Do fly parasites of bats and their hosts coevolve?speciation in Trichobius phyllostomae group (Diptera, Streblidae) and their hosts (Chiroptera, Phyllostomidae) suggests that they do not

Moscas parasitas de morcegos e seus hospedeiros co-evoluem?especiação do grupo Trichobius phyllostomae (Diptera, Streblidae) e seus hospedeiros (Chiroptera, Phyllostomidae) sugere que eles não fazem

SYSTEMATICS, MORPHOLOGY AND BIOGEOGRAPHY

Do fly parasites of bats and their hosts coevolve? Speciation in Trichobius phyllostomae group (Diptera, Streblidae) and their hosts (Chiroptera, Phyllostomidae) suggests that they do not

Moscas parasitas de morcegos e seus hospedeiros co-evoluem? Especiação do grupo Trichobius phyllostomae (Diptera, Streblidae) e seus hospedeiros (Chiroptera, Phyllostomidae) sugere que eles não fazem

Gustavo Graciolli^I; Claudio J. B. de Carvalho^II

^IDepartamento de Biologia, Centro de Ciências Biológicas e da Saúde, Universidade Federal de Mato Grosso do Sul, Cidade Universitária, Caixa Postal 549, 79070-900 Campo Grande-MS, Brasil. [e-mail]

^IIDepartamento de Zoologia, Universidade Federal do Paraná, Caixa Postal 19020, 81531-980 Curitiba-PR, Brasil. [e-mail]

ABSTRACT

We examined whether, like many parasite-host systems of coevolution, a group of obligate parasitic bat flies (Trichobius phyllostomae Kessel and related species) cospeciate with their hosts. We first did a cladistic analysis of the T. phyllostomae group and combined that analysis with a phylogenetic hypothesis from the literature for the Stenodermatinae bats. The cladistic analysis included, as outgroups, one species from each morphological group and complex of Trichobius Gervais, and one species from the following genera: Paratrichobius Miranda-Ribeiro, Megistopoda Macquart, Megistapophysys Dick & Wenzel, Neotrichobius Wenzel & Aitken, Speiseria Kessel and Strebla Wiedemann. The cladogram was rooted with a species of Strebla in the subfamily Streblinae. One cladogram was obtained and which found Trichobius to be polyphyletic. The phylogenetic hypothesis as follows: (Paratrichobius, (Neotrichobius, (Megistopoda, Megistapophysis)))) is the sister-group of the phyllostomae group and the following relationships within the ingroup, (((T. vampyropis Wenzel, Trichobius sp. 2) ((T. hispidus Wenzel, T. petersoni Wenzel) ((Trichobius sp. 1 (T. phyllostomae, T. brennani Wenzel))))). When we compared phylogenies through historical association analyses, cospeciation was uncommon, while host-switching was more common and better explained the association between the phyllostomae group and their bat hosts.

Keywords: Bat fly; cospeciation; macroevolutionary events; Trichobiinae; Sturnira.

RESUMO

Nós examinamos, como em muitos sistemas parasito-hospedeiro de coevolução, se um grupo de moscas parasitas obrigatórias de morcegos (Trichobius phyllostomae Kessel e espécies relacionadas) coespecia com seus hospedeiros. Primeiro, nós fizemos uma análise cladística do grupo T. phyllostomae e combinamos a análise com a hipótese filogenética da literatura para os morcegos da subfamília Stenodermatinae. A análise cladística incluiu, como grupo-externo, uma espécie de cada grupo morfológico de Trichobius Gervais, e uma espécie dos seguintes gêneros: Paratrichobius Miranda-Ribeiro, Megistopoda Macquart, Megistapophysys Dick & Wenzel, Neotrichobius Wenzel & Aitken, Speiseria Kessel e Strebla Wiedemann. O cladograma foi enraizado com a espécie de Strebla da subfamília Streblinae. Um cladograma foi obtido e no qual encontramos Trichobius como polifilético. A hipótese filogenética a seguir: (Paratrichobius, (Neotrichobius, (Megistopoda, Megistapophysis)))) é grupo-irmão do grupo phyllostomae e a relação dentro do grupo-interno a seguir, (((T. vampyropis Wenzel, Trichobius sp. 2) ((T. hispidus Wenzel, T. petersoni Wenzel) ((Trichobius sp. 1 (T. phyllostomae, T. brennani Wenzel))))). Quando comparamos as filogenias através de análises de associação histórica, co-especiação foi pouco comum, enquanto colonização de um novo hospedeiro foi mais comum e explicou melhor a associação netre o grupo phyllostomae e seus hospedeiros.

Palavras-chave: Dípteros ectoparasitos de morcegos; co-especiação; eventos macroevolucionários; Trichobiinae; Sturnira.

The history of host-parasite associations is often studied through reconciliation analysis and similar methods (Page 2003). For example, the Phthiraptera often have been used in this type of study because the entire lice life cycle is on their host and so their host specificity and alpha-taxonomy are very well examined and known (Table I). Although Phthiraptera is a good model system, there are some other groups that have the same potential to be a model system. Widening host-parasite analysis to include candidate groups, such as Streblidae, will enhance our knowledge of the coevolutionay process by testing hypothesis in a broad context. Of such groups, a few studies examine historical associations among fleas and ticks (parasites with a free life stage) and their hosts (Cumming 2000; Krasnov & Shenbrot 2002, respectively, Table I) and no one has yet been done in bat fly-bat system.

Bat flies (Diptera, Nycteribiidae and Streblidae) are blood-feeding, obligate ectoparasites (Dick & Miller 2010; Graciolli 2010). Of these, the streblid bat flies are mostly tropical and more species are found in the New World where their main hosts are bats in the family Phyllostomidae. In the fly family Streblidae, the American genus Trichobius Gervais, 1844 has the greatest number of species and subspecies. The 68 species and subspecies of Trichobius were divided into morphological groups and complexes by Wenzel et al. (1966) and Wenzel (1976) using mostly male and female post-abdomen characteristics for their classification, and in which each group or complex was monophyletic although the genus was not. This perspective places the phyllostomae group as a sister-group of the genera Paratrichobius Costa-Lima, 1921, Megistopoda Coquillett, 1899 and Neotrichobius Wenzel & Aitken, 1966 (Fig. 1).

A recent phylogenetic analysis of bat flies using molecular data supported the polyphyly of Trichobius, although the species of the dunni, pallidus, phyllostomae and uniformis groups and other streblid genera, with exception of Strebla Wiedemann, 1824 were not included in that analysis (Dittmar et al. 2006). In opposition to polyphyly of Trichobius, Jobling (1938) and Guerrero (1997) considered the genus a natural group and Guerrero (1997) further suggested that each morphological group within Trichobius should be considered as a subgenus. In regards to the phyllostomae group, T. vampyropis Wenzel, 1966 was related to T. hispidus Wenzel, 1976 while T. petersoni Wenzel, 1976 and T. phyllostomae Kessel, 1925 were closer to T. brennani Wenzel, 1966 (Wenzel 1976).

The five species of the phyllostomae group (not including two undescribed species) have been found above 800 meters elevation in Central America and northern South America and southern Brazil. These species are host-specific parasites of bats in the genera Sturnira Gray, 1842 and Platyrrhinus Saussure, 1860 (Fig. 2) (Wenzel et al. 1966; Wenzel 1976), both in the subfamily Stenodermatinae (Phyllostomidae). In that subfamily there are 15 species in two subgenera (Simmons 2005), all of which are frugivorous (Molinari & Soriano 1987; Gannon et al. 1989; Giannini 1999). Most species roost in the forest canopy or in hollow trees, while one species, Sturnira bidens (Thomas, 1915) uses caves (Molinari & Soriano 1987). Species of Sturnira may often be sympatric and syntopic (Tamsitt & Häuser 1985; Soriano & Molinari 1987; Gannon et al. 1989; Pacheco & Patterson 1992; Giannini 1999). The genus Platyrrhinus includes 14 species restricted to the Neotropics (Velazco 2005). All species are frugivorous and roost in the canopy or hollow trees (Novak 1994).

Here, we develop a phylogenetic hypothesis for the Trichobius phyllostomae group and examine their associations with their hosts using a historical perspective by means of reconciliation analysis.

MATERIAL AND METHODS

Host phylogeny. We used the phylogenetic hypotheses for the genera of stenodermatine bats of Wetterer et al. (2000) as a template. The phylogenetic hypothesis of Sturnira (Villalobos & Valerio 2002) and Platyrrhinus (Velazco 2005) were merge to the template phylogeny to improve historical scenarios.

Parasite (phyllostomae group) phylogeny. To achieve a parsimonious phylogenetic hypothesis for Trichobius (and thereby allow for the possibility of paraphyly) to associate with the bat flies, we included as outgroups one species from each morphological group and complex of Trichobius, including Trichobius parasiticus Gervais, 1844 the type species of the genus, and one species of the genera Paratrichobius, Megistopoda, Megistapophysis Dick & Wenzel, 2006, Neotrichobius, and Speiseria Kessel, 1925. To root the cladogram we used a species of Strebla Wiedemann, 1844 in the subfamily Streblinae (Appendix 1). Nexus 0.5.0 (Page 2001) was used for the morphological data matrix, which comprised 44 characters (19 binary, 25 multistate) of which one was polymorphic.

Phylogenetic analysis used NONA (Goloboff 1998) and Winclada ver. 1.00.08 (Nixon 2002). For tree reconstruction, all characters were equally weighted and unordered (non additive) with the following commands: Tree- bisection-reconnection (TBR) branch-swapping with 10,000 heuristic random addition replicates. Bootstrapping with 1,000 replicates with random addition with TBR branch-swapping was used to examine the resulting tree. To character optimization we used Acctran.

The specimens examined in this study were lent by Coleção de Entomologia Padre Jesus Santiago Moure, Curitiba, Brasil (DZUP); Museu de Zoologia da Universidade de São Paulo, São Paulo, Brasil (MZSP); Field Museum of Natural History (FMNH), Chicago, USA.

Cophylogenetic Analyses. To compare host and parasite trees we used a method reconciliation in which an event-based concept is used where each event is assigned a cost and costs are minimized in the program TARZAN 0.9 (Merke & Middendorf 2005). The events that are used by TARZAN are cospeciations, duplications, sortings, extinctions, and (host) switching. The cost assignment used for each cophylogentic event was: cospeciation = 2, duplication = 2, sorting = 1, switch = 2. TARZAN reconciliation analysis does not allowed for multi-host associations.

To avoid "noise" in our analysis we considered two terminal branches for the species of Platyrrhinus (lowland and andean) following Velazco (2005), because Trichobius vampyropis and Trichobius sp. 2 are each associated with > 1 species of Andean Platyrrhinus. Trichobius petersoni is associated with two species of Sturnira. We considered T. petersoni as two populations (T. petersoni 1 and T. petersoni 2), each with its own associated host species.

RESULTS

Phylogeny of phyllostomae group. One most parsimonious tree was found (Fig. 3), with 144 steps, consistence index 65 and retention index 77. Eight out of 21 branches in this tree were recovered in > 70% of the bootstrap replicates (Fig. 4). The genus Trichobius is not a monophyletic group, but rather is separated into three clades (Fig. 3). The first clade is formed by T. tiptoni Wenzel, 1976, T. parasiticus Gervais, 1844 (dugesii group), T. lonchophyllae Wenzel, 1966 (uniformis group), T. longipes (Rudow, 1871) (longipes group), and T. jubatus Wenzel, 1976 (dunni group) (Fig. 3). The second clade includes T. major Coquillett, 1899 (major group), T. galei Wenzel, 1966 (caecus group), T. pallidus (Curran, 1934) (pallidus group). The remaining species of phyllostomae group are the sister-group of Paratrichobius longicrus (Miranda-Ribeiro, 1907), Megistopoda aranea (Coquillett, 1899), Megistapophysis mordax Wenzel & Dick, 2006, and Neotrichobius delicatus (Machado-Allison, 1965) (Fig. 3). Synapomorphies shared among the phyllostomae group (P. longicrus, (N. delicatus, (M. aranea, M. modax))) are pointed maxillar palp, laterovertex without setae in front of the eyes, eyes having more than 18 facets, and transverse

and elliptical tergite 7 (Fig. 3). This branch has a support of 66% in the bootstrap replicates (Fig. 4). The species of the phyllostomae group are monophyletic, supported by two synapomorphies, which are the long, triangular metasternal lobe and the spiniform setae that are shorter than the tarsomere on the ventral side of basitarsum 3 (Fig. 3). These had robust branch support in > 80% of the bootstrap replicates (Fig. 4).

We presented below characters and character states with length (L), consistence index (CI), and retention index (RI) used in the cladistic analysis of the phyllostomae group.

1. Apex of maxillar palpus: (0) wide and irregular; (1) with a small concavity; (2) rounded; (3) pointed. L = 4; CI = 75, RI = 85.

2. Shape of laterovertex: (0) Streblinae-like; (1) trapezoid; (2) subtriangular; (3) trapezoid with a posterior lobe on inner margin; (4) trapezoid with a small posterior lobe on inner margin; (5) transverse. L = 6; CI = 83, RI = 87.

The long posterior lobe (state 3) is found in dugesii (T. parasiticus and T. tiptoni), uniformis (T. lonchophyllae) and dunni (T. jubatus) and is very evident and more developed in the brachypterous genera Aspidoptera Coquillett, 1899 and Exastinion Wenzel, 1966. Wenzel (with collaborators 1966, 1976) considered the dugesii group to be closely related to these genera (Fig. 1).

3. Number of short setae on anterior margin of laterovertex in front of the eye: (0) 1; (1) 3; (2) more than 3; (3) 2; (4) 0. L = 4; CI = 100, RI = 100 (

Figs. 6-9

).

Setae in front of the eyes seems to be a plesiomorphic character in Trichobiinae that is lost in the clade formed by Paratrichobius, Neotrichobius, Megistopoda, Megistapophysis, and phyllostomae group.

4. Setae 6 of laterovertex: (0) present; (1) absent. L = 3; CI = 33, RI = 75.

Several setae can be found in the laterovertex of all Trichobiinae, but some apparently have fixed positions and similar shapes. Here we tentatively consider these setae homologous and make a map of the setae. We name the setae 1 to 6 (Figs. 10-12). Only setae 6 may be absent in Speiseria and homoplasically in Megistapophysis mordax and the species of the phyllostomae group. It is not possible to make homologous associations of the laterovertex setae of Strebla wiedemanni, because they have various extra setae. Therefore we considered these as "-" (inapplicable) for this taxon.

5. Accessory setae on posterior margin of laterovertex: (0) absent; (1) 1; (2) 2; (3) 4; (4) 5. L = 6; CI = 66, RI = 33.

Normally, one accessory setae is found in the Trichobiinae. However, a single accessory setae can also be found in other genera, such as Aspidoptera and Exastinion. The number of accessory setae may be a good diagnostic character for generic or specific levels.

6. Number of eye facets: (0) 9 to 15; (1) one; (2) greater than 18. L = 3; CI = 66, RI = 87.

7. Occipital lobes: (0) sclerotized; (1) lesser sclerotized. L = 1; CI = 100, RI = 100.

8. Longitudinal and transverse sutures: (0) not united; (1) united. L = 4; CI = 25, RI = 25.

9. Number of episternal setae: (0) 2; (1) 3; (2) 4; (3) about 14; (4) 6; (5) 5. L = 7; CI = 71; RI = 50 (Figs. 22-28).

10. Setae next to episternal cleft: (0) one long and one short setae; (1) two long setae; (2) one long and two short setae; (3) two long and one short setae; (4) one long setae; (5) three long setae. L = 7; CI = 71; RI = 50. (Figs. 22-28).

Zeve & Howell (1963) studied the external thorax morphology in three species of the major group, describing the pattern of episternal setae, but they did not comment on chaetotaxy. Aside from that study, episternal setae have not been cited as being used for generic or specific identification. This character might be useful for specific identification, especially for the phyllostomae group. The state 0 (Fig. 22) is basal in Trichobiinae, in which length is modified and setae may be lost (Fig. 3), especially in ((T. major, (T. galei, T. pallidus)), ((P. longicrus, ((M. aranea, M. mordax) N. delicatus)), ((Trichobius sp. 2, T. vampyropis), ((T. petersoni, T. hispidus), (Trichobius sp. 1, (T. phyllostomae, T. brennani)))))).

11. Thoracic sternum: (0) with longitudinal bare areas on each side; (1) setose; (2) almost bare. L = 2; CI = 100, RI = 100.

12. Anterior margin of male thoracic sternum in ventral view: (0) past coxae 1 and over base of trochanter; (1) past the insertion of the trochanter; (2) not on insertion of the trochanter; (3) past and over coxae 1. L = 7; CI = 42, RI = 55.

13. Shape of the anterior margin of thoracic sternum: (0) concave; (1) truncate; (2) blunt. L = 3; CI = 66, RI = 80.

14. Posterior margin of the metasternum in ventral view: (0) not expanded over base of coxae 3, rounded margin; (1) not expanded over base of coxae 3, truncate margin; (2) expanded over base of coxa 3. L = 3; CI = 66, RI = 80.

15. Anterior margin of female thoracic sternum in ventral view: (0) past the coxae 1 and on base of trochanter; (1) past the trochanter insertion; (2) not on trochanter insertion; (3) past and on coxa 1. L = 6; CI = 50, RI = 62.

16. Metasternal lobe: (0) absent; (1) blunt and trapezoid; (2) united with metepimere; (3) long and triangular; (4) short. L = 5; CI = 80, RI = 90. (Figs. 13-16).

The metasternal lobe was considered by Wenzel (and collaborators 1966, 1976) to be an important character to identify the groups and complexes of Trichobius. The blunt, trapezoidal metasternal lobe arose three times independently (Figs. 2, 14). The lobe blunt and trapezoid have arisen homoplasically in T. tiptoni (dugesii group) and P. longicrus. In T. phyllostomae, long and triangular metasternal lobe is a synapomorphy (Figs. 2, 15).

17. Ventral face of coxa 3: (0) setose; (1) not setose. L = 1; CI = 100, RI = 100.

18. Diagonal rows of setae on inner surface of femur 1: (0) rows with similar setae; (1) inner row with thick spines. L = 1; CI = 100, RI = 100.

19. Number of setae on inner surface of femur 1 whose apex is directed towards anterior surface: (0) none; (1) 2; (2) 3; (3) more than 3. L = 3; CI = 100, RI = 100.

20. Lateroventral setae on femur 2: (0) as long as other setae: (1) longer and thick, similar to the long dorsal setae. L = 2; CI = 50, RI = 85.

21. Unpigmented transverse ventral stripe on femur 2: (0) absent; (1) present. L = 1; CI = 100, RI = 100.

22. Unpigmented transverse ventral stripe on femur 3: (0) absent; (1) present. L = 2; CI = 50, RI = 75.

23. Tibiae 1: (0) with anterolateral and posterolateral rows of spiniform setae higher than tibiae 1; (1) with an anterolateral row of setae that are as long as the dorsal setae; (2) with an anterolateral row of spiniform setae shorter than tibia 1. L = 2; CI = 100, RI = 100.

24. Tarsomere: (0) rectangular; (1) moniliform. L = 1; CI = 100, RI = 100.

25. Thorn-like setae on ventral surface of basitarsum 1: (0) as long as the basitarsum; (1) absent; (2) shorter than the basitarsum. L = 3; CI = 66, RI = 85.

26. Thorn-like setae on ventral surface of basitarsum 2: (0) as long as the basitarsum; (1) absent; (2) shorter than the basitarsum. L = 3; CI = 66, RI = 85.

27. Thorn-like setae on ventral surface of basitarsum 3: (0) two rows; (1) as long as the basitarsum; (2) absent; (3) shorter than the basitarsum; (4) two setae. L = 4; CI = 100, RI = 100.

28. Length of basitarsum 3: (0) as long as the two preceding tarsomeres together; (1) as long as the preceding tarsomere; (2) as long as the three preceding tarsomeres together. L = 4; CI = 50, RI = 60.

29. Wing: (0) macropterous; (1) stenopterous. L = 1; CI = 100, RI = 100.

Other genera of Trichobiinae, for example Aspidoptera, Joblingia Dybas & Wenzel, 1947, and Exastinion, have reduced wings. Normally these wings are oval-shaped or as wide as long and are called brachypterous (Wenzel et al. 1966, Wenzel 1976). In Megistopoda, Neotrichobius and Megistapophysis the wings are much longer than wide.

30. R vein: (0) straight; (1) sinuose, curved or bent. L = 2; CI = 50, RI = 50. (

Figs. 18-22

).

 

Although the sinuose R vein appears independently in T. dugesii and T. major groups, it is an important synapomorphy that groups the species of the T. dugesii group.

31. M1+2 vein: (0) setose throughout its length; (1) one long basal setae follow a bare area; (2) no basal setae. L = 5; CI = 40, RI = 40. (Figs. 17-21).

32. 1A(2A) vein: (0) basal setae absent; (1) setose throughout its length; (2) basal half without setae. L = 3; CI = 66, RI = 50. (Figs. 17-21).

Most species of Trichobiinae and Streblinae have no basal setae (state 0).

33. Dorsal abdominal connexivum: (0) almost totally bare; (1) setose on distal half; (2) almost totally setose. L = 2; CI = 100, RI = 100.

34. Male with long setae on the lateral abdominal connexivum posterior to sintergite 1+2 lobule: (0) absent; (1) present. L = 1; CI = 100, RI = 100.

35. Sternite VI: (0) wide; (1) tread-like; (2) absent. L = 4; CI = 50, RI = 60.

36. Gonopods: (0) symmetrical; (1) asymmetrical. L = 2; CI = 50, RI = 75.

Asymmetrical gonopods seem to be common in Trichobiinae, contrary to Streblinae, in which gonopods are symmetrical. Here, S. ambigua and the clade (T. major (T. galei, T. pallidus)) have symmetrical gonopods. Other taxa in the Trichobiinae that have symmetrical gonopods are the brachypterous genera, such as Aspidoptera, Exastinion, Noctiliostrebla Wenzel, and the full-winged Paratrichobius americanus Peterson & Ross, 1972 (Peterson & Ross 1972).

37. Gonopophyseal setae: (0) proximal seta longer than distal; (1) distal seta longer than proximal. L = 2; CI = 50, RI = 80.

38. Female with long setae on the lateral abdominal connexivum posterior to the sintergite 1+2 lobule: (0) absent; (1) a transversal row with three to four; (2) a patch. L = 5; CI = 40, RI = 57.

39. Shape of tergite 7: (0) pentagonal; (1) inverted drop-like with truncate posterior margin; (2) small and unsclerotized; (3) trapezoid; (4) pallidus form; (5) transverse and elliptical; (6) two elliptical sclerites; (7) absent. L = 8; CI = 87, RI = 80.

40. Tergite 7 and epiproct: (0) not united; (1) united. L = 4; CI = 25, RI = 50.

In addition to the taxa studied here, some species of Strebla and Anastrebla Wenzel, 1966 (Streblinae) also have united tergite 7 and epiproct. Perhaps this character is more useful for separating species than genera.

41. Setae across spiracle 7 in females. (0) present; (1) absent. L = 3; CI = 33; RI = 77.

42. Spiniform setae on posterior margin of sternite 2. (0) absent; (1) present. L = 1, CI = 100; RI = 100.

43. Female cerci: (0) free; (1) united with ventral arc. L = 2; CI = 50; RI = 80.

44. Postgenital sclerite: (0) absent; (1) present. L = 1; CI = 100; RI = 100.

Cophylogenetic Analyses. The cophylogenetic analysis yielded 813 scenarios, with costs ranging between 13 to 57. The number of cospeciations events in these scenarios ranged from 0 - 5, with 1 - 4 duplications, 0 - 6 host switches, 0 - 23 sorting events and 0 - 3 extinctions. The six scenarios with the lower cost (< 13) included two cospeciations events, one duplication event, four host-switches events and one sorting event (Fig. 5). Three scenarios showed that the phyllostomae group arose from the Sturnira clade (Figs. 5A, B, F) and three arose on the Platyrrhinus Andean group clade (Figs. 5C, D, E). In all scenarios, T. phyllostomae underwent speciation in moving to Sturnira lilium (E. Geoffroy, 1810) from S. ludovici Anthony, 1924 (Fig. 5).

 

DISCUSSION

Phylogeny of the T. phyllostomae group. The topological tree (Fig. 3) supported the hypothesis of relationships of the genera of Trichobiinae made by Wenzel et al. (1966) and Wenzel (1976) (Fig. 1). The major, pallidus, caecus, and phyllostomae groups and Paratrichobius, Megistopoda, Megistapophysis, and Neotrichobius are in the same large clade, while dugesii, dunni and longipes groups are included in another monophyletic group. However, contrary to one of the hypotheses of Wenzel et al. (1966), the uniformis group is unrelated to Speiseria, which indicates that some characters, such as setae on the base of the 1A(2A) vein (Fig. 17) and longer distal than basal gonapophyseal setae, may be homoplasic. Wenzel et al. (1966) also suggested the possibility that Speiseria may be related to phyllostomae group, which he later rejected (Wenzel 1976) and which is also not supported here.

The shape of thoracic sternum and the reduced wing are similar in Megistapophysis and the T. phyllostomae group, Paratrichobius, Megistopoda and Neotrichobius (Dick & Wenzel 2006). Thus, the following relationship among genera was predicted: (T. phyllostomae group (Paratrichobius (Megistapophysis (Megistopoda (Neotrichobius))))). Here, Megistapophysis and Megistopoda form a monophyletic group whose main synapomorphy is the ctenidium in the second abdominal sternite (Fig. 3).

Contrary to the morphological grouping of Trichobiinae, our results contradict the topology of Dittmar et al. (2006) based on molecular data. In their analysis, the dugesii group is paraphyletic, with the dugesii complex close to the caecus group, while the parasiticus complex is closer to the longipes group and T. hirsutulus Bequaert, 1933, a species of the major group. Here, we suggest that uniformis, dugesii, longipes, and dunni groups belong to the genus Trichobius, although the clade (T. major (T. galei, T. pallidus)) had low branch support (Fig. 4). The major group will require a more comprehensive analysis that includes other genera of the Trichobiinae to determine its status. Such an analysis should include the genera Joblingia and Anatrichobius because of the morphological similarity of the female terminalia among these genera. Species of the caecus and pallidus groups probably do not belong to Trichobius, but we prefer to wait for a more detailed study before renaming caecus group and re-validating Kesselia Curran, 1934 to comprise T. pallidus. A new genus should be given to the phyllostomae group. Thus, further study using both, morphological and molecular data, is required and should include all genera in the Trichobiinae for correct delimitation of Trichobius.

According to Wenzel (1976), T. phyllostomae is a sister-group of T. brennani, and T. petersoni is a sister-group of T. hispidus which is in turn a sister-group T. vampyopis. Here, we found that the species that are found on Sturnira form a monophyletic group whose sister group are found on Platyrrhinus.

Cophylogenetic Analysis. If host species are sympatric and syntopic, the likelihood increases for dispersal to different hosts. If survival and reproduction on these new hosts are reasonable, host-switching becomes much more likely (Clayton et al. 2003). Although species in the phyllostomae group have strong host specificity, which should favor cospeciation, and low prevalence and intensity, which should favor sorting, cospeciation and sorting are seldom encountered. Streblid bat flies are pupiparous, which means that instead of eggs, females deposit pupae around the bat roosting sites. Thus, in bat flies, host switching may have occurred during their coevolutionary histories when more than one host bat roosted in the same roosting site. Flies in the phyllostomae group are strong fliers, which also increases their chances of encountering a new host species. Nonetheless, because of host grooming behaviors and low rates of infestation (small population sizes of the flies), host switching may be minimal.

Grooming is an important cause of ectoparasite mortality and may reduce or eliminate colonization of new hosts (Marshall 1981). However, bats often have flies, so grooming may not be effective for bat flies and the importance of grooming for bat flies has not been studied. For example, Noctilio leporinus (Linnaeus, 1758) is often strongly infested by Noctiliostrebla Wenzel, 1966 and Paradyschiria Speiser, 1900 (Wenzel et al. 1966; Moura et al. 2003). Paradyschiria fusca Speiser, 1900 and Noctiliostrebla aitkeni Wenzel, 1966 remains were found in 10% and 2% (respectively) of the feces of N. leporinus (Bordignon 2006).

Sturnira evolved from a highland ancestor (Villalobos & Valerio 2002) such as the Andean Platyrrhinus (Velazques 2005). Platyrrhinus spp. and Stunira bidens (Thomas, 1915) have been found syntopically (Molinari & Soriano 1987; Lidner & Morawetz 2006) and both often use caves for roosting (Gardner 2007; Molinari & Soriano 1987). Other species of Sturnira do not roost in caves, but rather use tunnels, bridges and houses (Molinari & Soriano 1987). These roosting behaviors suggest that flies have had many opportunities for host switching (Fig. 5). Off-host behavior and longevity of the T. phyllostomae group, or of most Neotropical streblid bat flies, is unknown and should be studied to better understand how switching may occur.

Speciation in T. phyllostomae was the result of a switch from S. ludovici to S. lilium. Both bats are sympatric in Central America and northern South America, but T. phyllostomae is from southern Brazil, where S. ludovici is absent. A similar distribution pattern is found in two brachypterous genera of streblid bat flies, Joblingia and Anatrichobius. Joblingia includes two described species. Joblingia schmidti Dybas & Wenzel, 1947 parasitizes Myotis keaysi J. A. Allen, 1914 and M. oxyotus (Peters, 1867) from the highlands of Guatemala and Panama (Guerrero 1995). And the recently described Joblingia minuta Graciolli & Dick, 2012 parasitize Myotis ruber (E. Geoffroy, 1806), a vespertilionid bat that is restricted to southern South America (Graciolli & Dick 2012). Anatrichobius includes two described species (Graciolli 2003) and one undescribed species (Wenzel et al. 1966) from Peru. Anatrichobius scorzai Wenzel, 1966 is found on Myotis albescens Geoffroy, 1806, M. keaysi, M. nigricans (Schinz, 1820), M. oxyotus and occurs from Panama to Argentina (Guerrero 1995; Autino et al. 2009). Another species, A. passosi Graciolli, 2003 parasites Myotis albescens (E. Geoffroy, 1806), M. levis (I. Geoffroy, 1824), M. nigricans, and M. ruber and is restricted to southern Brazil (Graciolli 2003; Bertola et al. 2005). These patterns suggest a common biogeographical history among these streblid bat flies.

The lack of cospeciation in the two populations of T. petersoni and S. erythromos (Tshudi, 1844) and S. bogotensis Shamel, 1927 may be considered a failure to speciate even though some conditions may seem favorable. Such a scenario is possible if the parasites on both hosts continued to have genetic contact after having colonized the new host (Banks & Paterson 2005). Future studies using molecular data will be able to better understand this process.

Cospeciation may be uncommon in this group for a variety of reasons. Cophylogenetic associations among arthropods and their vertebrate hosts studied using component analysis may be divided into two groups with respect to the life-cycle biology of the parasites (Table I). In one group, the entire life cycle is on the host, and in the second group, part of the life cycle is not on the host (usually an immature stage). The second is that most common among these flies. Historical associations of lice and some mites with their hosts may have many cospeciation/codivergent events, exceptions to which may be explained by phoresia (Weckstein 2004). Cophylogenetic analyses of other, second-group, ectoparasitic arthropods (fleas and ticks) with preparasitic stages in the roost or environment of their mammal hosts, find a low probability of cospeciation. This suggests that sorting, duplication and host switching were more likely. In conclusion, we found that Trichobius is not monophyletic, and that during the coevolutionary history of Trichobius phyllostomae group and their hosts, cospeciation seldom occurred.

ACKNOWLEDGMENTS

We thank the following curators and their institutions for the specimens loaned: Márcia Árzua (MHNCI), Eliana Cancello and Mirian Marques (MZSP), and Rupert L. Wenzel and Carl W. Dick (FMNH). We are grateful to Luciane Marinoni, Walter A. P. Boeger, Aricio X. Linhares, Mauricio O. Moura, Marcela Lareschi, and Silvio S. Nihei for helpful comments. We are grateful to David Notton for sending digital images of the holotype of Trichobius phyllostomae. Maureen Turcatel prepared the illustration of Sturnira in Fig. 2. This study was funded by the CNPq (#140214/00-3) and National Science Foundation (DBI-0545051) -"Digitization, Conversion, and Accessibility of The Field Museum's Bat Fly Collection." Carl W. Dick, PI; Bruce D. Patterson, Joanna McCaffrey, Alfred F. Newton, CoPIs. CJBC is also grateful to the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), for the support provided (process # 304713/2011-2). Contribution number 1781 of Departamento de Zoologia, UFPR.

Received 18/4/2012; accepted 26/10/2012

Editor: Marcia Souto Couri

Appendix 1. Examined specimens.

Trichobius Gervais, 1844 phyllostomae group Trichobius phyllostomae Kessel, 1925

Type material examined. BRASIL. Santa Catarina. Humboldt [ = Corupá], VII/1916 (digital image of 1 female [11921-200 Trichobius phyllostomae Kessel TYPE], host Phyllostoma sp., BMNH), W. Erhardt leg.

Material examined. BRASIL. São Paulo. Ilha do Cardoso, 14/V/1991 (1 female, host Sturnira lilium, MZSP), S.F. Corrêa leg. 11/VI/1991 (1 female, host Sturnira lilium, MZSP), S.F. Corrêa leg. Paraná. Adrianópolis, Fazenda Primavera, Tatuapeva, 05/VI/2001 (1 male, host Sturnira lilium, MHNCI). Fazenda Rio Grande, Fazenda Gralha Azul, 22/II/2002 (1 female, host S. lilium, DZUP), G. Bianconi leg. Tijucas do Sul, Lagoa, 15/IV/2000 (1, host S. lilium, MHNCI), G. Bianconi leg. Santa Catarina. Humboldt [ = Corupá], 07/VII/1918 (digital image of wing, hosp. Phyllostoma sp., BMNH), W. Erhardt leg. Joinville, 08/V/1929 (1 male, host Artibeus lituratus, CNHM), Steiger leg. Rio Grande do Sul. Dom Pedro de Alcântara, 20/II/1997 (1 female, host S. lilium, DZUP), A. M. Rui leg. Maquiné: Estação de Pesquisa e Produção de Maquiné da Fundação de Pesquisa Agropecuária da Secretaria da Agricultura do Estado do Rio Grande do Sul (FEPAGRO), 27/IX/2000 (1 female, host S. lilium, DZUP), A. M. Rui leg; 17/XI/2000 (1 male, 2 females, host S. lilium, DZUP), A. M. Rui leg.

Trichobius brennani Wenzel, 1966

Type material examined. Paratypes. PANAMA. Chiriquí: Casa Lewis, Cerro Punta. Elev. 5600-5700 ft., 03/II/1960 (1 female [slide], host Sturnira ludovici [Tipton-Handley#5878], FMNH), C. M. Keenan & V.J. Tipton legs; (1 female, host Sturnira ludovici [Tipton-Handley#5884], FMNH), C. M. Keenan & V. J. Tipton leg.; Casa Tilley, Cerro Punta, elev. 5300-5600 ft., 23/IV/1961 (1 male [slide], host S. ludovici [Tipton-Handley#7303], C. M. Keenan & V. J. Tipton leg.; 12/III/1962 (2 females [slide] 1 female [Allo-type, slide], host S. ludovici [Tipton-Handley#10469], FMNH), C. M. Keenan & V. J. Tipton leg.; 14/III/1962 (2 males, 1 female, host S. ludovici [Tipton-Handley#10469], FMNH), C. M. Keenan & V. J. Tipton leg.; 01/V/ 1960 (1 female, host S. ludovici [Tipton-Handley#6176], FMNH), C. M. Keenan & V. J. Tipton leg. Darién: Cerro Malí, 09/II/1964 (2 males, 1 female, host S. ludovici [Tipton-Handley#12443], FMNH), C. O. Handley Jr. leg; Cerro Tacarcuna, el. 4100-4800ft., 29/II/1964 (1 male, 1 female, host S. ludovici, [Tipton-Handley#12797], MZSP), C. O. Handley Jr. leg.; 04/III/1964 (2 males, host S. ludovici [Tipton-Handley#5878], FMNH), C. O. Handley Jr.; 07/III/1964 (1 male, 3 females, host S. ludovici [TiptonHandley#12960], FMNH), C. O. Handley Jr. leg.

Trichobius vampyropis Wenzel, 1966

Type material examined. Holotype. PANAMA. Bocas del Toro: nr. Rancho Mojica, 13/IX/1961 (1 male [slide], host Vampyrops vitatus [ = Platyrrhinus vitattus] [Tipton-Handley#8115], FMNH), C. M. Keenan & V. J. Tipton leg. Paratypes. PANAMÁ. Bocas del Toro: nr. Rancho Mojica, 13/ IX/1961 (1 female [Allotype, slide], host Vampyrops vittatus [ = Platyrrhinus vittatus] [Tipton-Handley#8115], FMNH), C. M. Keenan & V. J. Tipton leg.

Material examined. VENEZUELA. Distrito Federal: Los Venados, 4km NNW Caracas, 1524m, 29/VII/1965 (1 female, host Vampyrops umbratus ( = Platyrrhinus dorsalis) [SVP#334], FMNH), A. Peterson & M. Tuttle leg.; 1400m, 02/VIII/1965 (2 males, 1 female, host Vampyrops umbratus ( = Platyrrhinus dorsalis) [SVP#495, 508], A. Peterson & M. Tuttle leg.; 1430m 02/VIII/1965 (1 male, 1 female, host Vampyrops umbratus ( = Platyrrhinus dorsalis) [SVP#499], A. Peterson & M. Tuttle leg.; 1463m, 03/VIII/1965 (3 males, 1 female, host Vampyrops vittatus [ = Platyrrhinus vittatus] [SPV#539], FMNH), A. Peterson & M. Tuttle leg. Pico Ávila, 5km NNE Caracas, near Hotel Humboldt, 2.092m, 20/VIII/1965 (1 male, host Vampyrops umbratus ( = Platyrrhinus dorsalis) [SVP#676], FMNH), A. Peterson & M. Tuttle leg.

Trichobius hispidus Wenzel, 1976

Type material examined. Paratypes. VENEZUELA. Mérida: Tabay, 6 km ESSE Tabay, Midlle Refugio, 2550 m, 06/IV/1966 (2 females [slide], host Sturnira bidens [SVP#4369, 4372], FMNH), N. E. Peterson leg.; 2640m, 08/IV/1966 (1 male, 1 female, host S. bidens [SVP#4391], FMNH), N. E. Peterson leg.; 2640m, 11/IV/1966 (1 male [slide], host Sturnira bidens [SVP#4411], FMNH), N. E. Peterson leg. 2640m,15/IV/1966 (1 male [slide], host Sturnira bidens [SVP#4447], FMNH), N. E. Peterson leg.

Trichobius petersoni Wenzel, 1976

Type material examined. Holotype. COLOMBIA. Cundimarca: Bogota, X/1968 (1 male [#46, slide], host Sturnira bogotensis [field#cjm 13139], FMNH), C. J. Marinkelle leg. Parátipos. COLOMBIA. Cundimarca: Bogota, X/1968 (1 female [Allotype #46, slide], host S. bogotensis [field#cjm 13139], FMNH), C. J. Marinkelle leg. VENEZUELA. Mérida: La Carbonera, 12 km SE La Azulita 2150m, 21/IV/1966 (2 male [1 slide] 1 female, host Sturnira erythromos [SVP#4471, 4482], FMNH), N. E. Peterson leg.; 2180m, 23/IV/1966 (1 female [slide], host S. erythromos [SVP#4536], FMNH), N. E. Peterson leg. Tabay: 4km E Tabay, La Mucuy, 2107m, 08/III/1966 (1 male [slide], host S. bogotensis [SVP#4012], N. E. Peterson leg.

Trichobius sp. 1

Material examined. COSTA RICA. San Jose: San Gerardo, 26/III/ 1985 (1 male [slide] 2 females [1 slide], host Sturnira sp., MZSP), R. Eckerlin leg. Herédia: Parque Nacional Braulio Carrillo, San Miguel, 3.5 km S, 11.5 km E, 1000 m, 09/IV/1986 (1 male, host Sturnira mordax [USNM562835], FMNH), R. L. LaVal leg. Alayuela: 4.2 km SE, Cariblanco (10º14´N 84º10'W), 12/II/1983 (1 female, host S. mordax [CM92487], FMNH), T. J. McCarthy leg.

Trichobius sp. 2

Material examined. PERU. Cuzco: Paucartambo, Pillahuata, 2460 meters, 03/IV/2001 (1 male [PMV#557], host Platyrrhinus albericoi, FMNH), P. M. Velazco leg.; 18/IV/2001 (1 male [SS#2087], host Platyrrhinus nigellus, FMNH), S. Solari leg.; 07/X/2001 (1 female [JCH#1168], host Platyrrhinus infuscus, FMNH), S. Solari leg.; 13/X/2001 (2 males, 2 females [BDP#4092], host P. nigellus, FMNH), B. D. Patterson leg.; 08/XI/2001 (2 females [CWD#P768], host Platyrrhinus masu, FMNH), E. Suazo leg.

pallidus group Trichobius pallidus (Curran, 1934)

Material examined. BRASIL. São Paulo. Capão Redondo, Parque Estadual de Intervale, sede de pesquisa, 14/XII/1999 (1 male, host Furipterus horrens, DZUP), F. C. Passos & W. A. Pedro leg.; 17/XII/1999 (1 male, 1 female, host F. horrens, DZUP), F. C. Passos & W. A. Pedro leg.

caecus group Trichobius galei Wenzel, 1966

Material examined. BRASIL. Mato Grosso do Sul. Corumbá, Instituto Agronômico, 29/I/2003 (4 males, 2 females, host Natalus stramineus, DZUP), M. O. Bordignon leg.

uniformis group Trichobius lonchophyllae Wenzel, 1966

Material examined. BRASIL. Distrito Federal. Brasília, Gruta Fenda II, 11/VI/1998 (1 male, 1 female, host Lonchophylla deykeseri, DZUP), D. C. Coelho leg. Brazlândia, Gruta da Cafurinca, 02/V/1998 (10 males, host L. deykeseri, DZUP), D. C. Coelho leg.; 27/V/1998 (2 males, host L. deykeseri, DZUP), D. C. Coelho leg.

dugesii group dugesii complex Trichobius tiptoni Wenzel, 1976

Material examined. BRAZIL. Rio Grande do Sul. Dom Pedro de Alcântara, 24/IV/1997 (4 males, 1 female, host Anoura caudifer, DZUP), Graciolli, Rui & Grillo leg.; 6-7/IX/1997 (2 males, host Anoura geoffroyi, DZUP), H. C. Z. Grillo leg.; 08/IX/1997 (4 males, 1 female, host A. caudifer, DZUP), H. C. Z. Grillo leg.

parasiticus complex Trichobius parasiticus Gervais, 1844

Material examined. ARGENTINA. Tucumán. Arroyo Aguas Chiquitas, Reserva Provincial Aguas Chiquitas, 31/X/1993 (1 male, 1 female, host Desmodus rotundus, DZUP), A. G. Autino leg. BRAZIL. Mato Grosso do Sul. Fazenda Rio Negro, (1 female, D. rotundus, MZSP).

longipes group Trichobius longipes (Rudow, 1871)

Material examined. BRAZIL. Distrito Federal. Brazlândia, Gruta Dois Irmãos, 02/V/1998 (2 females, host Phyllostomus hastatus, DZUP), D. C. Coelho leg. Paraná. Fênix, Parque Estadual Vila Rica do Espírito Santo, 22/VI/2001 (1 male, host P. hastatus, DZUP), G. Bianconi leg.

dunni group Trichobius jubatus Wenzel, 1976

Material examined. BRAZIL. Paraná. Diamante do Norte, 28/IX/1994 (1 male, 2 females, host Molossus rufus, MHNCI).

major group Trichobius major Coquillett, 1899

Material examined. USA. Florida: Alachua Co., bat cave near Newberry, 04/VIII/1953 (2 males, 2 females, host Myotis austroriparius, MZSP), Dale W. Rice leg.

Paratrichobius Costa-Lima, 1921 Paratrichobius longicrus (Miranda-Ribeiro, 1907)

Material examined. BRAZIL. Paraná. Londrina, Fundo de Vale Quebec, 13/XI/1998 (2 males, 2 females, host Artibeus lituratus, DZUP), G. Graciolli & I. Lima legs. Rio Grande do Sul. Dom Pedro de Alcântara, 1997 (1 female, host A. lituratus, DZUP), G. Graciolli & A. M. Rui legs.; 10/V/1997 (11 males, 3 females, host A. lituratus, DZUP), G. Graciolli & A.M. Rui legs.

Megistopoda Macquart, 1852 Megistopoda aranea (Coquillett, 1899)

Material examined. BRAZIL. Pará. Santarém, Alter de Chão, 07/V/ 2000 (1 male, 1 female, host Artibeus planirostris, DZUP), E. Bernard leg. Paraná. Foz do Iguaçu, Parque Nacional do Iguaçu, 12/I/1999 (1 male, host Artibeus fimbriatus, DZUP), Graciolli, Lima & Rocha legs. Rio Grande do Sul. Dom Pedro de Alcântara, 1997 (1 female, host Artibeus lituratus, DZUP), G. Graciolli & A. M. Rui legs.

Megistapophysis Wenzel & Dick, 2006 Megistapophysis mordax Wenzel & Dick, 2006

Material examined. COSTA RICA. Puntarenas. Monteverde, 1400m (2 males, host Sturnira mordax, DZUP), R. K. Laval leg. San Jose. Cerro de La Muerte 10.5 mi N San Isidro General, 1585m, 15/VIII/1963 (2 females, host Sturnira mordax, FMNH), R. S. Casebeer leg.

Neotrichobius Wenzel & Aitken, 1966 Neotrichobius delicatus (Machado-Allison, 1965)

Material examined. BRAZIL. São Paulo. Ilha do Cardoso, 14/III/1991 (1 male, 2 females, host Artibeus cinereus, MZSP), S. F. Corrêa leg.

Speiseria Kessel, 1925 Speiseria ambigua Kessel, 1925

Material examined. BRAZIL. Amazônia. Reserva 1501 km 41, PDBFF Amazonas, 04/VIII/2003 (1 male, 1 female, host Carollia sp., INPA), F. Rocha-Mendes leg. Gavião, Dynamical Biological Forest Fragments Project, 20/X/1997 (1 male, 1 female, host Carollia perspicillata, MZSP), E. Sampaio leg.

Strebla Wiedemann, 1824 Strebla wiedemanni Kolenati, 1856

Material examined. BRAZIL. Paraná. Almirante Tamandaré, 30/IV/ 1985 (1 male, 1 female, host Desmodus rotundus, DZUP), Viana leg. Rio Branco do Sul, 21/VI/1988 (5 males, host D. rotundus, DZUP), Dos Santos leg. São Jerônimo da Serra, 20/VII/1987 (3 males, 4 females, host D. rotundus, DZUP), Dos Santos leg.

• Autino, A. G.; G. L. Claps; M. S. Sánchez & R. M. Barquez. 2009. New records of bat ectoparasites (Diptera, Hemiptera and Siphonaptera) from northern Argentina. Neotropical Entomology 38:165-177
• Banks, J. C.; R. L. Palma & A. M. Paterson. 2006. Cophylogenetic relationships between penguins and their chewing lice. Journal of Evolutionary Biology 19:156-166.
• Banks, J. C. & A. M. Paterson. 2005. Multi-host parasite species in cophylogenetic studies. International Journal for Parasitology 35:741-746.
• Bertola, P. B.; C. C. Aires; S. E. Favorito; G. Graciolli; M. Amaku & R. Pinto-da-Rocha. 2005. Bat flies (Diptera: Streblidae, Nycteribiidae) parasitic on bats (Mammalia: Chiroptera) at Parque Estadual da Cantareira, São Paulo, Brazil: parasitism rates and host-parasite associations. Memórias do Instituto Oswaldo Cruz 100:25-32.
• Bochov, A. V.; P. B. Klimov & G. Wauthy. 2011. Phylogeny and coevolutionary associations of makialgine mites (Acari, Psoroptidae, Makialginae) provide insight into evolutionary history of their hosts, strepsirrhine primates. Zoological Journal of the Linnean Society162:1-14.
• Bordignon, M. O. 2006. Diet of the fishing bat Noctilio leporinus (Linnaeus) (Mammalia, Chiroptera) in a mangrove area of southern Brazil. Revista Brasileira de Zoologia 23:256-260.
• Clayton, D. H.; S. Al-Tamimi & K. P. Jonhson. 2003. The ecological basis of coevolutionary history, p. 310-341. In: R. D. M. Page (ed.). Tangled Trees: Phylogeny, Cospeciation and Coevolution. Chicago, The University of Chicago Press, 350 p.
• Cumming, G. S. 2000. Host use does not clarify the evolutionary history of African ticks (Acari: Ixodoidea). African Zoology 35:43-50.
• Dabert, J.; M. Dabert & S. V. Mironov. 2001. Phylogeny of feather mite Subfamily Avenzoariinae (Acari: Analgoidea: Avenzoariidae) inferred from combined analyses of molecular and morphological data. Molecular Phylogenetics and Evolution 20:124-135.
• Dick, C. W. & J. Miller. 2010. Streblidae (bat flies), p. 1249-1260. In: B. V. Brown; A. Borkent; J. M. Cumming; D. M. Wood, N. E. Woodley & M. A. Zumbado (eds.). Manual of Central American Diptera, Volume 2. Ottawa, NRC Research Press, xi+1442 p.
• Dick, C. W. & R. L. Wenzel. 2006. A new genus and species of trichobiine bat flies (Diptera: Hippoboscoidea: Streblidae) from Costa Rica. Insect Systematics and Evolution 37:433-442.
• Dittmar, K.; M. L. Porter; S. Murray & M. F. Whiting. 2006. Molecular phylogenetic analysis of nycteribiid and streblid bat flies (Diptera: Brachycera, Calyptratae): implications for host associations and phylogeographic origins. Molecular Phylogenetics and Evolution 38:155-170.
• Gardner, A. L. 2007. Genus Platyrrhinus, p. 329-341. In: A. L. Gardner (ed.). Mammals of South America, Volume 1, marsupials, xenarthrans, shrews, and bats. Chicago, The University Chicago Press, xiv+669 p.
• Gannon, M. R.; M. R. Willig & J. Knox Jones Jr. 1989. Sturnira lilium Mammalian species 333:1-5.
• Giannini, N. P. 1999. Selection of diet and elevation by sympatric species of Sturnira in an Andean rainforest. Journal of Mammalogy 80:1186- 1195.
• Goloboff, P. 1998. Nona, version 2.0 Computer program and documentation distributed by the author,Tucuman, Argentina. http://www.cladistics.com
• Graciolli, G. 2003. Nova espécie de Anatrichobius Wenzel, 1966 (Diptera, Streblidae) do Brasil meridional. Revista Brasileira de Entomologia 47:55-58.
• Graciolli, G. 2010. Nycteribiidae(bat flies, spider bat flies), p. 1261-1266. In: B. V. Brown; A. Borkent; J. M. Cumming; D. M. Wood, N. E. Woodley & M. A. Zumbado (eds.). Manual of Central American Diptera, Volume 2. Ottawa, NRC Research Press, xi+1442 p.
• Graciolli, G. & C. W. Dick. 2012. Description of a second species of Joblingia Dybas & Wenzel, 1947 (Diptera: Streblidae). Systematic Parasitology 81:187-193.
• Guerrero, R. 1995. Catálogo de los Streblidae (Diptera: Pupipara) parásitos de murciélagos (Mammalia: Chiroptera) del Nuevo Mundo. V. Trichobiinae con alas reducidas o ausentes y miscalaneos. Boletín de Entomología Venezolana, Nueva Serie 10:135-160.
• Guerrero, R. 1997. Catálogo de los Streblidae (Diptera: Pupipara) parásitos de murciélagos (Mammalia: Chiroptera) del Nuevo Mundo. VII. Lista de especies, hospedadores y países. Acta Biologica Venezuelica 17:9-24.
• Hughes, J.; M. Kennedy; K. P. Johnson; R. L. Palma & R. D. M. Page. 2007. Multiple cophylogenetic analyses reveal frequent cospeciation between pelecaniform birds and Pectinopygus lice. Systematic Biology 56:232-251.
• Jobling, B. 1938. A revision of the species of the genus Trichobius (Diptera Acalypterae, Streblidae). Parasitology 30:358-387.
• Johnson, K. P.; R. J. Adams; R. D. M. Page & D. H. Clayton. 2003.When do parasites fail to speciate in response to host speciation? Systematic Biology 52:37-47.
• Krasnov, B. R. & G. I. Shenbrot. 2002. Coevolution events in the history of association between jerboas (Rodentia: Dipopidae) and their flea parasites. Israel Journal of Zoology 48:331-350.
• Liang, L. & W. Houyoung. 2005. Morphological phylogeny of Geusibia Jordan, 1932 (Siphonaptera: Leptopsyllidae) and the host-parasite relationship with pikas. Systematic Parasitology 61:65-78.
• Lidner, A. & W. Morawetz. 2006. Seed dispersal by frugivorous bats on landslides in a montane rain forest in Southern Ecuador. Chiroptera Neotropical 12:232-237.
• Marshall, A. G. 1981. The Ecology of ectoparasitic insects. London, Academic Press, 459 p.
• Merke, D. & M. Middenfort. 2005. Reconstruction of the cophylogenetic history of related phylogenetic trees with divergence timing information. Theory in Biosciences 123:277-299.
• Molinari, J. & P. J. Soriano. 1987. Sturnira bidens Mammalian species 276:1-4.
• Morelli, M. & G. S. Spicer. 2007. Cospeciation between the nasal mite Ptilonyssus sairae (Acari: Rhinonyssidae) and its bird hosts. Systematic & Applied Acarology 12:179-188.
• Moura, M. O.; M. O. Bordignon & G. Graciolli. 2003. Host characteristics do not affect community structure of ectoparasites on the fishing bat Noctilio leporinus (L., 1758) (Mammalia: Chiroptera). Memórias do Instituto Oswaldo Cruz 98:811-815.
• Nixon, K. C. 2002. Winclada version 1.00.08 Computer program and documentation distribuded by the author. http://www.cladistics.com
• Novak, R. M.1994. Walker's Bats of the World. Baltimore, Johns Hopkins University Press, 287 p.
• Pacheco, V. & B. D. Patterson. 1992. Systematics and biogeographic analyses of four species of Sturnira (Chiroptera: Phyllostomidae), with emphasis on Peruvian forms. Memorias del Museo de Historia Natural 21:57-81.
• Page, R. D. M. 2001. NDE - Nexus Data Editor 0.5.0 http:// taxonomy.zoology.gla.ac.uk/rod/NDE/nde.html
• Page, R. D. M. (ed.). 2003. Tangled Trees. Phylogeny, Cospeciation, and Coevolution. Chicago, The University of Chicago Press, 350 p.
• Paterson, A. M.; G. P. Wallis; L. J. Wallis & R. D. Gray. 2000. Seabird and louse coevolution: complex histories revealed by 12S rRNA sequences and reconciliation analyses. Systematic Biology 49:383-399.
• Peterson, B. V. & A. Ross. 1972. A new species of Paratrichobius (Diptera: Streblidae) from Arizona. The Canadian Entomologist 104:781-784.
• Simmons, N. S. 2005. Order Chiroptera, p. 312-529. In: D. E. Wilson & D. M. Reeder (eds). Mammal species of the World: a taxonomic and geographic reference, Third Edition, Volume 1, Baltimore, Johns Hopkins University Press, 2142 p.
• Smith, V. S.; J. E. Light & L. A. Durden. 2008. Rodent louse diversity, phylogeny and cospeciation in the Manu Biosphere Reserve, Peru. Biological Journal of the Linnean Society 95:598-610.
• Soriano, P. J. & J. Molinari. 1987. Sturnira aratathomasi Mammalian species 284:1-4.
• Tamsitt, J. R. & C. Häuser. 1985. Sturnira magna Mammalian species 240:1-4.
• Velazco, P. M. 2005. Morphological phylogeny of the bat genus Platyrrhinus Saussure, 1860 (Chiroptera: Phyllostomidae) with the description of four new species. Fieldiana Zoology, New Series 105:1-53.
• Villalobos, F. & A. A. Valerio. 2002. The phylogenetic relationships of the bat genus Sturnira Gray, 1842 (Chiroptera, Phyllostomidae). Mammalian Biology 67:268-275.
• Weckstein, J. D. 2004. Biogeography explains cophylogenetic patterns in toucan chewing lice. Systematic Biology 53:154-164.
• Wenzel, R. L. 1976. The streblid batflies of Venezuela (Diptera: Streblidae). Brigham Young University Science Bulletin, Biological Series 20:1-177.
• Wenzel, R. L.; V. J. Tipton & A. Kiewlicz. 1966. The streblid batflies of Panama (Diptera: Calyptera: Streblidae), p. 405-675. In: R. L. Wenzel & V. J. Tipton (eds.). Ectoparasites of Panama. Chicago, Field Museum Natural History, xii+861 p.
• Wetterer, A. L.; M. V. Rockman & N. B. Simmons. 2000. Phylogeny of phyllostomid bats (Mammalia: Chiroptera): data from diverse morphological systems, sex chromosomes, and restriction sites. Bulletin of the American Museum of Natural History 248:1-200.
• Zeve, V. H. & D. E. Howell. 1963. The comparative external morphology of Trichobius corynorhini, T. major, and T. sphaeronotus (Diptera, Streblidae). Part II. The thorax. Annals of the Entomological Society of America 56:2-17.

Publication Dates

• Publication in this collection
  24 Jan 2013
• Date of issue
  Dec 2012

History

• Received
  18 Apr 2012
• Accepted
  26 Oct 2012