The Hawaiian islands are an extremely isolated oceanic archipelago, and their fauna has long served as models of dispersal in island biogeography. While molecular data have recently been applied to investigate the timing and origin of dispersal events for several animal groups including birds, insects, and snails, these questions have been largely unaddressed in Hawai'i’s only native terrestrial mammal, the Hawaiian hoary bat, Lasiurus cinereus semotus. Here, we use molecular data to test the hypotheses that (1) Hawaiian L. c. semotus originated via dispersal from North American populations of L. c. cinereus rather than from South American L. c. villosissimus, and (2) modern Hawaiian populations were founded from a single dispersal event. Contrary to the latter hypothesis, our mitochondrial data support a biogeographic history of multiple, relatively recent dispersals of hoary bats from North America to the Hawaiian islands. Coalescent demographic analyses of multilocus data suggest that modern populations of Hawaiian hoary bats were founded no more than 10 kya. Our finding of multiple evolutionarily significant units in Hawai'i highlights information that should be useful for re-evaluation of the conservation status of hoary bats in Hawai'i.
Citation: Russell AL, Pinzari CA, Vonhof MJ, Olival KJ, Bonaccorso FJ (2015) Two Tickets to Paradise: Multiple Dispersal Events in the Founding of Hoary Bat Populations in Hawai'i. PLoS ONE 10(6): e0127912. doi:10.1371/journal.pone.0127912
Academic Editor: Michelle L. Baker, CSIRO, AUSTRALIA
Received: February 3, 2015; Accepted: April 20, 2015; Published: June 17, 2015
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication
Data Availability: The data have been uploaded to GenBank, with the accession numbers KR349974-KR350142.
Funding: This research was supported financially by the Pacific Island Ecosystems Research Center, U.S. Geological Survey (#G12AC20117). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The Hawaiian islands are among the most isolated archipelagos in the world, and their native flora and fauna have served as long-standing models for island biogeography and adaptive radiation . The Hawaiian archipelago has been and continues to be formed by a volcanic hotspot under the Pacific tectonic plate, and has never been connected to continental landmasses. Islands erode and increase in age as they move with the plate towards the northwest, with ages of the main Hawaiian islands ranging from 5.1–4.7 MY for the islands of Ni'ihau and Kaua'i to <0.5 MYA for the island of Hawai'i . Due to their isolation, all native plant and animal species originally colonized the Hawaiian islands by long-distance dispersal via sea or air (e.g., on oceanic flotsam, passive dispersal by air currents, active flight, or indirectly hitching a ride such as seeds or snails attached to birds).
Molecular data have been key to unraveling the colonization history of several groups of terrestrial Hawaiian invertebrates and birds [3–7]. These analyses have shown that some taxa were established in the islands via multiple dispersal events from source populations outside Hawai'i, and others via a single colonization. The diversity of native Hawaiian terrestrial groups varies greatly, with some taxa being highly speciose (e.g., land snails, passerine birds, and insects), and others such as mammals being severely depauperate. With the exception of the native monk seal (Monachus schauinslandi), there is only one native terrestrial mammal species in Hawai'i—the hoary bat, Lasiurus cinereus semotus—and its origins and colonization history in the Hawaiian islands remain largely unknown.
The hoary bat, Lasiurus cinereus sensu lato, occupies a large continental distribution from near the latitudinal tree line in Canada, through most of North America to at least Guatemala and from Colombia to northern Argentina and Chile in South America . Also, it occurs in established populations on remote oceanic island groups such as Hawai'i and the Galapagos, and extralimital records exist from locations as far as Iceland and the Orkney Islands [9–11]. No other American bat has established island populations on a similar scale.
Three subspecies of L. cinereus are currently recognized: L. c. cinereus in North and Central America with Philadelphia as type locality, L. c. villosissimus in South America with Paraguay as type locality, and L. c. semotus restricted to the Hawaiian islands . Molecular genetic studies of L. cinereus are limited to an allozyme-based analysis of species-level variation within the genus Lasiurus by Baker et al. , and an analysis of mitochondrial RFLP data by Morales and Bickham  that supported the taxonomic distinction of North American, South American, and Hawaiian populations at the subspecific level. The latter authors included only one specimen from Hawai'i and concluded that L. c. semotus probably originated from North American populations, with dispersal occurring relatively recently based on low divergence values among haplotypes. A morphological study  documented significant divergence of Hawaiian and North American populations, and changes in wing shape and jaw mechanics that may have allowed Hawaiian hoary bats to use different habitats and prey than those of their North American counterparts. Modern sequencing tools and more robust analytical frameworks to reconstruct phylogeographic and demographic history are now available to test hypotheses regarding the number, regions of origin, and timing of L. c. semotus dispersal events to the Hawaiian islands.
Hoary bats have flight morphology that permits long distance dispersal and migration [16,17], including long-distance migration within the North American continent  and the regular colonization of oceanic islands by this species. Bonaccorso and McGuire  modeled energetics and water balance of simulated colonization flights for L. c. cinereus founders arriving in Hawai'i. They concluded that physical conditions (trade wind velocity and direction) and physiological conditions during fall migration (fat storage, energy consumption, and water balance) would allow for long distance dispersal from the Pacific coast of North America (rather than from other parts of its range), and suggested that multiple colonization events may have been possible despite the energetic and physical constraints on dispersers.
In this study we examine the phylogeography of L. cinereus from the Hawaiian islands to estimate colonization history, divergence times, and effective population sizes for Hawaiian hoary bats in the context of additional samples from North America, South America, and the Galapagos islands. We use multiple molecular markers (mitochondrial and nuclear) and analytical approaches (Bayesian and maximum likelihood) to test the following specific hypotheses:
- Hawaiian L. c. semotus originated from North American L. c. cinereus rather than from South American L. c. villosissimus, and
- Hawaiian L. c. semotus originated from a single colonization event.
Our study provides genetic data that can be used to guide conservation priorities for this endangered mammal, and adds to the growing body of evidence for the biogeographic origins of native Hawaiian taxa.
Materials and Methods
Live specimens of Hawaiian hoary bats (Lasiurus cinereus semotus; n = 44) were captured using mist nets in a variety of urban and forest sites on the island of Hawai'i during 2005–2012. Captured individuals were sexed, wing tissue was sampled, and bats were released on site. We used a sterile 3-mm biopsy punch to sample wing tissue ; tissue samples were stored in NaCl-saturated 20% DMSO or silica gel desiccant at ambient temperature in the field, and at—80°C upon return to the lab. Tissue samples from necropsied carcasses from O'ahu, Maui, and Kaua'i were donated by the U.S. Geological Survey’s National Wildlife Health Center Honolulu Field Station. These carcasses (n = 15) were frozen on discovery and subsequent tissue samples were stored at -80°C. Sampling locations are shown in Fig 1.
The locations for L. cinereus specimens used in CO1 analysis.
Samples of L. c. cinereus from continental North America (n = 85) represent a combination of museum tissue collections (pectoral muscle or organs stored in ethanol), mist-netted live bats, and turbine-killed bats from wind energy facilities (wing biopsies from the latter two stored in NaCl-saturated 20% DMSO). Sampling locations at the state, provincial, and island levels are provided in Table 1; more precise sampling information is available upon request.
Wing biopsy samples from several South American L. c. villosissimus specimens were taken from the collection at the American Museum of Natural History. We used two specimens, from Santa Cruz, Bolivia (AMNH catalog # M-260258) and Galapagos Islands, Ecuador (AMNH catalog # M-268079), which were originally collected in 1989 and 1991, respectively. Lasiurus intermedius and L. xanthinus were used as outgroups for phylogenetic analyses, with samples of L. intermedius from the American Museum of Natural History (AMNH 215319) and the Angelo State Natural History Collection (ASK421 = ASNHC 1408), and of L. xanthinus from Centro de Investigaciones Biológicas del Noroestre, La Paz, Mexico.
All live animal sampling was carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Committee on the Ethics of Animal Experiments of the University of Hawai'i at Hilo (permit number 04-039-5). An Endangered Species Federal Fish and Wildlife permit (number TE003483-25) was issued by the U.S. Fish and Wildlife Service. A Protected Wildlife permit (number WL13-04) was issued by the State of Hawai'i, Department of Land and Natural Resources.
DNA isolation and sequencing
DNA was isolated from all tissues using a DNeasy DNA isolation kit (Qiagen) according to the manufacturer’s instructions. DNA extraction from L. c. vilosissimus specimens used a previously described modified DNeasy protocol with extended 4 day digestion time and reduced elution volume . PCR for the cytochrome c oxidase I (COI) gene used primers LCO1490 and HCO2198 and cycling conditions outlined by Hebert et al. . PCR products were cleaned of excess nucleotides and primers using Exo-SAP (Affymetrix), and were sequenced at the University of Arizona Genetics Core facility, yielding a 657 bp fragment. South American specimens from the AMNH were sequenced at the Sackler Institute for Comparative Genomics core facility. Raw sequences were edited and trimmed of primer sequences using CodonCode Aligner. The mitochondrial COI dataset was also supplemented with sequences from GenBank [21,24]. The full dataset was aligned by eye.
Nuclear sequences from the chymase (CHY, 695 bp) and recombination activating gene 2 (RAG2, 706 bp) loci were amplified and cloned from a subset of individuals using primers from Venta et al.  and Stadelmann et al. , respectively. Comparisons of these data against the published horse genome (UCSC Genome Browser) suggest that the CHY locus is an EPIC marker, while the RAG2 locus represents an exon. These loci were chosen based on pilot studies indicating consistent amplification and high variability of these markers in lasiurines (MJV). Initial PCR was conducted using illustra Hot Start mix PCR beads (GE Healthcare), with 0.5 μL of each appropriate primer and 1.0 μL of genomic DNA. Diploid amplicons were cloned using a TOPO TA cloning kit (Life Technologies) according to the manufacturer’s instructions. Five to eight colonies were picked per individual and used as template in a PCR reaction using the same primers. Reactions yielding a target-sized product were sequenced at the University of Arizona Genetics Core facility. Sequences were edited using Sequencher v.5.1 (GeneCodes) and aligned by eye. All novel sequences generated in this study have been accessioned with GenBank (accession numbers KR349974-KR350142).
We used a dataset of 150 COI sequences to construct a phylogeny for all L. cinereus subspecies. First, redundant haplotypes were identified using Collapse v.1.2  and removed, resulting in a dataset of 47 unique haplotypes (see supporting information for details on collapsed haplotypes). A maximum likelihood phylogenetic analysis was conducted using RAxML v.8 . Based on the results of a jModelTest  analysis, we specified a GTR + Γ model, with four gamma rate categories. Support for major nodes was assessed via 1000 rapid bootstrap replicates. Resulting phylogenies were visualized using FigTree v.1.4.0, with L. intermedius designated as the outgroup. Support for alternative topologies was assessed using the approximately unbiased , Shimodaira-Hasegawa , and Kishino-Hasegawa  tests implemented in Consel v.0.1 . To clarify relationships among haplotypes within L. cinereus, we constructed a maximum parsimony network for the mitochondrial COI dataset using TCS v.1.21 [34,35].
The history of population size changes in Hawaiian populations was explored using extended Bayesian skyline plots  in Beast v.1.7.4. These multilocus analyses used DNA sequence data from the mitochondrial COI locus and nuclear CHY and RAG2 introns, and were conducted separately on the Hawaii1 and Hawaii2 clades identified in the maximum likelihood phylogeny (see Results). Evolutionary models for each locus and each clade were identified using jModelTest v.2.1.2 . These models were approximated for the Hawaii1 clade as a TN93 model  with empirical nucleotide frequencies for COI, and an HKY model  with equal nucleotide frequencies for the nuclear loci. For the Hawaii2 clade, models were specified as an HKY model with empirical nucleotide frequencies for the COI and RAG2 loci, and an HKY model with equal nucleotide frequencies for the CHY locus. Substitution rates were scaled relative to a rate of 2% per million years (My) for COI . Substitution rate priors for nuclear loci were set to a U[0,1] distribution with an initial value of 0.02 (= 2% per My). Operator values were adjusted to a weight of 2 for all kappa values, a weight of 15 for all substitution rates and heights, a weight of 40 for the demographic.populationMeanDist variable, a weight of 100 for the demographic.indicators variable, and a weight of 60 for the demographic.scaleActive variable. Three independent runs of 20 million steps each were conducted for each population to verify consistency of results. Runs were checked for convergence by monitoring ESS values in Tracer v.1.5. Estimates of Ne were calculated assuming an average generation time of 2 years.
Historical biogeography of hoary bats in Hawai'i
A maximum parsimony network was constructed for the mitochondrial dataset of 47 unique haplotypes (Fig 2). Lasiurus cinereus samples clearly segregated into three clusters, referred to as Hawaii1, Hawaii2/North America, and South America. Hawaiian hoary bats were divided among two clusters, one of which also included all mainland North American samples. The Hawaii1 cluster represented 48 individuals, including all 44 from the island of Hawai'i, two samples from Kaua'i, one sample from Maui, and one from O'ahu. The haplotypes in this Hawaii1 cluster were endemic to the Hawaiian islands. The Hawaii2/North America cluster included eight samples from Maui, three from O'ahu, and all those (n = 85) from North America. Lasiurus cinereus semotus samples segregating within this cluster were very closely related to L. c. cinereus samples; in fact, one haplotype (labelled FJB18M in Fig 2) was found on Maui and O'ahu as well as in Nebraska, Ontario, Saskatchewan, California, Georgia, and Michigan (see S1 Table). The Hawaii1 and Hawaii2/North America clusters were clearly distinct from one another (Table 2), with an average pairwise distance of 19.98 differences between them. The single haplotype detected in South America was much more distinct from the other clusters in the network, with an average distance of 55.71 and 52.45 differences between it and the Hawaii1 and Hawaii2/North America clusters, respectively.
Each haplotype is represented by a circle, the relative size of which roughly corresponds to the haplotype frequency. The number of mutations between haplotypes is indicated only for those connections spanning more than one mutation. Hawaii1 haplotypes grey, Hawaii2/North America haplotypes white, South American haplotype black. Hawaiian islands haplotypes in bold.
A maximum likelihood analysis recovered a similar phylogenetic backbone (S1 Fig). The most likely topology described the South American subspecies L. c. villosissimus as sister to a clade containing both the Hawaii1 and Hawaii2/North America clades as defined in the network analysis. However, bootstrap support was low for some key internal branches, including those defining the Hawaii2/North America (75%) and Hawaii1 (70%) clades and the branch setting all L. c. cinereus and L. c. semotus samples as sister to South American L. c. villosissimus (60%). Topological tests (S2 Fig, S2 Table) were not able to reject any alternative topologies.
Historical demography of Hawaiian populations
Extended Bayesian skyline analyses supported different demographic histories for the Hawaii1 and Hawaii2 populations of L. c. semotus (Fig 3). Analysis of the Hawaii1 population supported a single period of growth starting around 10 kya (Fig 3A). Ancestral effective population size prior to demographic expansion was estimated at approximately 10,000 individuals, while the current effective population size was estimated to be approximately 132,000 individuals. We emphasize that this estimate of current Ne should not be interpreted as an estimate of the present-day census population size. The Hawaii2 population, on the other hand, is consistent with a model of population stasis with an effective size of approximately 21,000 individuals (Fig 3B). There does seem to be a slight increase in the median Ne starting approximately 800 years ago. The significance of this increase is unclear, but it is a consistent pattern across replicate runs.
Results of three runs are shown as gray lines, bounded by 95% confidence intervals in black. A. Skyline plot for the Hawaii1 population, showing a pattern of population growth starting at ca. 10 kya. B. Skyline plot for the Hawaii2 population, fitting a model of population stasis.
Biogeography of Hawaiian hoary bats
We show that Hawaiian populations of hoary bats derive from at least two independent dispersal events, both originating from populations in North America. The presence in the Hawaiian archipelago of two genetically distinct lineages is indicated by both phylogenetic and parsimony network analyses of mitochondrial DNA sequence data, and is further supported by multilocus demographic analyses characterizing the distinct evolutionary histories of these two Hawaiian lineages. Previous authors have also alluded to morphological differentiation between populations on the islands of Hawai'i and Maui .
Our results are consistent with the physiological model of Bonaccorso and McGuire , who predicted that multiple colonizations of Hawai'i by North American L. cinereus were possible based on prevailing physical conditions and energetics of migrating individuals. They are also consistent with other phylogeographic studies showing that species groups of some terrestrial Hawaiian taxa were founded by multiple historic colonizations, including ducks (Anas sp., ), succineid snails , and katydids , suggesting that multiple taxa have been able to repeatedly overcome the significant challenges associated with the colonization of these remote islands.
Our results point to the colonization of Hawai'i by hoary bats on two occasions by lineages that experienced distinct evolutionary trajectories. The Hawaii1 lineage was sampled primarily from the island of Hawai'i, but was also present in low numbers on O'ahu, Maui, and Kaua'i. EBSP analyses of this lineage support a model of population growth starting around 10 kya. A single fossil specimen described from Kaua'i confirms the presence of L. cinereus in the Hawaiian islands as early as 6760 yrBP , consistent with our genetically-derived dates. This lineage has diverged significantly from the mainland lineage (Table 2). Given the short branch lengths within L. cinereus reconstructed in our phylogeny, we infer that the demographic expansion detected for the Hawaii1 lineage in the EBSP analysis represents population growth following dispersal to Hawai'i. This inference is also supported by the star-like topology of this clade in the haplotype network (Fig 2).
In contrast, the Hawaii2 lineage shows no significant signal of population size change. Hawaii2 mitochondrial haplotypes are not clearly differentiated from North American haplotypes, with one shared haplotype (FJB18M) found at sampling locations across North America as well as on the islands of Maui and O'ahu (S1 Table). These results suggest that the Hawaii2 lineage represents a more recent dispersal to Hawai'i.
While we present the first examination of the colonization history and evolution of Hawaiian hoary bats, there are a number of opportunities for further research. First, the network and phylogenetic analyses were necessarily limited to the mtDNA data because most of our North American mitochondrial data were acquired via GenBank and we had access to only a limited number of tissue samples for generating new data from the continent. This also explains why our nuclear intron data were limited to Hawaiian samples. Future studies utilizing coalescent-based species tree/gene tree analyses of multilocus data [45,46] would be particularly informative for addressing taxonomic questions that are naturally raised by our study.
The fact that both the Hawaii1 and Hawaii2 lineages are found in apparent sympatry on Maui and O'ahu also requires further examination. Despite a strong sampling effort on the island of Hawai'i (n = 44 from 11 locations), the Hawaii2 lineage was not detected there. Because these clades are defined and differentiated solely by the mitochondrial COI gene, it is possible that the lack of congruent structure at the nuclear loci results from ongoing gene flow between the two populations (S3 Table). Given the recent timescales over which these dispersal events probably occurred, it is also possible that the lack of congruent structure results from incomplete lineage sorting at the nuclear loci . Future work examining patterns of structure at nuclear microsatellite loci or large numbers of single nucleotide polymorophism (SNP) markers would be informative for addressing these alternatives. Furthermore, the presence of the Hawaii1 lineage and, to a lesser extent, the Hawaii2 lineage on multiple islands suggests the occurrence of significant inter-island dispersal. Microsatellite genotyping and SNP typing, along with improved sampling of the older islands, would prove useful in quantifying rates of movement among islands. Ongoing work by our research group is addressing these questions.
Implications for hoary bat conservation
We strongly caution against the interpretation of our estimates of Ne as being indicative of the present-day census size for Hawaiian populations. If these populations were founded recently by individuals representative of the genetic diversity of the North American population, then diversity measures might well exceed that expected at mutation-drift equilibrium and our estimates of Ne would thus be much larger than the current census size [48,49]. As the founding population establishes and grows, it is possible that census size could eventually overtake effective size, but it is unclear how long that process would take.
Understanding the conservation genetics of hoary bats in Hawai'i is particularly timely and important because L. c. semotus is listed by the U.S. Fish and Wildlife Service  as Endangered, and the species recovery plan [40,51] cites molecular genetics as an area of data deficiency needed for adaptive management. Furthermore, both North American and Hawaiian hoary bats are under emerging threats from rapid expansion of the wind energy industry  with “take” mortalities having triggered mitigation actions in Habitat Conservation Plans for several wind facilities in Hawai'i and North America (data obtained by A. Pereira, April 2014, via a Freedom of Information Act request to the U.S. Fish and Wildlife Service).
Given the threats to the recovery of Hawaiian hoary bats, most notably those from wind energy development, barbed wire fences , and habitat loss , there is an urgent need for additional work on the conservation genetics of hoary bats to understand, among other things, their estimated effective population sizes, dispersal abilities, and genetic differentiation among and within islands. One area of importance is the resolution of potential evolutionary significant units that probably are represented by the Hawaii1 and Hawaii2 clades. The Endangered Species Act of 1973 (7 U.S.C. § 136, 16 U.S.C. § 1531 et seq.) defines species to include “any subspecies of fish or wildlife or plants, and any distinct population segment of any species of vertebrate fish or wildlife which interbreeds when mature”. Increasing field efforts to census and sample individuals from populations on islands other than the island of Hawai'i, combined with multilocus genotyping  of these and additional specimens should help to further clarify the taxonomic status and ESU designation of this island mammal.
S1 Fig. Maximum likelihood phylogeny for sampled L. cinereus subspecies, with bootstrap support measures.
Hawaiian hoary bats (L. c. semotus) cluster into two distinct clades, Hawaii1 and Hawaii2/North America, the latter of which is more closely related to mainland North American (L. c. cinereus) samples.
S2 Fig. Constraint topologies analyzed in the topological tests.
S1 Table. Location information for redundant COI haplotypes used in the network analysis.
S2 Table. Results of tests of alternative tree topologies.
Significance values are provided for approximately unbiased, Kishino-Hasegawa, and Shimodaira-Hasegawa tests.
S3 Table. Locus-specific AMOVAs quantifying genetic structure between individuals assigned to the Hawaii1 and Hawaii2 clades.
We thank R. Bernard, Y. Castaneda, N. Cortez-Delgado, A. Hart, A. Hubancheva, K. Lahaela, A. Miles, K. Montoya-Aiona, R. Moseley, C. Todd, B. Yuen, and V. Zrncic for assistance with field work in Hawai'i. Tissue collection from carcasses was facilitated by R. Breeden, M. Craig, F. Duvall, M. Hagemann, and T. Work. L. Kirby and J. Pontow assisted with cloning and sequencing of nuclear loci. We are grateful to the following people and institutions for providing access to continental specimens: American Museum of Natural History, Angelo State Natural History Collection, Carnegie Museum of Natural History, Centro de Investigaciones Biológicas del Noroestre, Louisiana State University Museum of Natural Science, Museum of Vertebrate Zoology at Berkeley, T. Alvarez, G. Amato, L. Ammerman, R. Benedict, R. Hersh, C. Lausen, K. Miner, H. Rice, N. Simmons, and C. Stihler. Permissions to conduct research and access lands under their stewardship were granted by Island Princess Estates, Hakalau Forest National Wildlife Refuge, and Laupahoehoe Hawai'i Experimental Tropical Forest, Hawai'i Department of Land and Natural Resources. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
Conceived and designed the experiments: ALR CAP MJV KJO FJB. Performed the experiments: ALR MJV KJO. Analyzed the data: ALR KJO. Contributed reagents/materials/analysis tools: CAP FJB. Wrote the paper: ALR CAP MJV KJO FJB.
- 1. Ziegler AC (2002) Hawaiian Natural History, Ecology, and Evolution. Honolulu: University of Hawaii Press.
- 2. Price JP, Clague DA (2002) How old is the Hawaiian biota? Geology and phylogeny suggest recent divergence. Proceedings of the Royal Society B: Biological Sciences 269: 2429–2435. pmid:12495485
- 3. Lerner HRL, Meyer M, James HF, Hofreiter M, Fleischer RC (2011) Multilocus resolution of phylogeny and timescale in the extant adaptive radiation of Hawaiian honeycreepers. Current Biology 21: 1838–1844. doi: 10.1016/j.cub.2011.09.039. pmid:22018543
- 4. Holland BS, Hadfield MG (2004) Origin and diversification of the endemic Hawaiian tree snails (Achatinellidae: Achatinellinae) based on molecular evidence. Molecular Phylogenetics and Evolution 32: 588–600. pmid:15223040
- 5. Cowie RH, Holland BS (2008) Molecular biogeography and diversification of the endemic terrestrial fauna of the Hawaiian Islands. Philosophical Transactions of the Royal Society of London B 363: 3363–3376. doi: 10.1098/rstb.2008.0061. pmid:18765363
- 6. Croucher PJP, Oxford GS, Lam A, Mody N, Gillespie RG (2012) Colonization history and population genetics of the color-polymorphic Hawaiian happy-face spider Theridion grallator (Araneae, Theridiidae). Evolution 66: 2815–2833. doi: 10.1111/j.1558-5646.2012.01653.x. pmid:22946805
- 7. Haines WP, Schmitz P, Rubinoff D (2014) Ancient diversification of Hyposmocoma moths in Hawaii. Nature Communications 5: e3502.
- 8. Shump KA, Shump AU (1982) Lasiurus cinereus. Mammalian Species 185: 1–5.
- 9. Hayman RW (1959) American bats reported in Iceland. Journal of Mammalogy 40: 245–246.
- 10. Hill JE, Yalden DW (1990) The status of the hoary bat, Lasiurus cinereus, as a British species. Journal of Zoology 222: 694–697.
- 11. Findley JS, Jones C (1964) Seasonal distribution of the hoary bat. Journal of Mammalogy 45: 461–470.
- 12. Simmons NB (2005) Order Chiroptera. In: Wilson DE, Reeder DM, editors. Mammal Species of the World: a Taxonomic and Geographic Reference. Washington, DC: Smithsonian Institution Press. pp. 312–529.
- 13. Baker RJ, Patton JC, Genoways HH, Bickham JW (1988) Genic studies of Lasiurus (Chiroptera: Vespertilionidae). Occasional Papers, The Museum, Texas Tech University 117: 1–16.
- 14. Morales JC, Bickham JW (1995) Molecular systematics of the genus Lasiurus (Chiroptera: Vespertilionidae) based on restriction-site maps of the mitochondrial ribosomal genes. Journal of Mammalogy 76: 730–749.
- 15. Jacobs DS (1996) Morphological divergence in an insular bat, Lasiurus cinereus semotus. Functional Ecology 10: 622–630.
- 16. Reimer JP, Baerwald EF, Barclay RMR (2010) Diet of hoary (Lasiurus cinereus) and silver-haired (Lasionycteris noctivagans) bats while migrating through southwestern Alberta in late summer and autumn. American Midlands Naturalist 164: 230–237.
- 17. Findley JS, Studier EH, Wilson DE (1972) Morphologic properties of bat wings. Journal of Mammalogy 53: 429–444.
- 18. Cryan PM (2003) Seasonal distribution of migratory tree bats (Lasiurus and Lasionycteris) in North America. Journal of Mammalogy 84: 579–593.
- 19. Bonaccorso FJ, McGuire LP (2013) Modeling the colonization of Hawaii by hoary bats (Lasiurus cinereus). In: Adams R, Pedersen SC, editors. Bat Evolution, Ecology, and Conservation. New York: Springer Science Press. pp. 187–206.
- 20. Worthington Wilmer J, Barratt E (1996) A non-lethal method of tissue sampling for genetic studies of chiropterans. Bat Research News 37: 1–3.
- 21. Nadin-Davis SA, Guerrero E, Knowles MK, Feng Y (2012) DNA barcoding facilitates bat species identification for improved surveillance of bat-associated rabies across Canada. Open Zoology Journal 5: 27–37.
- 22. Olival KJ (2008) Population genetic structure and phylogeography of Southeast Asian flying foxes: implications for conservation and disease ecology: Columbia University.
- 23. Hebert PDN, Cywinska A, Ball SL, deWaard JR (2003) Biological identifications through DNA barcodes. Proceedings of the Royal Society B: Biological Sciences 270: 313–321. pmid:12614582
- 24. Streicker DG, Turmelle AS, Vonhof MJ, Kuzmin IV, McCracken GF, Rupprecht CE (2010) Host phylogeny constrains cross-species emergence and establishment of rabies virus in bats. Science 329: 676–679. doi: 10.1126/science.1188836. pmid:20689015
- 25. Venta PJ, Brouillette JA, Yuzbasiyan-Gurkan V, Brewer GJ (1996) Gene-specific universal mammalian sequence-tagged sites: application to the canine genome. Biochemical Genetics 34: 321–341. pmid:8894053
- 26. Stadelmann B, Lin LK, Kunz TH, Ruedi M (2007) Molecular phylogeny of New World Myotis (Chiroptera, Vespertilionidae) inferred from mitochondrial and nuclear DNA genes. Molecular Phylogenetics and Evolution 43: 32–48. pmid:17049280
- 27. Posada D (2011) Collapse: describing haplotypes from sequence alignment.
- 28. Stamatakis A (2014) RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30: 1312–1313. doi: 10.1093/bioinformatics/btu033. pmid:24451623
- 29. Posada D (2008) jModelTest: phylogenetic model averaging. Molecular Biology and Evolution 25: 1253–1256. doi: 10.1093/molbev/msn083. pmid:18397919
- 30. Shimodaira H (2002) An approximately unbiased test of phylogenetic tree selection. Systematic Biology 51: 492–508. pmid:12079646
- 31. Shimodaira H, Hasegawa M (1999) Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Molecular Biology and Evolution 16: 1114–1116.
- 32. Kishino H, Hasegawa M (1989) Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Hominoidea. Journal of Molecular Evolution 29: 170–179. pmid:2509717
- 33. Shimodaira H, Hasegawa M (2001) CONSEL: for assessing the confidence of phylogenetic tree selection. Bioinformatics 17: 1246–1247. pmid:11751242
- 34. Clement M, Posada D, Crandall K (2000) TCS: a computer program to estimate gene genealogies. Molecular Ecology 9: 1657–1660. pmid:11050560
- 35. Templeton AR, Crandall KA, Sing CF (1992) A cladistic analysis of phenotypic associations with haplotypes inferred from restriction endonuclease mapping and DNA sequence data. III. Cladogram estimation. Genetics 132: 619–633. pmid:1385266
- 36. Heled J, Drummond AJ (2008) Bayesian inference of population size history from multiple loci. BMC Evolutionary Biology 8: e289.
- 37. Tamura K, Nei M (1993) Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution 10: 512–526. pmid:8336541
- 38. Hasegawa M, Kishino H, Yano T-a (1985) Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22: 160–174. pmid:3934395
- 39. Hickerson MJ, Meyer CP, Moritz C (2006) DNA barcoding will often fail to discover new animal species over broad parameter space. Systematic Biology 55: 729–739. pmid:17060195
- 40. U.S. Fish and Wildlife Service (2011) Ope'ape'a or Hawaiian hoary bat (Lasiurus cinereus semotus): 5-year review summary and evaluation. In: U.S. Fish and Wildlife Service, editor. Honolulu, HI. pp. 13.
- 41. Fleischer RC, McIntosh CE (2001) Molecular systematics and biogeography of the Hawaiian avifauna. Studies in Avian Biology 22: 51–60.
- 42. Holland BS, Cowie RH (2009) Land snail models in island biogeography: a tale of two snails. American Malacological Bulletin 27: 59–68.
- 43. Shapiro LH, Strazanac JS, Roderick GK (2006) Molecular phylogeny of Banza (Orthoptera: Tettigoniidae), the endemic katydids of the Hawaiian Archipelago. Molecular Phylogenetics and Evolution 41: 53–63. pmid:16781170
- 44. Burney DA, James HF, Pigott Burney L, Olson SL, Kikuchi W, Wagner WL (2001) Fossil evidence for a diverse biota from Kaua'i and its transformation since human arrival. Ecological Monographs 71: 615–641.
- 45. Ence DD, Carstens BC (2011) SpedeSTEM: a rapid and accurate method for species delimitation. Molecular Ecology Resources 11: 473–480. doi: 10.1111/j.1755-0998.2010.02947.x. pmid:21481205
- 46. Yang Z, Rannala B (2010) Bayesian species delimitation using multilocus sequence data. Proceedings of the National Academy of Sciences, USA 107: 9264–9269.
- 47. Maddison WP, Knowles LL (2006) Inferring phylogeny despite incomplete lineage sorting. Systematic Biology 55: 21–30. pmid:16507521
- 48. Crandall KA, Posada D, Vasco D (1999) Effective population sizes: missing measures and missing concepts. Animal Conservation 2: 317–319.
- 49. Luikart G, Ryman N, Tallmon DA, Schwartz MK, Allendorf FW (2010) Estimation of census and effective population sizes: the increasing usefulness of DNA-based approaches. Conservation Genetics 11: 355–373.
- 50. U.S. Fish and Wildlife Service (1970) Conservation of endangered species and other fish or wildlife. Federal Register. pp. 16047–16048.
- 51. U.S. Fish and Wildlife Service (1998) Recovery plan for the Hawaiian hoary bat. In: U.S. Fish and Wildlife Service, editor. Portland, OR. pp. 1–50.
- 52. Cryan PM, Jameson JW, Baerwald EF, Willis CKR, Barclay RMR, Snider EA, et al. (2012) Evidence of late-summer mating readiness and early sexual maturation in migratory tree-roosting bats found dead at wind turbines. PLoS ONE 7: e47586. doi: 10.1371/journal.pone.0047586. pmid:23094065
- 53. Zimpfer J, Bonaccorso F (2010) Barbed wire fences and Hawaiian hoary bats: what we know. Hawaii Conservation Conference. Honolulu, HI.
- 54. Tomich PQ (1986) Mammals in Hawaii. Honolulu: Bishop Museum Press. 375 p.
- 55. Knowles LL, Carstens BC (2007) Delimiting species without monophyletic gene trees. Systematic Biology 56: 887–895. pmid:18027282