Bat flies (Nycteribiidae and Streblidae) have been used to study co-evolutionary patterns between ectoparasites and bats. In the world, Nycteribiidae and Streblidae are represented by approximately 276 and 237 species, respectively. In regions such as the Orinoquia located in the north of South America (Colombia and Venezuela), the richness of bats is high (more than 100 documented species), but studies on Nycteribiidae and Streblidae are scarce and discontinuous. To contribute to the knowledge of ectoparasitic flies in the Orinoquia, records of flies and their interactions with bats were reviewed, including new records and associations using interaction networks. We documented 124 species of Streblidae and only 12 of Nycteribiidae for the Orinoquia in approximately 102 bat species reported in Colombia and Venezuela. New records for six species of bat flies in Colombia were found (Mastoptera guimaraesi, Noctiliostrebla maai, Paradyschiria parvuloides, Trichobius jubatus, Trichobius parasiticus, and Basilia ferrisi) associated with six species of bats (Cynomops planirostris, Desmodus rotundus, Myotis handleyi, Molossus rufus, Noctilio albiventris, and Phyllostomus hastatus). The bat-ectoparasite interaction networks in the Orinoquia revealed a pattern of antagonistic relationships, with high specialization, modularity, and low connectivity and nesting. The identified networks are between bat fly species belonging to different ecomorphological groups with unique host species. This supports the idea of ecological niche partitioning among ectoparasitic bat flies and hosts. Our study expanded the knowledge of the distribution of some fly species and the associations with bat hosts in Colombia, by presenting morphological descriptions and new observations, which are key to understanding the ecology, diversity, and distribution of these species.

Bat flies, ecological networks, ecomorphology, Neotropics, Nycteribiidae, parasitism, Streblidae

Bat flies (Diptera: Hippoboscoidea) are obligate blood-feeding ectoparasites of bats (Chiroptera) and an interesting evolutionary system for studying co-evolutionary patterns between hosts and ectoparasites (Porter et al. 2021). Taxonomically bat flies are divided into two cosmopolitan families, Nycteribiidae and Streblidae (Wenzel et al. 1966; Dick and Patterson 2006; Patterson et al. 2007), but the latter is considered paraphyletic with New World and Old-World lineages (Dittmar et al. 2006; Petersen et al. 2007). Both families of bat flies show greater diversity in tropical latitudes and are less diverse in subtropical and temperate regions (Dick and Dittmar 2014). However, Nycteribiids (three subfamilies, 11 genera and ~ 276 species for 2018) are more numerous in the Eastern Hemisphere, while, for the Streblids (five subfamilies, 33 genera, and ~ 237 species for 2018) the richness is greater in the Western Hemisphere (Dick and Patterson 2006; Dick and Miller 2010; Soares et al. 2013; Graciolli and Dick 2018). Based on dispersion-vicariance analysis, Dittmar et al. (2006) suggested that the Neotropical region is the ancestral area for all New World Streblidae, while the Oriental region is considered the ancestral area of Nycteribidae and Old-World Streblidae.

Both Nycteribiidae and Streblidae present a wide variety of morphological and physiological adaptations for their ectoparasitic lifestyle, among which adenotrophic viviparity stands out (Hagan 1951; Lehane 2005; Dick and Miller 2010; Dick and Dittmar 2014). In this process, the larvae develop individually in the female oviduct until the third instar (called prepupa) (Hagan 1951). Later, when the development of the prepupa is complete, female bat flies of both families deposit a single prepupa on substrates in the roost (Petersen et al. 2007; Dick and Dittmar 2014). Once the prepupa is deposited, it immediately forms a puparium, which, after a pupal stage, emerges as an unfed adult fly (teneral) and must locate and colonize a host (Ching and Marshall 1968).

In general, the life strategy of bat flies reflects their association with bats, and it has been found that some Streblidae flies can form specific parasite assemblages for each bat species due to their high host specificity (Wenzel 1976; Hiller et al. 2018). Each community can consist of two to five fly species, and each of them shows a preference for a specific part of the bat’s body (Wenzel 1976; Ter Hofstede et al. 2004; Dick and Gettinger 2005; Tello et al. 2008; Patterson et al. 2009). Dick (2005) defined three ecomorphological groups of bat flies based on behavioral observations and morphological traits: 1) “wing crawler” primarily live on wing membranes, with non-flattened bodies and legs of the same size, 2) “fur runners” that live on the hairy membranes of the body and move by running on the fur, with well-developed wings and long legs (especially the hind legs), and 3) “fur swimmers” that inhabit areas of long fur such as the neck and have compressed bodies and heads, and possessing Ctenidia. Hiller et al. (2018) supported this classification, focusing on differences in the morphology and size of the hind legs, and found evidence of density-dependent competition among species within the same ecomorphological group.

In Neotropical countries with a high number of bat species such as Colombia (~ 217 species) and Venezuela (172 species) (Delgado-Jaramillo et al. 2016; Ramírez-Chaves et al. 2021), studies on Streblidae and Nycteribiidae are few and discontinuous (e.g., Tamsitt and Fox 1970; Marinkelle and Grose 1981; Guerrero 1994). In recent years, new research has expanded the knowledge about these ectoparasites in both countries (Herrera-Sepúlveda 2013; Tarquino-Carbonell 2014; Tarquino-Carbonell et al. 2015; Dick et al. 2016; Durán et al. 2018; Calonge-Camargo and Pérez-Torres 2018; Guerrero 2019; Liévano-Romero et al. 2019; Raigosa Álvarez et al. 2020; Cañizales and Guerrero 2022). Nonetheless, there are very few studies in the Orinoquia Region, a hydrographic basin that converges into the Orinoco River, covering an area of almost 989,000 km² (Domínguez 1998), and shared by both countries (Guimarães 1972; Wenzel 1976; Dick et al. 2016; Liévano-Romero et al. 2019; López Rivera et al. 2022). Approximately 65% of this area is located in Venezuela, while the remaining 35% is in Colombia (León 2005). The Venezuelan portion extends from the Andes and the Cordillera de la Costa to the north-western bank of the Orinoco River, forming most of the Venezuelan plains and the delta Orinoco (Domínguez 1998; León 2005). The Orinoquia is one of the most diverse areas in terms of mammal diversity with 318 species, including 150 species of Chiroptera (Ferrer Pérez et al. 2009; Pardo and Rangel-Ch. 2014).

Of the 125 Streblidae species reported in Colombia and Venezuela (81 and 107 respectively; Wenzel 1976; Tarquino-Carbonell et al. 2015; Dick et al. 2016; Duran et al. 2017; Calonge-Camargo and Pérez-Torres 2018; Guerrero 2019; Liévano-Romero et al. 2019; Ascuntar-Osnas et al. 2020; Raigosa Álvarez et al. 2020; López Rivera et al. 2022), 124 have been documented for the Orinoquia in ~ 96 bat species of the families Emballonuridae, Molossidae, Mormoopidae, Natalidae, Noctilionidae, Phyllostomidae, and Vespertilionidae (Wenzel 1976; Dick et al. 2016; Guerrero 2019; Liévano-Romero et al. 2019; López Rivera et al. 2022). Similarly, of the 16 Nycteribiidae reported in Colombia and Venezuela (11 and 13, respectively; Wenzel and Tipton 1966; Graciolli et al. 2016; Pastrana-Montiel et al. 2019; Raigosa Álvarez et al. 2020; López Rivera et al. 2022), 12 have been found in the Orinoquia (two in Colombia, 12 in Venezuela; Wenzel and Tipton 1966; Graciolli et al. 2016; López Rivera et al. 2022). Besides the high number of bats and bat flies documented in the Orinoquia, the interaction structure that might be evaluated using ecological or interaction networks (Dormann et al. 2009), has not been addressed. Interaction networks are formed by interactions (links) between species (nodes) that make up a community (network); they are considered as a synthesis tool in the study of ecological interactions that allow understanding the functioning of megadiverse systems (Blüthgen et al. 2006). The study of the host-parasite specificity interactions is essential to understand the mechanisms behind parasitism and its relationship with biodiversity functioning (Frainer et al. 2018) since parasites play an important role in the regulation of populations of host species (Poulin et al. 2006). For those reasons, and to contribute to the knowledge of ectoparasitic flies of the Orinoquia, we present novel records (including morphological description of ectoparasitic flies) and new associations with bats from this region in Colombia, and a review of records of ectoparasitic flies and the interactions with hosts bats in the whole Orinoquia (Colombia-Venezuela).

Our study highlights the high diversity of bat flies that inhabit the Orinoco Region in northern South America. In other neotropical countries such as Brazil, Paraguay, and Panama, several studies have found a directly proportional relationship between the richness of Streblidae flies and bats (Dick and Gettinger 2005; Graciolli and Bianconi 2007). Considering that Colombia and Venezuela, are among the countries with the greatest diversity of bats in the Neotropics (with 217 and 172 species, respectively; Delgado-Jaramillo et al. 2016; Ramírez-Chaves et al. 2021), the finding of 136 species of bat flies in the Orinoquia Region was expected and will increase with additional studies in poorly sampled areas the Andean foothills of the Orinoco basin were endemic bat species such as Vampyressa voragine inhabit (Morales-Martínez et al. 2021). Besides the elevated number of the species documented in the Orinoquia, there are still gaps of information about these flies confirming the need for greater sampling efforts, particularly in some states/departments and biomes such as Portuguesa, Cojedes and Delta Amaruco in Venezuela and Guaviare in Colombia. The richness of bat flies in Colombia and Venezuela is also higher and surpasses the number of species documented in countries like Brazil, which despite having the largest area in South America (48%), has 181 species of bats (Garbino et al. 2022) and > 100 species of Streblidae and 26 of Nycteribiidae (da Silva et al. 2023).

Although the high specialization of some species of Streblidae has been controversial and some researchers have previously described them as mostly host-nonspecific (Dittmar et al. 2006), the evaluation of the quality of the non-primary associations reported in the literature for Orinoquia showed that several of them (~ 10% of the records) were probably due to accidental transfers or contamination of the samples during collection. These results support more recent studies suggesting that most species are host-specific (de Vasconcelos et al. 2016; Estrada-Villegas et al. 2018), a change attributable to methodological improvements in sample collection and taxonomic updates of flies and hosts. In addition, we have found controversial non-primary associations that could not be classified as possible cases of contamination. Some of these associations occurred in congeneric species, as is the case of the reports of M. minuta (primary host: L. silvicola) on T. maresi in Casanare, Colombia (Liévano-Romero et al. 2019). Considering that some unusual interaction occurred in non-congeneric species of the primary hosts, explaining the reasons behind is complicated. This is the case of N. albiventris parasitized by M. proxima (primary host: Sturnira species), and S. giannae parasitized by N. maai (primary host: N. albiventris) (López Rivera et al. 2022). Furthermore, the nature of the specificity in some non-primary associations for certain fly species complexes could not be determined (Table 1). In this sense, we suggested that the determination of the true specificity of ectoparasitic flies will only be achieved when: 1) large samples of hosts and parasites are available, 2) sampling protocols strictly control cross-contamination between hosts, and 3) association cases are evaluated statistically (Dick 2005).

Although most studies on the interaction between bats and ectoparasites have been descriptive, these are crucial for understanding host-parasite dynamics in different environments (Patterson et al. 2007; Fagundes et al. 2017; Salinas-Ramos et al. 2018). Previous studies focused on interactions networks only covered few localities in the Colombian Orinoquia (López Rivera et al. 2022) or were limited to enumerating species and abundances in specific locations in Colombia (departments of Arauca, Casanare, and Meta) and Venezuela (states of Anzoátegui, Apure, Aragua, Barinas, Carabobo, Guárico, Lara, Mérida, Monagas, Trujillo and Yaracuy) (Wenzel and Tipton 1966; Wenzel 1976; Marinkelle and Grose 1981; Dick et al. 2016; Liévano-Romero et al. 2019; López Rivera et al. 2022). Considering that, our study represents the first attempt to define the ecological interactions between ectoparasitic flies and their host bats in the Orinoquia Region of Colombia and Venezuela. However, previously reported modularity and specialization values for the interaction networks in the Department of Arauca, in Colombia (Q = 0.61 and 0.69; H₂’=0.78 and 0.91) (López Rivera et al. 2022), were similar to the values found for Nycteribiidae in the network of the entire Orinoquia (Q = 0.67 and H₂’=0.88), but lower compared to the values found in the Streblidae network of the entire Orinoquia (Q = 0.85 and H₂’=0.94). These findings match suggestions in the literature where the size of the network can influence the modularity values, being higher for larger networks (Júnior et al. 2020). This can be explained by the highly specialized nature of the interaction between bats and bat flies (Dick and Patterson 2007), where most fly species are associated with a single host (monoxenic), and even species that parasitize multiple bat species (oligoxenic and polyxenic) are restricted to phylogenetically close hosts, leading to the formation of groups with a similar composition of ectoparasitic fly species (Falcão 2015; López Rivera et al. 2022).

Ectoparasites typically display some host specificity, implying that some ectoparasitic fly species are adapted to parasitize only one or a few bat species (Wenzel 1976; Dick and Patterson 2007). The 11 bat species that were parasitized by more than five bat fly species might have characteristics or behaviors that make them prone to being parasitized by multiple bat fly species. Several studies have shown the common presence of various bat fly species on the same bat (Wenzel 1976; Dick 2005; Hiller et al. 2018). According to Wenzel (1976), 63% of the infested Venezuelan bat species harbored two to four species of bat flies. Several species of Streblidae coexist on the same hosts, most of these associations are negatively correlated in abundance, but competition is not strong enough to lead to local extinction (Dick 2005). In fact, some of the highly connected bat host species within the interaction network for the Orinoquia were reported to be involved in some of the previously documented cases of positive co-occurrence between bat flies (Dick 2005), such as: E. clovisi and Trichobius propinquus on Anoura geoffroyi (Fig. 7), Trichobius joblingi and Strebla guajiro on C. perspicillata (Fig. 8), and Trichobius lionycteridis with A. spurrelli on L. spurrelli (Fig. 9).

In a lesser extent, positive correlations have also been reported in abundance, indicating mutualistic relationships (Dick 2005; Hiller et al. 2018; Alcantara et al. 2022). It is possible that the presence and greater abundance of one species of parasite facilitates the presence and abundance of the other species, reciprocally eliminating the pressure of grooming (one of the main causes of mortality of ectoparasites) in each population (Marshall 1981). Our interaction networks for the Orinoquia also revealed positive co-occurrences previously reported for ectoparasitic flies from Venezuela (Dick 2005). For example, a relatively equivalent association in the size of links and nodes is observed between T. parasiticus and Strebla wiedemanni, both associated with D. rotundus (Fig. 8). Similarly, an equivalent relationship between S. diaemi and T. diaemi is presented in its primary host D. youngii (Fig. 9). Cooccurrences of Neotropical Streblids have been reported mainly between species belonging to different genera (Dick 2005; Hiller et al. 2018) that differ in their general morphology facilitating the coexistence of species by utilizing different regions on the host’s body surface (Wenzel 1976; ter Hofstede et al. 2004; Dick 2005; Hiller et al. 2018). Our interaction networks reveal the separation of the three ecomorphological groups (wing crawler – WC, fur runner – FR, and fur swimmer – FS), associated in a single bat species (Dick 2005; Hiller et al. 2018). Some notable examples of these associations include: i) Metalasmus pseudopterus (FS) with A. phyllostomatis (WC) or with M. aranea (FR), and Aspidoptera phyllostomatis (WC) with Megistopoda aranea (FR) in Artibeus planirostris; ii) the association of Speiseria ambigua (FR) with S. guajiro (FW) or with Trichobius joblingi (WC) on Carollia perspicillata, and the association of Strebla guajiro (FS) with Trichobius joblingi (WC) on the same host; iii) the presence of T. parasiticus (WC) and Strebla wiedemanni (FS) on D. rotundus, and iv) the association of Strebla hertigi (FS) with Trichobius costalimai (WC) or with Trichobioides perspicillatus (WC) on Phyllostomus discolor (Fig. 8). Therefore, our results suggest ecological niche partitioning of ectoparasitic flies on bat hosts in the Orinoco region. However, due to the lack of a detailed phylogeny of bat flies, it is unclear whether these morphological differences reflect the evolutionary history of bat flies or represent convergent adaptations to host habitat type (Hiller et al. 2018).

In this study, new records are presented that describe the co-occurrence of N. maai and P. parvuloides in N. albiventris in the Department of Arauca, Colombia, an association previously reported by López Rivera et al. (2022) (Table 2). The positive interactions between Noctiliostrebla and Paradyschria species in Noctilio spp. are well known and documented (Wenzel 1976; Guerrero 1995; Guerrero 1998; Moura et al. 2003; Schad et al. 2012). For example, in the case of N. albiventris, N. maai coexists with Paradyschiria curvata or Paradyschiria parvula, and in the case of Noctilio leporinus, each individual host is infested with Noctiliostrebla aitkeni and Paradyschiria fusca or Noctiliostrebla traubi and Paradyschiria lineata (Wenzel 1976; Dick 2005). In this sense, Paradyschiria almost always presents the highest values of prevalence and abundance, while Noctiliostrebla occurs more commonly in the presence of Paradyschiria (Moura et al. 2003; Presley 2007; Schad et al. 2012; Guerrero 2019). Our results agree with this pattern since we observed a higher abundance of P. parvuloides compared to N. maai in N. albiventris (Fig. 8, Table 2).

Paradyschiria species are fur runners, while Noctiliostrebla spp. are wing-crawlers, again supporting the niche partitioning of these species on Noctilio spp. The ecomorphological classification for Paradyschiria has been controversial, as they had previously been classified as wing-crawlers (Dick 2005) for not having longer hind legs, a morphological feature associated with the fur runner microhabitat (Ter Hofstede et al. 2004). However, Paradyschiria spp. Have an extremely strong forefemur compared to other streblid species, which allows them greater agility to move in the fur and and hold on the host’s hair (Wenzel et al. 1966; Wenzel 1976; Alcantara et al. 2022). Noctilio bats are considered to be among the shortest-haired of Neotropical and Nearctic bats (Dick and Miller 2010), so Paradyschiria flies may not need longer hind legs, otherwise a strong forefemur can be used to hold on to the host’s hair tightly if it cannot avoid grooming with rapid movements (Alcantara et al. 2022).

Finally, the analysis of ecological networks has been fundamental in the understanding of complex biological systems, providing information on how species are organized and connected in a community (Bascompte 2009; Butts 2009; Thébault and Fontaine 2010). In this sense, the study of host-parasite specificity interactions is essential to understand the mechanisms behind parasitism and its relationship with the functioning of biodiversity (Frainer et al. 2018), since parasites play an important role in the regulation of populations of host species (Poulin et al. 2006). Our results indicated that highly connected bat species in the Orinoquia interaction network act as hosts for a diversity of ectoparasitic fly species belonging to different ecomorphological groups. These groups differ in how they interact and are located on the bat’s body. Therefore, the presence of these bats is not only important for ectoparasitic fly species in general, but also for maintaining the coexistence and interaction of different ecomorphological groups of ectoparasitic flies in the network.

The Orinoco Region, located in northern South America, harbors a remarkable diversity of bat flies. This diversity is largely attributed to the rich bat fauna found in Colombia and Venezuela, two countries known for their high bat diversity in the Neotropical region. In the Orinoco Region, associations have been identified between bat fly species belonging to different ecomorphological groups and unique host species. This supports the idea of a potential niche partitioning among ectoparasitic bat flies on their bat hosts. However, due to the lack of a detailed phylogeny of bat flies, it is unclear whether the observed morphological differences are the result of evolutionary history or convergent adaptations to different host habitat types. In this study, we present new records of associations between bat flies and their bat hosts, thereby expanding our knowledge of these interactions in the Orinoco Region. Overall, this study contributes to our understanding of the diversity, specificity, and ecological interactions between bat flies and their host bats in this region. These findings underscore the need for further research and sampling efforts to fill knowledge gaps in this field.

We thank the Museo de Historia Natural de la Universidad de Caldas, for allowing the work in the facilities of the MHN-UCa collections. This project was funded by the Vicerrectoría de Investigaciones, Universidad de Caldas (code 0318322 and 0277620). We thank the Convocatoria de subvención del Fondo de Ciencia, Tecnología e Innovación del Sistema General de Regalías para componer una lista de proyectos elegibles para su viabilidad, priorización y aprobación por parte de la OCAD como parte del Programa de Becas de Excelencia Doctoral cohorte 1 – 2019.. We thank the Ecology Laboratory of the Universidad de Caldas for lending us their space and stereomicroscope with camera. We thank Liping Yan, Gustavo Graciolli, and two anonymous reviewers for comments on an early version of this manuscript and for the editorial support.