common than suggested by analysis of capture methods. A combination of acoustic sampling and capture methods recorded 37 species for the north of the Yucatan Peninsula (Table 1; Fig. 2).

image

Figure 2. Bat species accumulation curves for the four habitats sampled and landscape. Δ, species recorded with capture methods only; and •, species recorded with capture methods and acoustical sampling.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We described and analysed the echolocation calls of bats and compared species inventories derived from traditional capture methods with those derived from acoustic sampling. Fourteen species and five phonic types belonging to the families Emballonuridae, Mormoopidae, Molossidae and Vespertilionidae were identified based on five echolocation call characteristics. Discriminant function analysis showed a high level of correct classification of the calls, indicating that identification of species by their echolocation calls is feasible. At all habitats, acoustic sampling and capture methods sampled significantly more species each night than capture methods alone. Capture methods failed to sample 30% of the bat fauna. This component, which consisted entirely of aerial insectivores, was sampled only by bat detectors.

species identification

The correct identification to bat species was possible from the echolocation call characteristics of the bats we recorded. Most call parameters were consistent with those reported in the literature for the few species whose calls have been analysed (O’Farrell & Miller 1997, 1999; Rydell et al. 2002). Some differences were detected between our recorded calls and those published, particularly in call duration, for example, Mormoops megalophylla Peters, 1864 and P. personatus (O’Farrell & Miller 1997). Such differences are probably due to calls being recorded in habitats that differ in levels of background clutter (forest vs. open-space habitats) since clutter affects intra-specific call variability. Alternatively, reported differences in calls may be attributed to researchers using different recording systems (frequency division vs. time-expansion).

In recent years, there has been a lively debate about the comparative efficiency of the frequency division (i.e. the Anabat system) and time expansion systems in detecting echolocation calls (O’Farrell, Gannon & Miller 1999; Fenton 2000). The bat detector's time expansion facility retains most information of the call in contrast to frequency division systems which often fail to detect all harmonics (Fenton 2000). This was particularly important in successfully identifying species in the present study, because observation of flight behaviour, an important factor discriminating species in the field (Kalko 1995), was not possible for many species. The strongest harmonics are, however, efficiently detected with frequency division recordings (O’Farrell & Miller 1999), and we were able to compare them with our time-expanded recordings.

echolocation call design

Differences in call design among species can be explained by their different foraging habits. Within the Emballonuridae, Peropteryx macrotis Wagner, 1843, which forages mainly in open spaces, emits narrowband calls, whereas S. bilineata, which typically forages close to vegetation, also emits a narrowband call but with broadband components that helps to determine the position of prey relative to background clutter (Jung et al. 2007). Within the Mormoopidae, species displayed multiharmonic calls of short duration, with the exception of P. parnellii. The calls of P. parnellii consist of a long narrowband component terminating with a broadband sweep, a signal design that gives this species not only high performance in target detection but also enables classification of targets in cluttered habitats (Jones & Teeling 2006). Species within the Molossidae displayed long narrowband or broadband–narrowband calls which are typically used for hunting in uncluttered or open spaces (Schnitzler & Kalko 1998). Finally, vespertilionids emitted broadband calls with a distinct shallow component (e.g. Eptesicus furinalis d’Orbigny 1847, M. keaysi) used for foraging when the background is cluttered (Schnitzler & Kalko 1998).

sampling methods and inventory completeness of bat assemblages

Exhaustive inventories of Neotropical bat faunas are not feasible objectives for short-term fieldwork: too many species are rare or hard to capture (Simmons & Voss 1998). Our results demonstrate that, with the addition of acoustic methods, new species are sampled at a faster rate than with the use of capture methods alone, and inventory completeness increased. This has important implications for sampling design as the use of bat detectors might reduce the time, effort and money required for surveys. Comparison between acoustic sampling and traditional sampling methods (mist netting and harp traps) conducted simultaneously every sampling night in the variety of habitats shows that insectivorous species, particularly within the family Molossidae, are under-sampled by netting methods. Mist nets were highly effective at capturing phyllostomids, whereas harp traps were effective at sampling mormoopids that forage close to the ground such as P. parnellii and P. davyi. However, species that fly and forage at higher altitudes, above the height of nets and harp traps (e.g. M. sinaloae, N. laticaudatus, molossid phonic types) were not caught. Other species such as S. bilineata and P. macrotis visually and/or acoustically detected the mist nets and harp traps, as we acoustically detected these bats flying close to but avoiding nets and harp traps.

It has been suggested that in the Neotropics where phyllostomids predominate, mist netting is the most appropriate sampling method (e.g. Fenton et al. 1992; Simmons & Voss 1998). Following the present analysis of the acoustic recordings of free-flying bats, the checklist of bats in northern Yucatan increased from 26 species, based on captures only, to 37 species. Twenty of these species are non-phyllostomids and were more easily recorded with acoustic sampling. These findings demonstrate that the use of traditional capture methods such as mist nets alone, would give a misleading picture of the bat assemblage and that only studies employing both multiple capture methods and acoustical sampling increase inventory completeness and produce representative species assemblages in habitats and landscapes. Accurate species lists and abundance estimates are essential to carry out effective forest management and conservation of bat diversity. In the Neotropics, this only can be achieved with the combination of both capture and acoustic methods. We strongly recommend the adoption of acoustic sampling in all future studies of bat assemblages in the Neotropics.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

This study was funded by the Consejo Nacional de Ciencia y Tecnología, Mexico (no.168790) and the University of Aberdeen. We thank B. Bolívar, P. Vilchis, J. Pech, C. Gelati and many other field assistants for invaluable help, and owners and staff of the El Eden Ecological Reserve, Santa María Ranch and private ranches for their hospitality.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Barnett, A.A., Sampaio, E.M., Kalko, E.K.V., Shapley, R.L., Fischer, E., Camargo, G. & Rodríguez-Herrera, B. (2006) Bats of Jaú National Park, central Amazonia, Brazil. Acta Chiropterologica, 8, 103128.
  • Berry, N., O’Connor, W., Holderied, M.W. & Jones, G. (2004) Detection and avoidance of harp traps by echolocating bats. Acta Chiropterologica, 6, 335346.
  • Clarke, F.M., Pio, D.V. & Racey, P.A. (2005) A comparison of logging systems and bat diversity in the Neotropics. Conservation Biology, 19, 11941204.
  • Clarke, F.M., Rostant, L.V. & Racey, P.A. (2005) Life after logging: post-logging recovery of a Neotropical bat community. Journal of Applied Ecology, 42, 409420.
  • Estrada, A., Coates-Estrada, R. & Meritt, Jr., D. (1993) Bat species richness and abundance in tropical rain forest fragments and in agricultural habitats at Los Tuxtlas, Mexico. Ecography, 16, 309318.
  • Favila, M.E. & Halffter, G. (1997) The use of indicator groups for measuring biodiversity as related to community structure and function. Acta Zoológica Mexicana, nueva serie, 72, 125.
  • Fenton, M. B. (2000) Choosing the ‘correct’ bat detector. Acta Chiropterologica, 2, 215224.
  • Fenton, M.B. & Bell, G.P. (1979) Echolocation and feeding behaviour in four species of Myotis (Chiroptera). Canadian Journal of Zoology, 57, 12711277.
  • Fenton, M.B., Acharya, L., Audet, D., Hickey M.B., Merriman, C., Obrist, M.K. & Syme, S.M. (1992) Phyllostomid bats (Chiroptera: Phyllostomidae) as indicators of habitat disruption in the Neotropics. Biotropica, 24, 440446.
  • Francis, C.M. (1989) A comparison of mist nets and two designs of harp traps for capturing bats. Journal of Mammalogy, 70, 865870.
  • Gorresen, P.M. & Willig, M.R. (2004) Landscape responses of bats to habitat fragmentation in Atlantic forest of Paraguay. Journal of Mammalogy, 85, 688697.
  • Heithaus, E.R., Fleming, T.H. & Opler, P.A. (1975) Foraging patterns and resource utilization in seven species of bats in a seasonal tropical forest. Ecology, 56, 841854.
  • Jones, G. & Teeling, E.C. (2006) The evolution of echolocation in bats. Trends in Ecology & Evolution, 21, 149156.
  • Jones, G., Vaughan N. & Parsons, S. (2000) Acoustic identification of bats directly sampled and time expanded recordings of vocalisations. Acta Chiropterologica, 2, 155170.
  • Jung, K., Kalko, E.K.V. & Von Helversen, O. (2007) Echolocation calls in Central American emballonurid bats: signal design and call frequency alternation. Journal of Zoology (London), 272, 125137.
  • Kalko, E.K.V. (1995) Echolocation signal design, foraging habitats and guild structure in six neotropical sheath-tailed bats (Emballonuridae). Symposia of the Zoological Society of London, No. 67, 259273.
  • Lang, A.B., Kalko, E.K.V., Römer, H., Bockholdt, C. & Dechmann, D.K.N. (2006) Activity levels of bats and katydids in relation to the lunar cycle. Oecologia, 146, 659666.
  • MacSwiney G., M.C., Vilchis, P., Clarke, F.M. & Racey, P.A. (2007) The import-ance of cenotes in conserving bat assemblages in the Yucatan, Mexico. Biological Conservation, 136, 499509.
  • Medellín, R.A. (1993) Estructura y diversidad de una comunidad de murciélagos en el trópico húmedo mexicano. Avances en el Estudio de los Mamíferos de México (eds R.A.Medellín & G.Ceballos), pp. 333350. Publicaciones Especiales, Asociación Mexicana de Mastozoología, México.
  • Moreno, C.E. & Halffter, G. (2000) Assessing the completeness of bat biodiversity inventories using species accumulation curves. Journal of Applied Ecology, 37, 149158.
  • O’Farrell, M.J., Gannon, W.L. & Miller, B.W. (1999) Confronting the dogma: a reply. Journal of Mammalogy, 80, 297302.
  • O’Farrell, M.J. & Miller, B.W. (1997) A new examination of echolocation calls of some Neotropical bats (Emballonuridae and Mormoopidae). Journal of Mammalogy, 87, 954963.
  • O’Farrell, M.J. & Miller, B.W. (1999) Use of vocal signatures for the inventory of free-flying Neotropical bats. Biotropica, 31, 507516.
  • Russ, J. (1999) The Bats of Britain and Ireland: Echolocation Calls, Sound Analysis, and The Identification. Alana Books, Bishop's Castle.
  • Rydell, J., Arita, H.T., Santos, M. & Granados, J. (2002) Acoustic identification of insectivorous bats (Order Chiroptera) of Yucatan, Mexico. Journal of Zoology (London), 257, 2736.
  • Schnitzler, H.U. & Kalko, E.K.V. (1998) How echolocating bats search and find food. Bat Biology and Conservation (eds T.H.Kunz & P.A.Racey), pp. 183196. Smithsonian Institution Press, Washington, DC.
  • Simmons, N.B. (2005) Order Chiroptera. Mammal Species of the World (eds D.E.Wilson & D.M.Reeder), pp. 312529. Johns Hopkins University Press, Baltimore, MD, USA.
  • Simmons, N.B. & Voss, R.S. (1998) The mammals of Paracou, French Guiana: a neotropical lowland rainforest fauna. I. Bats. Bulletin of the American Museum of Natural History, 237, 1219.
  • Vaughan, N., Parsons, S., Barlow, K.E. & Gannon, M.R. (2004) Echolocation calls and wing morphology of bats from the West Indies. Acta Chiropterologica, 6, 7590.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Appendix S1. Detailed methods

Appendix S2. Description of calls by family

Table S1. Call characteristics of insectivorous bat species

Table S2. Discriminant function analysis results

Please note: Blackwell Publishing are not responsible for the content or functionality of any supplementary materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
JPE_1531_sm_AppendixS1.doc34KSupporting info item
JPE_1531_sm_AppendixS2.doc40KSupporting info item
JPE_1531_sm_TableS1.doc87KSupporting info item
JPE_1531_sm_TableS2.doc80KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.