Townsend, Victor R. Jr. ¹ ; Schaus, Maynard H. ² ; Roberto, Elizabeth A.³ and Proud, Daniel N. ⁴

¹✉ Department of Biology, Virginia Wesleyan University, 5817 Wesleyan Drive, Virginia Beach, Virginia 23455 USA.
²Department of Biology, Virginia Wesleyan University, 5817 Wesleyan Drive, Virginia Beach, Virginia 23455 USA.
³Department of Biology, Virginia Wesleyan University, 5817 Wesleyan Drive, Virginia Beach, Virginia 23455 USA.
⁴Department of Biological Sciences, Moravian University, 1200 Main Street, Bethlehem, Pennsylvania, 18018 USA.

2024 - Volume: 64 Issue: 2 pages: 413-424

Original research

Keywords

Abstract

Introduction

The red larvae of the mite Leptus spp. Latreille, 1796 (Erythraeidae) are conspicuous, cosmopolitan ectoparasites of terrestrial arthropods including a diverse variety of insects (Baker 1982; Southcott 1992; Wendt et al. 1992; Haitlinger 2004, 2007; Mayoral and Barranco 2011; Pereira et al. 2012 ; Bernard et al. 2019) and arachnids (Åbro 1988; Cokendolpher 1993; Cokendolpher and Mitov 2007 ; Gabryś et al. 2011; Judson and Mąkol 2011; Mąkol and Felska 2011; Haitlinger et al. 2020). Several studies have investigated the ecological interactions of these mite larvae with harvestmen (Opiliones) hosts in Nearctic (McAloon and Durden 2000; Townsend et al. 2006) and Neotropical forests (Townsend et al. 2008). Erythraeid larvae do not form a stylostome, but rather use the chelicerae to penetrate the cuticle of the host (Baker 1982), becoming firmly anchored at the original site of attachment without any apparent aid from the pedipalps (Åbro 1988). For dragonfly hosts in Zambia, larvae of Leptus appear to select attachment sites with a softer cuticle (Bernard et al. 2019). Upon insertion, the chelicerae become ensheathed in glue-like cement that externally forms a cone while the distal-most parts of the mite's chelicerae become distended within the tissue of the host (Baker 1982; Åbro 1988). The buccal apertures of attached larvae possess ''tentacular processes'' that may aid in feeding (Åbro 1988). Mites are hypothesized to feed upon hemolymph plasma and cellular debris of the host for an undetermined period (Åbro 1988). Differences in size and engorgement of larvae on the same host suggests asynchronous infestation and repeated infestation events (Bernard et al. 2019). Eventually, the larvae detach from the host and molt to become nymphs, then adults, that act as predators of small insects (reviewed by Cokendolpher 1993). Relatively little is known about the impact of parasitic larvae upon the survival, locomotion, or reproduction of their harvestmen hosts (Guffey 1998; Cokendolpher and Mitov 2007; Bernard et al. 2019).

Larval mites attach to the appendages (chelicerae, pedipalps, legs) and body (dorsal scutum, free tergites) of harvestmen (Åbro 1988; McAloon and Durden 2000; Townsend et al. 2008). Intraspecific variation in body size among mite larvae has been attributed to varying degrees of engorgement (Åbro 1988) and different levels of success with regards to piercing the integument of the host (Åbro 1991), but could also result from the asynchronous infestation by larvae with extended emergence times (Bernard et al. 2019). More than 54 species of Leptus are known from North, Central and South America, with most taxonomic descriptions based upon larval characters (Welbourn and Young 1987; Southcott 1992; Haitlinger 2004 ; Haitlinger et al. 2020) and there are probably more undescribed species in Neotropical forests. Mites preferentially attach to the femora of the legs of harvestmen, but are rarely observed on the tarsi (McAloon and Durden 2000). The general absence of mites from distal leg segments may reflect the propensity of harvestmen to perform ''leg threading'' or ''grooming'' behavior in which the tarsi and metatarsi are brought to the mouth and passed laterally between the bases of the palpi and leg I (Edgar 1971; Pereira et al. 2004). Harvestmen frequently engage in this behavior when active, often after feeding, and they may spend up to several minutes at a time grooming themselves (Edgar 1971). Prior studies have reported intensities of infestation varying from 1-3 larvae to over a dozen mites per host (McAloon and Durden 2000; Townsend et al. 2008). Prevalence of larval mite infestation has been observed to vary interspecifically (Townsend et al. 2008), seasonally (Townsend et al. 2006), and geographically (Townsend et al. 2008). Relatively little is known about intersexual or ontogenetic variation among harvestmen in the prevalence of erythraeid mite infestations.

In this study, we examined the interactions of larval erythraeid mites with the Neotropical cosmetid harvestman Erginulus clavotibialis (Pickard-Cambridge 1905). E. clavotibialis is one of the largest species of harvestmen in Belize (6-7.5 mm body length as adults) and is relatively widely distributed, occurring throughout central and northern Belize as well as in Mexico (the Yucatan Peninsula) and Guatemala (Goodnight and Goodnight 1976). As is the case with many gonyleptoidean harvestmen (Buzatto and Machado 2014), there is considerable sexual dimorphism among adult E. clavotibialis and there are two distinct male morphotypes. In general, females are slightly larger in total body length, but have relatively smaller chelicerae and lack armature on leg IV (Goodnight and Goodnight 1976; Townsend and Enzmann 2018). In contrast, males have relatively larger chelicerae and patella-tibia IV is armed with large protuberances or spines on the prolateral and retrolateral surfaces (Goodnight and Goodnight 1976; Townsend and Enzmann 2018). Large males that possess relatively prominent chelicerae and robust spines on leg IV are known as α males, whereas individuals with smaller chelicerae and less prominent spines are β males (Buzatto and Machado 2014). We used scanning electron microscopy (SEM) to examine the mode of attachment to the host and compared our observations with prior descriptions (Baker 1982; Åbro 1988, 1991). In addition, we surveyed larval mite infestations for three different populations of harvestmen in Belize and thus investigated geographic variation in prevalence and intensity of mite infestation. We also examined potential differences in parasitism of larvae upon hosts of different life history stages (nymphs and adults), sex, and morphotypes (α and β males). Finally, we analyzed the distribution of mites upon different regions of the host body to determine whether the larvae exhibited a preference for attachment sites for specific regions or leg segments of the harvestman.

Material and methods

From 19-30 July 2018, we captured 116 individuals (penultimate nymphs and adults) of the cosmetid harvestman Erginulus clavotibialis by hand from beneath logs and rocks in forested habitats adjacent to hiking trails at three sites in Belize (Table 1): Clarissa Falls, Cayo District (17.1155, -89.1120); Las Cuevas Research Station, Chiquibul National Forest Reserve, Cayo District (16.7349, -88.9867); and the Columbia River Forest Reserve, Toledo District (16.3171, -88.9344). Earlier nymphal instars or larvae of harvestmen were not observed in the field. Prior to preservation in 70% ethanol, we examined each harvestman and determined the sex, morphotype (α or β, if male), and age (penultimate nymph or adult). In addition, we noted the presence, location and number of mite larvae that were attached to each of the appendages (chelicerae, pedipalps or walking legs) and different body regions (Tables 1, 2). With the aid of morphological characters from published keys (Southcott 1992; Haitlinger 2004; Mayoral and Barranco 2011), we identified the mite larvae as Leptus sp., but were unsuccessful in further determining the species present in our samples.

We prepared 30 larvae for examination with scanning electron microscopy (SEM), including 20 individuals that were found dislodged from their hosts in the storage containers and an additional 10 mites that were still attached to their hosts (in these instances, we removed the host's appendage with the attached mite and prepared them for SEM together). All larvae (detached and attached) were carefully dehydrated in a graded ethanol series, dried with hexamethyldisilazane (HMDS) according to the protocol described by Nation (1983), and mounted on aluminum stubs with carbon adhesives. Specimens were sputter-coated with 10-20 nm of gold and examined with the Hitachi S-3400 VP SEM on the campus of Virginia Wesleyan University. We viewed and photographed the mites at accelerating voltages of 5-10 kV.

Prevalence of infestation was defined as the percentage of harvestmen with at least one mite, whereas the mean intensity of infestation was the average number of mites per infested host (Schmidt and Roberts 1989). Our original dataset had observations for harvestmen from three sites (Clarissa Falls, Columbia River, and Chiquibul), two ontogenetic stages (adults and penultimate nymphs), and three adult morphotypes (α males, β males, and females). We report prevalence and intensity for all harvestmen that we collected, but restricted our statistical analyses to cases where we had sufficient sample sizes for that comparison. This ensured a sample size of at least 27 for each group. Because we collected only seven specimens from Colombia River (6 females, 1 male, 0 nymphs) we omitted those observations from our dataset. Additionally, because nymphs were only collected at Clarissa Falls (but not the other sites) we were only able to test for differences between ontogenetic stages at that site. Data for the two male morphotypes (α and β males) were combined because we only collected 11 β males, but are reported separately (Tables 1, 2).

To test whether there was an association between the prevalence of infestation and each of the three predictor variables (site, sex, and ontogenetic stage) we used Fisher's Exact Test implemented in R (R Core Team 2023) with'fisher.exact()' in package'exact2x2' (Fay 2010). We next created a subset of data that included individuals infested with at least one mite to test whether the intensity of infestation was explained by site (Clarissa Falls and Chiquibul) or sex (male and female). We fit a generalized linear model using a negative binomial distribution implemented using the'glm.nb()' function in package'MASS' (Venables & Ripley 2002). We compared four models (null, sex only, site only, and sex + site) and selected the best fit model using the corrected Akaike Information Criterion (AICc).

Finally, we wanted to know whether larvae were more likely to attach to specific regions of the body or legs in comparison to others. We counted the number of mites attached to the leg segments of the four walking legs (trochanter, femur, tibia, patella, metatarsus, tarsus), chelicerae, pedipalps, and body. We calculated the percent surface area for each body region and leg segment following the methods described by McAloon and Durden (2000) with the following modifications. Two specimens (β males) were dissected and the dorsal scutum, ventral coxae, free tergites, chelicerae, pedipalps and legs I-IV (trochanter-tarsus) were photographed using a Leica EZD4 digital stereomicroscope. The photomicrographs were calibrated using the Leica Application Suite software, version 2.0 with one pixel = 0.0066 mm. With the aid of Adobe Photoshop 11.02, we measured the surface area for each body region (dorsal scutum, ventral coxae, and posterior free tergites). For the flattened appendage segments (femur and tibia of the pedipalp and tibia IV), we measured surface area from the lateral perspective and multiplied by a factor of two to account for both surfaces of the segment. We also measured the length and width of the other segments for each appendage (chelicerae, pedipalps, legs I-IV). For these segments, we calculated surface area using > www.calculator.net> based upon the assumption that these structures were cylindrical. We used mite count data and surface area for two analyses.

In the first analysis, we used a Χ² goodness-of-fit test (α = 0.05) to test whether the observed number of mites differed from the expected number found on three regions of the body (dorsal sclerites, free tergites, and coxae including the ventral surface) and six appendages (chelicerae, pedipalps, and the four walking legs). We used a proportional model that assumes that the probability of attachment of a larva to the cuticular surface of the harvestman is random, and is therefore proportional to the surface area of the body or leg region of the host (Table 4). Post-hoc Χ² tests were used to compare each of the nine attachments sites to the sum of the other eight. The most conservative option, a Bonferroni correction, was applied to the α level for post-hoc tests (α = 0.05/9 = 0.0056).

In the second analysis, we used a Χ² goodness-of-fit test (α = 0.05) based on a proportional model to test whether mites were more likely to attach to a particular segment of the leg. To construct the proportional model, we recalculated the surface area for each leg segment relative to the surface area of all the legs (Table 5), and excluding the surface area of the body, chelicera, and pedipalps. We then combined the data for each segment of the four legs (i.e., trochanters I + II + III + IV). Post-hoc Χ² tests were then used to compare each of the six leg segments to the sum of the other five. A Bonferroni correction was applied to the α level for post-hoc tests (α = 0.05/6 = 0.0083).

Results

In Belize, adults and nymphs of Erginulus clavotibialis are commonly infested by larval erythraeid mites of at least one unidentified species of Leptus. With the aid of SEM, we examined the attachment sites of mites on various appendages of their hosts (Figure 1). Our observations confirm those of Baker (1982) and Åbro (1988) in that larvae attach by only their chelicerae and produce a cementing substance that forms an externally visible cone that helps hold them in place. In addition, we were able to observe the morphology of the distal region of the chelicerae within the cone in several specimens that were still attached (Figure 2A-D) as well as many that had been dislodged (Figure 2E-H). In these specimens, we found two clusters of 20-40 cilia/stereocilia-like processes on the dorsal aspect of the oral cavity. These processes were generally obscured and thus difficult to characterize in the attached mite (Figure 2A-D), but readily visible in the detached specimens. Åbro (1988) alluded to these structures as ''tentacular processes'' but provided no further details as to their size or number. In our specimens, these structures measured 20-30 μm in length and were each less than 1 μm in diameter. Because we did not use TEM to determine the composition of the cytoskeleton, we are unable to definitively classify these structures as mobile cilia (with an axoneme composed of microtubules) or immobile stereocilia (supported internally by actin microfilaments).

In addition to observing the components of the attachment organ, we also observed the presence of old cones, which may represent abandoned attachment sites , often near the attached larvae (Figure 3). These abandoned cement cones were generally similar in surface texture to the cones of active attachments sites and most were similar in size or infrequently smaller than those associated with the attached mites. Of the ten larvae that were still attached to the host cuticle, we observed the presence of old cones near these mites in six instances.

The proportion of adult males and females infested by larval mites (Tables 1, 2) did not differ significantly (P = 0.18). At Clarissa Falls, there was no difference between the proportion of adults and nymphs that were infested by mites (P = 0.23). Although there were no differences between sex or ontogenetic stage, we did find a significant difference between field sites. Adults at Chiquibul were 6.1 times more likely (95% CI = 2.1-19.1) to be infested by at least one larva compared to adults from Clarissa Falls (Fisher's Exact Test, P < 0.0004).

The model that best explained variation in the intensity of infestations was the site only model (AICc = 146.42, AIC weight = 0.59). Along with this model, the null model and the site+sex model had a cumulative weight of 0.94 (Table 3). The model that included only sex had a relatively low AIC weight of 0.06.

The observed proportion of mites attached to nine regions of the body and appendages (Table 4) was significantly different from the expected proportion of larvae based on surface area (Χ² = 85.79, df = 8, P < 0.0001). Post-hoc comparisons revealed that the proportion of mites attached to the chelicerae (Χ² = 15.14, df = 1, P < 0.0001) and pedipalps (Χ² = 12.56, df = 1, P < 0.0001) were significantly lower than expected while the proportion of mites attached to leg IV was significantly greater than expected (Χ² = 61.87, df = 1, P < 0.0001). The proportion of larvae attached to the dorsum, free tergites, coxae, and legs I-III were not significantly different from the expected values based on the proportional model (Table 4).

The observed proportion of mites attached to six different segments of the legs (Table 5) was significantly different from the expected proportion of mites based on surface area (Χ² = 60.15, df = 5, P < 0.0001). The proportion of larvae attached to the femur was significantly higher than expected (Χ² = 52.92, df = 1, P < 0.0001) while the proportion of mites attached to the tarsus was lower than expected (Χ² = 10.15, df = 1, P = 0.001). The proportion of larvae attached to the trochanter, patella, tibia, and metatarsus were not different from the expected proportions (Table 5).

Discussion

Our SEM-based observations of detached larval mites yielded considerably clearer views of the ''tentacular processes'' than previous studies (e.g., Åbro 1988). While these structures superficially appear somewhat ''tentacle-like'', the processes associated with the distal tips of the chelicerae of larval mites are remarkably similar in diameter and relative length to stereocilia and cilia. The primary differences between these specializations of the cell membrane are in relation to their cytoskeletal composition and general mobility. Stereocilia are supported by actin microfilaments and are nonmotile, whereas cilia feature a 9+2 internal arrangement of microtubules and are specialized to oscillate (Fawcett 1994). Åbro (1988) commented on the position of the ''tentacular processes'' of the chelicerae surrounding the buccal aperture, presumably these structures aid in feeding upon hemolymph and cellular debris of the host. Unfortunately, the primary application of SEM is the examination of surface features rather than cellular ultrastructure. Identification of the internal cellular structure and composition of the ''tentacular processes'' of larval mites will require transmission electron microscopy or confocal laser microscopy in a future study.

Our observations of larval mites still attached to the leg segments of harvestmen provide at least one additional interesting insight into these parasite-host interactions. In six of ten samples, we observed the presence of old cement cones near those of attached larvae. These cones were similar in size or smaller than those with mites attached, indicating that larvae may preferentially attach to areas on the host previously used by other mites and also suggests that harvestmen host may be infested asynchronously through the emergence of larval mites over an extended period of time as hypothesized for dragonflies infested by Leptus in Zambia (Bernard et al. 2019). The life history of cosmetid harvestmen is poorly known, but it appears likely that they live for several years following the final molt to adulthood (Goodnight and Goodnight 1976; Gnaspini 2007). It is not known how long larval mites remain attached to their hosts (Åbro 1988), but in our samples, it seems likely that hosts are infested by larvae over an extended period of time and that the emergence of larvae may be affected by local environmental conditions. Given the fact that adults do not molt, old cones may persist for weeks, months or perhaps longer periods of time. Late emerging larvae may use the presence of old cones produced by earlier emerging mites as indicators of suitable attachment sites, signaling to the parasite that these areas represent sites where the cuticle is relatively softer or can be penetrated effectively by their chelicerae.

In our study, the overall prevalence of infestation was 34%. These data are generally consistent with observations reported in prior studies of larval mites infesting harvestmen (Guffey 1998; McAloon and Durden 2000; Townsend et al. 2006, 2008). Most harvestmen that we captured had relatively few mites (the mean intensity was less than four larvae per host) and even the harvestmen that were relatively highly infested (\textgreater5 mites per host) did not act or move in any manner that suggested that they were burdened or negatively impacted by the mites. Although we did not find any significant ontogenetic or sexual variation in prevalence or intensity, our study differs from other published works in that we are the first to compare the prevalence and intensity of mites infesting males, females, and nymphs. Although our collection times were limited and uneven across locations, we observed significant variation in the prevalence of mite infestations between two field sites. Harvestmen at Chiquibul were much more likely to be parasitized by larvae than individuals at Clarissa Falls. The Clarissa Falls reserve is a much smaller forest fragment (a contiguous tract of approximately 150 ha), whereas the Chiquibul Forest Reserve is part of one of the largest tracts of intact forest north of the Amazon (over 500,000 ha in size). While we did not collect environmental data that clearly shows a dramatic difference in environmental conditions of these sites over an extended period preceding our collections, we did notice that the forested habitat was generally drier at Clarissa Falls. Forest fragmentation has been shown to impact humidity by increasing desiccation (Laurence 2004) and most of our sampling at Clarissa Falls occurred within 50 m of the forest edge, adjacent to a livestock pasture. During night sampling, we observed surface activity by harvestmen at Chiquibul and collected several individuals moving in the leaf litter and in the grass growing on the hiking trails. In contrast, we only captured harvestmen at Clarissa Falls from beneath logs and rocks and did not observe any individual moving through the leaf litter (night or day). In addition, Clarissa Falls was the only field site in Belize where we collected nymphs. We hypothesize that the drier conditions at this site may have affected the ecological interactions between mite larvae and their hosts. Low humidity may have slowed growth and foraging activity of harvestmen nymphs affecting their potential exposure to infestation by mite larvae. Drier conditions may have also impacted (and restricted) the emergence times of larval mites, thus reducing overall infestation rates. Wetter conditions at Chiquibul that favored activity and growth by harvestmen may also have supported a much larger population of erythraeid mites or created an extended period of emergence by larval mites. Wendt et al. (1992) noted a preference for humid and temperate conditions for all life history stages by Leptus trimaculatus.

Previous studies of larval mite-harvestmen interactions have noted significant preferences by Leptus spp. for attachments sites on the femora (McAloon and Durden 2000; Townsend et al. 2008) and tibiae (Townsend et al. 2008) of the legs and a general scarcity of mites infesting chelicerae, pedipalps and the tarsi of legs (McAloon and Durden 2000). In our study, we found a preference of mites for attaching to femur IV and a significant lack of mites on the tarsi of the legs, chelicerae and pedipalps. The external surfaces of femur IV of adult Erginulus clavotibialis feature several rows of setae. Prior studies (Åbro 1988; McAloon and Durden 2000) have also hypothesized that the non-random distribution of larvae on the host represents selection for specific attachment sites on the host by the mite and that these areas are weaker (or relatively thinner) than other locations (i.e., mites preferentially attach to leg segments but are rarely observed attached to the dorsal scutum). In addition, harvestmen exhibit grooming behavior (leg threading) in which the distal leg segments (e.g., the tarsi and metatarsi) are passed between the chelicerae and bases of the pedipalps and coxae I (Edgar 1971; Pereira et al. 2004; Acosta and Machado 2007). This action may enable the harvestman to effectively remove mites from these segments before they attach, whereas larvae attached to proximal segments (i.e., trochanter, femur, patella) are not affected by ''grooming'' and remain attached to the host. In contrast to the suggestions of Baker (1982) and Wendt et al. (1992), we believe that the distribution of larval mites upon their hosts is not random. On the bases of our observations, we conclude that multiple factors contribute to the selection of attachment sites by larval mites including variation the presence of more setae on leg segments as compared to the body, the ability of the host to remove larval mites from distal leg segments during grooming behavior, and the relative attractiveness of old, abandoned cones from prior mite infestations.

Acknowledgments

We thank Amanda Albert and Monika Metro for assistance with the field collections of harvestmen in Belize. We are grateful to Diana Vick and Kennedi Jones for assistance with SEM specimen preparation, Azucena Galvez of Clarissa Falls Resort and Rafael Mesh, station manager of Las Cuevas Research Station, for logistical support. We thank two anonymous reviewers for their constructive comments on an earlier version of this manuscript. This research was supported by funding from a VWU Faculty summer development grant (VRT) and the VWU SEM laboratory. Research was conducted with permission of the Belize Ministry of Forestry under scientific collecting permit WL/2/1/18(29).

References

1. Åbro A. 1988. The mode of attachment of mite larvae (Leptus spp.) to harvestmen (Opiliones). J. Nat. Hist., 22: 123-130. https://doi.org/10.1080/00222938800770091
2. Åbro A. 1991. Unsuccessful parasitic associations of mite larvae (Leptus spp.) to harvestmen (Opiliones). Fauna Norv. Ser. B., 38: 43.
3. Acosta L.E., Machado G. 2007. Diet and foraging. In: Pinto-da-Rocha R., Machado G., Giribet G. (Eds.). Harvestmen: The biology of Opiliones. Cambridge: Harvard University Press. p. 309-338. https://doi.org/10.2307/j.ctv322v442.12
4. Baker G.T. 1982. Site attachment of a protelean parasite (Erythraeidae: Leptus sp.). Experientia, 38:923. https://doi.org/10.1007/BF01953655
5. Bernard R., Felska M., Mąkol J. 2019. Erythraeid larvae parasitizing dragonflies in Zambia-description of Leptus (Leptus) chingombensis sp. nov. with data on biology and ecology of host-parasite interaction. Syst. Appl. Acarol., 24: 790-813. https://doi.org/10.11158/saa.24.5.6
6. Buzatto B.A., Machado G. 2014. Male dimorphism and alternative reproductive tactics in harvestmen (Arachnida: Opiliones). Behav. Proc., 109: 2-13. https://doi.org/10.1016/j.beproc.2014.06.008
7. Cokendolpher J.C. 1993. Pathogens and parasites of Opiliones (Arthropoda: Arachnida). J. Arachnol., 21: 120-146. https://www.jstor.org/stable/3705822
8. Cokendolpher J.C., Mitov P.G. 2007. Natural enemies. In: Pinto-da-Rocha R., Machado G., Giribet G. (Eds.). Harvestmen: The biology of Opiliones. Cambridge: Harvard University Press. p. 339-373. https://doi.org/10.2307/j.ctv322v442.13
9. Edgar A.L. 1971. Studies on the biology and ecology of Michigan Phalangida (Opiliones). Misc. Pubs. Mus. Zool. Univ. Mich., 144: 1-64. https://deepblue.lib.umich.edu/bitstream/handle/2027.42/56388/MP144.pdf?sequence=1
10. Fay M.P. 2010. Confidence intervals that match Fisher's exact or Blaker's exact tests. Biostatistics, 11: 373-374. https://doi.org/10.1093/biostatistics/kxp050
11. Fawcett D.W. 1994. Bloom and Fawcett: A textbook of histology, 12^th Edition. New York: Chapman & Hall. pp. 964.
12. Gabryś G., Felska M., Kłosińska A., Staręga W., Mąkol J. 2011. Harvestmen as hosts of Parasitengona (Acari: Actinotrichia, Prostigmata) larvae. J. Arachnol., 39: 349-351. https://doi.org/10.1636/CP10-93.1
13. Gnaspini P. 2007. Development. In: Pinto-da-Rocha R., Machado G., Giribet G. (Eds.). Harvestmen: The biology of Opiliones. Cambridge: Harvard University Press. p. 455-472. https://doi.org/10.2307/j.ctv322v442.17
14. Goodnight M.L., Goodnight C.J. 1976. Observations on the systematics, development, and habits of Erginulus clavotibialis (Opiliones: Cosmetidae). Trans. Am. Micro. Soc., 95: 654-664. https://doi.org/10.2307/3225390
15. Guffey C.A. 1998. The behavioral ecology of two species of harvestmen (Arachnida: Opiliones): The effects of leg autotomy, parasitism by mites, and aggregation. Ph.D. Dissertation, University of Louisiana at Lafayette, Lafayette, Louisiana, pp 179.
16. Haitlinger R. 2004. Three new species of Leptus Latreille, 1796 and the first record of Leptus onnae Haitlinger, 2000 (Acari: Prostigmata: Erythraeidae) from Brazil. Syst. Appl. Acarol., 9: 147-156. https://doi.org/10.11158/saa.9.1.20
17. Haitlinger R. 2007. A new genus and nine new larval species (Acari: Prostigmata: Erythraeidae, Eutrombidiidae) from Benin, Ghana and Togo. Rev. Ib. Aracnol., 14: 109-127. http://sea-entomologia.org/Publicaciones/RevistaIbericaAracnologia/RIA14/109_128_RHaitlinger_AcarsFromTogo.pdf
18. Haitlinger R., Šundić M., Ázara L., Bernardi L.F.O. 2020. A new species of larval Leptus (Leptus) (Trombidiformes: Erythraeidae) from Brazil with list of host-parasite associations between Leptus and arthropods in America. Biologia, 75: 1921-1930. https://doi.org/10.2478/s11756-020-00453-7
19. Judson M.L.I., Mąkol J. 2011. Pseudoscorpons (Chelonethi: Neobisiidae) parasitized by mites (Acari: Trombidiidae, Erythraeidae). J. Arachnol., 39: 345-348. https://doi.org/10.1636/CHa10-69.1
20. Laurence W.F. 2004, Forest-climate interactions in fragmented tropical landscapes. Phil. Trans. R. Soc. Lond. B., 359: 345-352. https://doi.org/10.1098/rstb.2003.1430
21. Mąkol J., Felska M. 2011. New records of spiders (Araneae) as hosts of terrestrial Parasitengona mites (Acari: Actinotrichida: Prostigmata). J. Arachnol., 39: 352-354. https://doi.org/10.1636/CP10-72.1
22. Mayoral J.G., Barranco P. 2011. A new species of larval Charletonia (Parasitengona: Erythraeidae) and new records of larval Erythraeidae parasitizing Orthoptera and Phasmida from Costa Rica. Acarologia, 51: 219-227. https://doi.org/10.1051/acarologia/20112023
23. McAloon F.M., Durden L.A. 2000. Attachment sites and frequency distribution of erythraeid mites, Leptus indianensis (Acari: Prostigmata), ectoparasitic on harvestman Leiobunum formosum. Exp. Appl. Acarol., 24: 561-567. https://doi.org/10.1023/A:1026554308826
24. Nation J.L. 1983. A new method using hexamethyldisilazane for preparation of soft insect tissues for scanning electron microscopy. Biotech. Histochem., 58: 347-351. https://doi.org/10.3109/10520298309066811
25. Pereira W., Elpino-Campos A., Del-Claro K., Machado G. 2004. Behavioral repertory of the Neotropical harvestman Ilhaia cuspidata (Opiliones, Gonyleptidae). J. Arachnol., 32: 22-30. https://doi.org/10.1636/S02-35
26. R Core Team. 2023. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.
27. Schmidt G.D., Roberts L.S. 1989. Foundations of Parasitology, 4^th Edition. St. Louis, C.V. Mosby Co. pp. 750.
28. Southcott R.V. 1992. Revision of the larvae of Leptus Latreille (Acarina: Erythraeidae) of Europe and North America, with descriptions of post-larval instars. Zool. J. Linn. Soc., 105: 1-153. https://doi.org/10.1111/j.1096-3642.1992.tb01228.x
29. Townsend Jr. V.R., Enzmann III B.P. 2018. Ontogenetic variation in the sensory structures on the pedipalps of cosmetid harvestmen (Arachida, Opiliones, Laniatores). J. Morphol., 279: 109-131. https://doi.org/10.1002/jmor.20758
30. Townsend Jr. V.R., Mulholland K.A., Bradford J.O., Proud D.N., Parent K.M. 2006. Seasonal variation in parasitism by Leptus mites (Acari, Erythraeidae) upon the harvestman Leiobunum formosum (Opiliones, Sclerosomatidae). J. Arachnol., 34: 492-494. https://doi.org/10.1636/T05-44.1
31. Townsend Jr. V.R., Proud D.N., Moore M.K., Tibbetts J.A., Burns J.A., Hunter R.K., Lazarowitz S.R., Felgenhauer B.E. 2008. Parasitic and phoretic mites associated with Neotropical harvestmen from Trinidad, West Indies. Ann. Entomol. Soc. Am., 101: 1026-1032. https://doi.org/10.1603/0013-8746-101.6.1026
32. Venables W.N., Ripley B.D. 2002. Modern applied statistics with S. Fourth Edition. New York: Springer. pp. 510. https://doi.org/10.1007/978-0-387-21706-2
33. Welbourn W.C., Young O.P. 1987. New genus and species of Erythraeinae (Acari: Erythraeidae) from Mississippi with a key to genera of North American Erythraeidae. Ann. Entomol. Soc. Am., 80: 230-242. https://doi.org/10.1093/aesa/80.2.230
34. Wendt F.E., Olomski R., Leimann J., Wohltmann A. 1992. Parasitism, life cycle and phenology of Leptus trimaculatus (Hermann, 1804) (Acari: Parasitengonae: Erythraeidae) including a description of the larva. Acarologia, 33: 55-68.

instructions