1The Pennsylvania State University, Department of Entomology, 501 ASI Building University Park, PA 16802, USA.
2Canadian National Collection of Insects, Arachnids, and Nematodes, Science and Technology Branch, Agriculture and Agri-Food Canada, 960 Carling Avenue, K.W. Neatby bldg., Ottawa, Ontario, K1A 0C6, Canada.
3The Pennsylvania State University, Department of Entomology, 501 ASI Building University Park, PA 16802, USA.
4Canadian National Collection of Insects, Arachnids, and Nematodes, Science and Technology Branch, Agriculture and Agri-Food Canada, 960 Carling Avenue, K.W. Neatby bldg., Ottawa, Ontario, K1A 0C6, Canada.
5✉ The Pennsylvania State University, Department of Entomology, 501 ASI Building University Park, PA 16802, USA.
2024 - Volume: 64 Issue: 2 pages: 575-591
https://doi.org/10.24349/40qb-lv0mParantennulidae is one of 21 families of mesostigmatic mites (Acari: Parasitiformes) that belong to the infraorder Antennophorina (Trigynaspida), a taxonomically diverse lineage of symbiotic mites (Kim 2004). Most antennophorine mites are associated with passalid, carabid and scolytine beetles, as well as ants and myriapods, and are generally considered commensals – if not mutualists, when feeding on potential parasites or pathogens on the host or in its nest – with little or no evidence for hurting their hosts, although many species are kleptoparasitic, with unknown impact on their host (Seeman and Walter 2023). The taxonomic diversity of Antennophorina is concentrated in tropical-subtropical regions, although a few families have representatives in temperate regions, including Parantennulidae. The family includes two monotypic genera, Parantennulus Berlese, 1903, and Diplopodophilus Willmann, 1940, which are both associated with myriapods, and three described species of Micromegistus Trägårdh, 1948, all of which are associated with ground beetles (Coleoptera: Carabidae) (Kim and Castagnoli 2010). Micromegistus gourlayi Womersley, 1958 was described from New Zealand on Mecodema sp. (Broscinae) (Kim and Castagnoli 2010; Womersley 1958) and the same species, or a close relative (identified as Micromegistus cf. gourlayi), was found in Australian caves in the same area as two large pterostichine carabids (Hamilton-Smith 1967). Micromegistus viduus (Berlese, 1888) occurs on Scarites sp. in South America (Berlese 1888, Kim and Castagnoli 2010), whereas M. bakeri Trägårdh, 1948, the focus of this paper, is known primarily from Scarites subterraneus Fabricius, 1775. Additionally, there are undescribed Micromegistus species associated with ground beetles in Australia (Seeman and Nahrung 2000; O. Seeman pers. comm. 2024) and leafcutter ants in Brazil (Kim and Castagnoli 2010), and a possible Micromegistus that was photographed on a tenebrionid in Australia (Lambert 2023).
Micromegistus bakeri was first described from adult specimens collected from Scarites subterraneus (Carabidae: Scaritinae; Figures 1–2) in Mississippi (Trägårdh 1948). Following this discovery, M. bakeri was reported from Kansas by Nickel and Elzinga (1970a), who redescribed the species and detailed most of our current knowledge about its natural history, including feeding biology. They confirmed that M. bakeri occurred most frequently on S. subterraneus, but also reported a few individuals from two other carabid species: Cyclotrachelus sodalis (LeConte, 1846) (reported as Evarthrus sodalis colossus in their study), and Patrobus longicornis (Say, 1823). Since then, M. bakeri was subsequently reported from Padre Island, Texas (McDaniel and Bolen 1979), but has otherwise received no other scientific study. It has not been reported from any of the six other species of Scarites in North America north of Mexico.
We address this knowledge gap in several ways. First, we present the first record of M. bakeri from Pennsylvania, which represents the easternmost record of this species, and report the results of a survey of local carabid fauna for M. bakeri. We also summarize previously unreported museum records of M. bakeri from Ohio, Michigan, and Missouri, along with suspected records based on photographs of specimens from community science websites. Using this data, we provide a distributional map of all confirmed and suspected M. bakeri localities overlaid on the range of Scarites spp. found in the United States and Canada. Furthermore, we provide digital photographs of live M. bakeri specimens, as well as the first publicly available COI barcode sequence for this taxon.
Major specimen collections in the USA were consulted for data on Micromegistus bakeri, including the Ohio State University Acarology Collection (OSAL) and the United States National Museum (USNM). The species identity of specimens was not verified by us but was considered reliable as they were identified by previous experts, including J.H. Camin, Beverly Gerdeman, Donald E. Johnston, and Richard C. Funk. The Canadian National Collection of Insects, Arachnids and Nematodes (CNC) and Frost Entomological Museum (PSUC) were also consulted but had no specimens of M. bakeri. Museum abbreviations follow Evenhuis (2024).
Ground beetles were collected at Stone Valley East Entrance in Pennsylvania, Centre County, State College (40°39′46.7″ N 77°54′33.4″ W) (Riggins 2023). Beetles were located by walking along a concrete path with a flashlight after dark, after which they were collected by hand. Non-Scarites specimens were collected from April through October 2021. Two Scarites were collected in September 2021, while the remaining Scarites specimens were collected in May and June 2022. Non-Scarites carabids were euthanized via freezing and stored at -20 °C until they could be examined under an Olympus SZ-6145TR Trinocular Stereo Microscope for mites. Scarites specimens were kept alive and reared following the protocols outlined below. Ground beetles were identified to species using published keys (Ball and Bousquet 2000, Ciegler 2000).
No collection permit was needed due to the field site being on Pennsylvania State University property.
Scarites subterraneus were kept in individual plastic containers containing Eco Earth Loose Coconut Fiber Substrate (Zoo Med, San Luis Obispo, CA) and fed Organix® Organic Chicken Flavor Dog Cookies (Castor & Pollux, Clackamas, OR). A folded paper towel was dampened and placed on one side of the container to add moisture to the habitat. Water was also added directly to one side of the container to keep half the substrate moist. The food and towel were changed twice weekly or when mold became evident in the enclosure. The substrate was changed completely once a month. The container was kept in the lab on a shelf near a window. Beetles and mites lived in the lab from mid-September 2021 to late March 2022 and May 2022 to mid-April 2023, representing two different sets of beetles.
The specimens were observed and photographed in petri dishes using an Olympus SZ-6145TR Trinocular Stereo Microscope 0.67x – 4.5x with an attached 10 MP Olympus Camera. The photographs were processed, and measurements conducted using the Olympus cellSens Standard 1.18 (Build 16686) Imaging Software.
Several mite specimens were removed from a single S. subterraneus specimen, stored in 70% ethanol and sent to the Ottawa Research and Development Centre (ORDC), Agriculture and Agri-Food Canada (Ottawa, Ontario, Canada) for identification. Mites were slide-mounted in a polyvinyl alcohol medium and identified under a Leica DM5500B compound microscope equipped with differential interference contrast, using the literature (Kethley 1977; Kim and Castagnoli 2010; Nickel and Elzinga 1970a).
The DNA barcode region of cytochrome c oxidase I gene (Hebert et al. 2003) was sequenced from a single mite (CNC1898834) at the ORDC's Sequencing Facility using the LCO1490 and HCO2198 primer pair (Folmer et al. 1994) following protocols outlined in Nowell and Schwarzfeld (2019) with the following exceptions. PCR amplification was performed in 7.5 µL reaction volumes comprised of 3.75 µL 2X Multiplex PCR Master Mix (QIAGEN Inc.), 0.075 µL (10 µM) of each primer, 2.35 µL of ddH2O, and 1.25 µL of template DNA. Sequencing reactions were performed in 10 µL reaction volumes with 9 µL of 1:8 diluted (per the manufacturers protocol) BrilliantDye Terminator v3.1 (Edge Biosystem), 0.34 µL ddH2O, 0.16 µL primer, and 1 µL purified PCR product. Following extraction, the DNA was archived at −80 °C and the voucher specimen was recovered and deposited in the Canadian National Collection of Insects, Arachnids, and Nematodes (CNC), Agriculture and Agri-Food Canada (Ottawa, Canada). The sequence was assembled and edited in Geneious Prime (v. 11.0.3), validated against the Barcode of Life Datasystem's (BOLD) reference library through the Identification Engine, and is publicly accessible through BOLD (SampleID = CNC1898834) and GenBank (Accession = ON257217).
Known localities for Micromegistus bakeri in North America were compiled from published literature (Trägårdh 1948; Nickel and Elzinga 1970a; McDaniel and Bolen 1979), as well as specimens housed in the OSAL and vouchered in BOLD. Photographs of S. subterraneus; S. lissopterus Chaudoir; S. marinus Nichols; S. ocalensis Nichols; S. quadriceps Chaudoir; S. stenops Bousquet & Skelley; S. vicinus Chaudoir; and unidentified Scarites posted to iNaturalist (https://www.inaturalist.org
) and BugGuide (https://bugguide.net
) were examined for visible mites to document additional suspected locality records (Table 1). The identities of all Scarites that had mites were confirmed by the authors. The map was created in Adobe Illustrator using Google Maps to mark localities. Because (1) the range of S. subterraneus encompasses the ranges of the other six Scarites species found in the United States and Canada, (2) the species identity of many Scarites posted on iNaturalist could not be confirmed beyond genus due to the absence of morphologically informative characters in photographs, and (3) Micromegistus-like mites were found on Scarites species other than S. subterraneus (see ''Utility of citizen science photographs'' below), the range map is based on all Research-grade Scarites observations available on iNaturalist.
In total, 573 ground beetles, representing 32 species, were collected. Micromegistus bakeri were found on 10 of 37 (27%) of S. subterraneus, but none of the 31 other carabid species (Table 2). Micromegistus bakeri was the only mite species observed on the field-collected ground beetles, although the subelytral space was not examined for mites. Although most carabid taxa were collected in low abundances (17 species plus unidentified Amara spp.), we collected at least 10 specimens from each of 13 non-Scarites taxa (12 species plus unidentified Agonum spp.). Furthermore, six of these species were found in abundances (33–147 individuals) similar to or greater than S. subterraneus, suggesting that M. bakeri do not use these carabids as hosts despite their similar prevalence in this locality. To our knowledge, Nickel and Elzinga (1970a) conducted the only other assessment of host range in M. bakeri, which encompassed a single locality in Kansas. They observed a high prevalence of M. bakeri on S. subterraneus (52 of 112 specimens; 46.4%). They also observed the mites on two other carabid beetles, Cyclotrachelus sodalis LeConte (reported as Evarthrus sodalis colossus in their study), and Patrobus longicornis, that occurred at the same site. Infestation rates on these two carabid species were high (25% and 33% of beetles had mites, respectively), but these figures were based on the inspection of only four and six specimens, respectively. With such limited data on the association between M. bakeri and these two apparent hosts, it remains unclear whether these occurrences represent accidental phoresy or a true symbiotic relationship.
Nickel and Elzinga (1970a) did not indicate whether they surveyed other carabid species to test the host range of M. bakeri. However, their data are part of a larger dataset of mites collected from a total of 18 carabid species in the frame of Nickel's thesis (Nickel 1969), from which results on other mite species were published (Nickel and Elzinga 1969, 1970b). Those publications indicate the absence of M. bakeri from the other 15 carabid species studies. Other than S. subterraneus, five carabid species had more than 10 individuals collected and four had more than 100 individuals collected – none of which carried M. bakeri. This, and two additional elements, further support a strong host preference for Scarites or specifically S. subterraneus over other potential carabid hosts. Firstly, from 44 records of M. bakeri in other states (Table 3), 42 (95%) had Scarites identified as host, including 5 (11.3%) identified specifically as S. subterraneus. Secondly, based on male morphology, M. bakeri is most closely related to M. viduus, which occurs on Scarites sp. in South America (Berlese 1988, Kim and Castagnoli 2010).
The specimens from Pennsylvania represent a new state record. Sixty-three specimens are housed in the OSAL collection, of which 42 include locality data (Table 3) and represent new state records for Ohio, Michigan, Missouri, and new country and province records for Ontario, Canada (Figure 3). The specimen discovered through BOLD (see DNA results below) represents a new state record for Arizona.
Micromegistus bakeri favor the ventral surface of their hosts (Figures 1–2) and can remain somewhat hidden under the prothorax and beneath or around the legs. Nickel and Elzinga (1970a) found that the mites prefer the coxae, trochanters and femora of legs II, and will secondarily occupy the same segments of legs I and III, as well as the thorax near the narrow junction between the prothorax and mesothorax of the beetle. Our inspection of two S. subterraneus infested with M. bakeri revealed that the mites favored the area around the junction of the prothorax and mesothorax (Figure 1b) compared to the other regions of the beetle noted by Nickel and Elzinga (1970a). However, as the mite population grew on these two S. subterraneus kept in the lab, increased activity on the entire ventral surface was also noted, presumably because preferred sites were overcrowded. Immatures were often observed interacting with adult mites (Figure 1c shows a larva on an adult, probably a female), and were frequently found on the beetle's mandibles as well as occasionally walking across the elytra.
Photographs on BugGuide and iNaturalist of 6,692 individual Scarites were screened for mites. Of those, 122 beetles (1.82%) hosted Micromegistus-like mites (Table 1, Figure 2). Although these mites had the correct gestalt for M. bakeri, we cannot confirm the species identity of the mites from the photographs alone. We therefore report them here as ''suspected'' M. bakeri records that require confirmation through examination of physical specimens but do not include them in the new state records listed above.
A majority of the mites (73/122; 59.8%) found in citizen science photographs were on S. subterraneus, which is part of the S. subterraneus species group. The three other North American species of this group – S. marinus, S. ocalensis, and S. stenops – have restricted ranges in Florida and Louisiana so are rarely encountered or photographed. A few mites (12/122; 9.8%) were found on beetles of the S. quadriceps species group, which comprises three species: S. quadriceps, S. vicinus and S. lissopterus. Scarites quadriceps occurs from Maryland and Florida west to Texas. Scarites vicinus occurs from Ontario and Alabama west to Iowa and Texas, so has a somewhat more northern range compared to S. vicinus but they co-occur across much of the eastern United States. Scarites lissopterus occurs Louisiana and Kansas west to New Mexico. All three species co-occur in Louisiana, Kansas, and Texas. They are distinguished based on the length of the metasternum compared to the length of the metacoxa and differences in absolute body length, which are often difficult to see or determine from photographs; so, in most cases we could not confirm the identification of beetles beyond the S. quadriceps species group in areas where they co-occur. The remaining mites (37/122; 30.3%) were on Scarites whose species identification could not be confirmed beyond genus due to the lack of morphologically informative characters in the photographs.
Presuming the mites seen in citizen science photographs are indeed M. bakeri, their prevalence on Scarites spp. appears to be much lower than has been reported in previous studies (Nickel and Elzinga 1970a: 46.4%) and in this study (27%). This is likely due to the relatively poor quality of many photographs on iNaturalist and BugGuide and the mites' tendency to favor the beetles' venter, which is often hidden from view in photographs and impeded our ability to detect more S. subterraneus specimens harboring mites in these databases. It could also be that photographers avoided specimens with obvious mites or removed mites prior to photographing them, either intentionally for more visually compelling photographs or unintentionally by handling the beetles prior to photographing them, although we suspect that this happens infrequent enough that it is not the sole cause of the discrepancy. Alternatively, many S. subterraneus records on citizen science websites originated from urban areas, and lower mite infestation rates on other ground beetle species have been reported from urban compared to rural forest sites (e.g., Mizser 2016), so the discrepancy may reflect a real ecological trend, at least partially.
In addition to Micromegistus, we also detected other mites on photographed Scarites, although at much lower rates (Table 4). Four records consist of mesostigmatans that cannot be identified to family with confidence, and one represents a species of Astigmata, possibly of the family Histiostomatidae. One of the records (Minnesota) appears to represent at least two species of Mesostigmata, one of which, the most numerous, could be deutonymphs of the family Parasitidae. Their high numbers clearly indicate that it's not accidental phoresy.
The small size and the inherently limited public interest in mites restrict the use of crowd-sourced photographic data in acarology, which is highlighted by our inability to identify non-Micromegistus mites in photographs. Indeed, the only publication using citizen science photographs to study mesostigmatic mites compared the diversity of field-collected arthropods to citizen science photographs from the same region and reported an absence of mesostigmatic mite in the photographs (Dolezal 2019, Skvarla and Fisher 2023). This study, by comparison, is the first to use presence data to study mesostigmatic mites and demonstrates that citizen science photographs can provide a useful resource when certain criteria are met. In this case, the hosts are commonly photographed, so even a low detection rate of 1.8% produced more than 100 putative records of M. bakeri, and the mites have a fairly unique gestalt (e.g. idiosoma almost as wide as long and ovate, i.e. broader posteriorly, legs spread laterally and front legs antennae-like) that distinguish them from other Scarites-inhabiting mites, and from most other mites known in North America. While there are potential pitfalls associated with citizen science data such as their limited reliability for species identification and the impermanence of such data (see Skvarla and Fisher (2023) for a detailed discussion of this), it can provide additional data that would otherwise be unavailable, as shown by our expanded distribution map (Figure 3).
This study presents the first COI DNA barcode record for Micromegistus bakeri available in any public sequence repository, representing a 658 bp (32.5% GC) sequence lacking any ambiguous bases. In BOLD, there were six COI sequences representing potentially two other species of Parantennulidae. The nearest neighbor (6.6% p-distance) of this sequence on BOLD [Accessed May 17th, 2022] belonged to a previously unidentified Parantennulidae specimen (Sample ID = DPMIT-41-72) collected from Deadhorse Ranch State Park in Arizona, United States; we borrowed the specimen (from the Centre for Biodiversity Genomics) and identified it as M. bakeri based on the absence of any significant morphological differences with our specimen from Pennsylvania and on comparisons with published descriptions of M. bakeri. This finding indicates a relatively high intraspecific divergence in M. bakeri, which exceeds the typical divergence values for mesostigmatic mites (Li et al. 2012; Young et al. 2019). This high divergence may be in part due to the geographic distance between the specimens (~3,050 km), but without additional specimens it is unclear if these represent genetically divergent specimens on the edges of the species range or rather cryptic species that we cannot yet distinguish. The other parantennulid species represented on BOLD is M. gourlayi, which originated from four GenBank records (EU825790-EU825793) collected in New Zealand; it was at least 19.3% divergent (p-distance) from our new record of M. bakeri.
Scarites subterraneus are morphologically convergent with passalid beetles, sharing with them a large robust body, a stalked prothorax, strong mandibles, and a burrowing behavior. The species burrows under logs, rocks or litter, where it hides during the day, mostly motionless, and hunts at night, bringing their prey underground before feeding (Hlavac 1967; Larochelle and Larivière 2003). It occurs in anthropogenic and open natural habitats, such as lawns, pastures, open forests, and near water bodies. Their burrows can be as deep as 15 cm (Larochelle and Larivière 2003). Larvae live in the adult's burrows, but also leave the burrows at night for hunting. It overwinters as adults in burrows (Hlavac 1967).
Nickel and Elzinga's (1970a) study showed that Micromegistus bakeri essentially completes its entire life cycle on S. subterraneus. Female mites carry two large eggs within the opisthosoma which develop concurrently and are born viviparously as larvae. In laboratory conditions, it takes about two weeks before two other larvae are born. Nickel and Elzinga (1970a) observed the birth of a larva, and stated that after emergence, the larva clung onto its mother's venter for a few minutes, and then to its dorsum for a similar period of time. Finally, the larva ''descended to the beetle and wandered about for several minutes on the carabid's venter before settling near the junction of trochanter and femur II''. This indicates that larvae are born while the adult female is on the host. As much as 30% of all M. bakeri collected from S. subterraneus by Nickel and Elzinga (1970a) being larvae further supports that. Our own observations (Fig. 1) and museum data (Table 3) are compatible with this.
The relationship between M. bakeri and S. subterraneus is unclear. Published observations suggest that M. bakeri is an opportunistic scavenger, a kleptoparasite, and possibly also a'paraphage' that feeds on the host's dermal secretions; it was also proposed to be a commensal (Nickel and Elzinga 1971a). The overall net impact on their host, whether detrimental (antagonistic), neutral (commensal), or even beneficial (mutualistic), is unknown. The impact may even shift when various parameters, such as mite density on a host, change. The relationship between M. bakeri and S. subterraneus may represent a context-dependent commensalism, as recorded in other systems, including density-dependent changes from commensalism to antagonism and commensalism to mutualism (Mathis and Bronstein 2020). In the S. subterraneus-M. bakeri system, mites have been observed feeding on the remains of host-killed prey (a mealworm) near the host beetle, feeding on food particles near the beetle's mouth, feeding on debris on the surface of the beetle (Nickel and Elzinga 1970a), and even lapping up beetle regurgitation fluid off the host (Johnston et al. 1947). Due to the size difference between the mites and their hosts, a few mites may not steal enough nutrients from a beetle to have negative effects (i.e., they are commensals), but may negatively impact host nutrient uptake (i.e., they are antagonists) when populations are higher. Nickel and Elzinga (1970a) observed that the mites were ''in constant motion, with their chelicerae making lateral, wiping movements'' across the body of the beetle and performed a preliminary experiment where radioactive sodium acetate was injected into the beetles to test if M. bakeri would assimilate the sodium acetate and thereby indicate that the mite was feeding on the host's secretions. Their results were positive but inconclusive since the radioactive substance could have been absorbed without actual feeding on the host's secretions. This purported feeding on the host's secretions may have no effect on the host, be negative if it forces the host to expend energy replacing the excretions, or perhaps even beneficial if the mites are removing fungal spores or harmful bacteria. Finally, M. bakeri eliminates its waste onto the host (Nickel and Elzinga 1970a). During heavy mite infestations the waste accumulates as a soft mass on the anterior portion of the beetle's mesosternum and between coxae II. If this accumulation encourages detrimental microbe growth, it could negatively impact the beetle. In short, there are many potential impacts on the beetles that may vary with mite population size, so further experiments should be conducted to determine what, if any, impact the mites have on their hosts.
The mites appear to occasionally disembark their host, presumably for seeking food, and are sometimes brushed off by the beetle. Mites have been observed remounting their host, and the adults can even jump back onto the host by leaping a distance sometimes greater than four times their body length (Nickel and Elzinga 1970a). This suggests that M. bakeri is adapted to wander off the host, probably during the day when the host rests in its burrows. The mite could then scavenge on invertebrate remains left by its host in its burrow or possibly even prey on invertebrates inhabiting the burrows, such as nematodes, although such facultative predatory behavior has not been observed yet. A strange blattisociid mite, Krantzoseius walteri Seeman (2012), recently described in Australia may have a somewhat analogous lifestyle to M. bakeri. Like M. bakeri, all life stages of K. walteri appear to live on the host, Trichosternus perater (Pterostichini), a burrowing carabid unrelated to Scarites spp. Like S. subterraneus, T. perater brings captured prey into its burrow, where K. walteri possibly feeds on microfauna associated with the prey items, such as nematodes, but may also scavenge the rotting prey itself (Seeman 2012; O. Seeman pers. comm. 2024).
Phoretic histiostomatid deutonymphs, which are occasionally found sharing the same host as M. bakeri, were not attacked by M. bakeri when both were found on the same host (Nickel and Elzinga 1970a, Nickel 1969, Riggins 2023). Indeed, the weakly, nearly edentate chelicerae of M. bakeri, with their mop-like appendages, do not appear optimal for seizing active prey, but rather for mopping up fluids, such as those oozing out of dead invertebrates, a behavior seen in other Trigynaspida (Seeman 2000). The larvae, nymphs and adult females and males all having similar chelicerae (Nickel and Elzinga 1970a), which suggests similar feeding habits throughout ontogeny.
Micromegistus bakeri probably colonize new individuals of S. subterraneus during mating, if beetles rest near each other during the day (e.g., under a rock or log), or if new hosts enter the burrow as mites were seen to transfer between adult beetles when reared together (Nickel 1969, Riggins 2023).
The life history of other Micromegistus species is unknown beyond basic collecting information. Immatures of M. viduus have been found on ground beetle hosts, as have immatures of undescribed Micromegistus species from Brazil (on leafcutter ants) and Australia (on carabids) (Kim and Castagnoli 2010; Seeman and Nahrung 2000; O. Seeman pers. comm. 2024). This suggests that the immatures of all species in the genus live on the host, indicating an intimate association with the host beyond phoresy, like for M. bakeri.
Parantennulidae is one of at least 24 mite families that have species associated with carabid beetles worldwide (Haitlinger 1988, Fain et al. 1995, Trach 2016, Seeman and Baker 2013, Trach and Seeman 2014). Most of these mites are phoretic on their host, using it mostly or only for dispersal as deutonymphs or adult females, or as both adult males and females in the case of Antennophorina. Other than that, they spend their life in soil-litter or other detrital habitat (e.g., dead wood) as predators of small invertebrates (most Mesostigmata), scavengers (some Antennophorina) or fungivores (e.g., Acaridae, Scutacaridae). These phoretic mites are associated with a few to many species of Carabidae (e.g., Antennoseius spp., Ascidae) or form loose associations with various beetle groups (e.g., some Macrochelidae) or even other arthropods (many Acaridae; Parasitidae). Fewer families include permanent associates of carabids that live under the elytra of their hosts as parasites (e.g., Podapolipidae: Eutarsopolipus) or putatively commensals or paraphages (e.g., Canestriniidae) (Nickel and Elzinga 1969, Katlav et al. 2015, Seeman 2021, Haitlinger 1988).
Nickel's work in Kansas shows that Scarites subterraneus is no exception and has many (at least eight) mite associates, including phoretic symbionts such as occasional predators (Ascidae: Antennoseius sp.; Parasitidae sp.), fungivores (Acaridae sp.) and common microbivores (Histiostomatidae sp.), as well as more intimate symbionts, namely two antennophorines, M. bakeri and Echinomegistus wheeleri Wasmann (Table 5; Nickel 1969, Nickel and Elzinga 1969, 1970a, 1970b). Based on Nickel's survey (1969), Micromegistus bakeri is the most common mite species occurring on S. subterraneus, with 57% of all (263) mites found on that carabid species; histiostomatids were second, with 34% of all mites, and then E. wheeleri with 6%; other mites were relatively rare, with 1–4 mites found in total on S. subterraneus. One of those rare mites was Caraboacarus karenae (Caraboacaridae), although common on other carabid species (Nickel and Elzinga 1969). Caraboacarids have unknown feeding habits and obscure relationships with their carabid hosts, with only adult females found on them (Katlav et al. 2015). Like M. bakeri, Echinomegistus wheeleri appears primarily associated with S. subterraneus and seem to have a similar feeding behavior as M. bakeri, although only adult males and females are found on the beetle, with immatures probably developing in soil or the carabid's tunnels (Nickel and Elzinga 1970b). More recently, a parasitic mite, Eutarsopolipus scariteus Husband (2001) (Podapolipidae) was described from S. subterraneus in Louisiana.
We hypothesize that carabids that have a stronger tendency to burrow have more mite symbionts, especially those that use the host for more than phoresy, as indicated by the presence of both sexes or immatures on the host. Such intimate symbionts would also be favored by a long-term usage of the burrows by the host, since this facilitates access to food resources such as partly eaten prey and to potential mates, with a relative ease of finding and reembarking on the host. It is unclear how much the same shelters are re-used by Scarites species in natural habitats, besides those used for overwintering in the northern parts of its geographic range. Among the approximately 200 described species of Scarites worldwide (Ball and Bousquet 2000, Bousquet 2012), only a few (four species plus two identified only as Scarites sp.) are recorded to host mites (Table 5). That >2% of Scarites species have recorded mite associates is striking when S. subterraneus has eight taxa of mites associated with it, including three that are host specialists (or nearly so) like M. bakeri (Nickel 1969, Husband 2001). Presuming this lack of mite associates is an artifact of sampling and reporting bias, it suggests that there are many Scarites-mite associations left to discover, depending on the number of mite species per beetle species and their degree of host specificity. The same could be said about the \textgreater1800 species of carabids of the subfamily Scaritinae (Bousquet 2012), most of which also have subterranean lifestyle, but for which we are aware of only a few mite species reported on them (Eidelberg 1994; Hogan 2012; Katlav et al. 2015, 2021; Arndt et al. 2016; Seeman 2020).
Micromegistus bakeri appears to prefer Scarites subterraneus given its prevalence and abundance on that species compared to non-Scarites carabids. We also detected what are likely M. bakeri on other Scarites species, although these observations need to be confirmed through field-collected specimens examined in hand. Additional work is needed to understand the relationship between M. bakeri and its host(s), as well as infestation rates across geographical space and habitat types. The new confirmed and suspected records presented herein demonstrate that the geographic distribution of M. bakeri is larger than previously documented and likely mirrors that of S. subterraneus, which is widely distributed across the continental United States (Figure 3). Finally, the high divergence between the COI sequence presented here and the sequence available on BOLD suggests that there may be an interesting story to these easily collected mites if anyone cares to investigate it.
We thank Hans Klompen for providing locality information for M. bakeri specimens housed in the Ohio State University Acarology Collection; Owen Seeman and another reviewer for valuable comments in a previous version of the manuscript; Eleanna Vasquez Cerda, Chloe Crabb, Joe Gallardo, Uma Riggins, Jim Crab, Carolyn Riggins, Rosa Walker, and Lem Riggins for their help in collecting specimens. This material is based upon work supported by the National Science Foundation under Grant No. NSF DEB 1762760/1556898 (T. Renner, PI). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. This work is supported by the USDA National Institute of Food and Agriculture and Hatch Appropriations under Project #PEN04974 and Accession #7006543.

