Protura in Arctic Regions, with Description of Mastodonentomon n. gen. (Acerentomidae, Nipponentominae) and a Key to Known Arctic Taxa

Protura are widespread, but their presence in the Arctic was first noted only ca. 70 years ago and is still little acknowledged. This work compiles taxonomic information on proturans in the Arctic regions and adds unpublished data from Northern Siberia. Currently, this fauna is represented by 23 species in two orders and 14 genera. The large cosmopolitan genus Eosentomon is represented by only four species, whereas Acerentomidae is much more diverse, with 19 species in 13 genera (eight Nipponentominae, five Acerentominae). Most of the Arctic species possess a larger number of setae than species living in temperate regions. Based on several unique characters, a new genus, Mastodonentomon, is erected for Nipponentomon macleani, and the species is re-described with the original description supplemented with new characters, including head chaetotaxy, seta length, and porotaxy. Proturan occurrence in the Arctic is limited to Beringia, but the majority of species have restricted distributions and none have been found in both the American Arctic and Siberia. This implies relict origins and high levels of proturan endemism in the Arctic. This emerging view on biogeographical history is, however, hampered by the limited extent of available data, which highlights the need for considerably greater survey efforts. A key to Arctic proturans is provided to facilitate further studies.


Introduction
Protura, known as coneheads, are among the smallest of soil-dwelling microarthropods. These hexapods are wingless, eyeless, and have slender elongate bodies ranging between 0.7-2 mm. Due to their minuteness and cryptic edaphic lifestyle, they are easily overlooked and were discovered relatively late in the history of entomology [1]. Originally regarded as an order belonging to the class Insecta [2], Protura was later placed among the basal hexapod taxa as the sister taxon to either Collembola or Diplura. Their phylogenetic position, however, is still a point of controversy [1,3]. Since their first description in 1907 [2], over 800 valid species belonging to three orders, seven families, and

History of Proturan Research in the Arctic
Protura were first collected inside the Arctic Circle by M. Hammer, who, in 1948, collected well over 100 specimens from two different locations above 68 • N in Northern Canada [10]. The majority of these proturans were described as Acerentulus canadensis Tuxen, 1955, and six specimens from one sample as Nosekiella condei Tuxen, 1955 (Table 1). Currently, these species are in the genera Verrucoentomon and Vesiculentomon, respectively [19,22]. Hammer's collection contained one specimen belonging to the genus Eosentomon. It was, however, a "larva 2" and, therefore, indeterminable to species [10]. Weber [23] mentioned a record of "Acerentulus sp. (Family Acerentomidae)" collected in 1949 from two locations in Alaska, extending the northernmost proturan record past 69 • N. Then it was not until summer 1976 when a notable collection of proturans from a number of localities across Alaska was made by S.F. MacLean and A. Fjellberg (Institute of Arctic Biology, Fairbanks AK) [11] with the northernmost localities above 70 • N. Based on that collection, Nosek [11][12][13] described 14 species, including three Eosentomidae and 11 Acerentomidae, and wrote a key [13] for determining all proturan species known from Alaska. Lastly, Bernard (1985) [14] described Eosentomon adakense from samples collected in 1978 on Adak Island (Aleutian Islands, Alaska). Some of the Acerentomidae species have been transferred to other genera: Orinentomon, Sugaentulus, and Imadateiella [19,20,24]. The genus Nosekientomon was erected for the species Vesiculentomon ruseki Nosek, 1977 [19]. Two Alaskaentomon species and Verrucoentomon imadatei Nosek, 1977 are valid and were re-described recently [21,22]. Revision of the type material of Nipponentomon macleani Nosek, 1977 shows that this species has some unique characters within Acerentomidae and a new genus has to be established for it.
The first known Arctic proturan in the Palearctic was collected from Snow Valley in the Magadan district in Siberia: Acerella sharovi Martynova, 1977 [15]. This species was transferred to Imadateiella by Imadaté [25]. Three proturan species were described from the taiga belt near Turukhansk City, close Insects 2020, 11, 173 3 of 17 to the Arctic Circle [16,18,26], and Nienna chukotka Shrubovych, 2019, was described from Northern Chukotka above 69 • N [27]. This record and a new record of Yamatentomon yamato Imadaté & Yosii, 1956, from northern Yakutia above 72 • N, are currently the northernmost known Palearctic localities of Protura.

Faunistic Dataset
A dataset of the distribution of proturans in Arctic regions was created based on the publications concerning these species. In addition, new data from Northern Yakutia (Siberia) reported in the current work were also included. Geographically, the dataset covers all studies conducted in the Arctic, as defined by the Arctic Monitoring and Assessment Programme (AMAP) [28], including all the terrestrial areas north of the Arctic Circle (66 • 33 N) as well as most areas north of 60 • N in Asia and North America with continuous permafrost, and the Aleutian Islands. It should be borne in mind that a significant proportion of these areas are often defined as Subarctic, but we follow the AMAP terminology for simplicity.
The dataset (Table 1) includes: (1) the valid species names and synonyms, and their taxonomic position, (2) sites where proturans were collected and their coordinates (with the type locality marked with *), (3) type of habitat in which they were found, as given in the original descriptions, (4) remarks on their known distribution, and (5) literature sources of the original records and species re-descriptions if available.

Material Examined
Proturan specimens from Northern Yakutia (Siberia) were extracted from soil and moss samples with Berlese-Tullgren funnels into 95% ethanol. The specimens were mounted on glass slides in Faure's medium [29] for taxonomic evaluation. The examined specimens were deposited in the collection of the Institute of Systematics and Evolution of Animals, Polish Academy of Sciences, Kraków (ISEA). In addition, the type material (holotype female and paratype male) of Nipponentomon macleani, preserved in the National Museum of Natural History (Smithsonian Institution) in Washington, DC, USA, was also studied using oil immersion and phase contrast on an Olympus microscope.

Species and Their Taxonomic Position Records from the Arctic Environmental Settings Distribution References
Subfamily Acerentominae Silvestri, 1907 Genus Fjellbergella Nosek, 1978 Fjellbergella
Dry alder litter in a mixed alder-aspen-birch forest.
Dry exposed hill with Dryas sp. and Carex rupestris.
Litter in shrubs with Betula sp. and Equisetum sp., N slope, rather moist.
1200 m elev., on exposed cliff in Dryas sp.

Species and Their Taxonomic Position Records from the Arctic Environmental Settings Distribution References
Verrucoentomon canadense (Tuxen, 1955)  Soil, litter, and moss in mixed and taiga forest.
Moist lichen slope in paper birch stand.

Faunistic Dataset
The faunistic dataset contains 37 records of Protura from 24 sites within the Arctic regions (Table 1), published originally in 11 papers. Of these, 26 records from 17 sites and 6 papers [10][11][12][13][14]23] concern Alaska and the Canadian Arctic and only 11 records from 7 sites and 5 papers [15][16][17]26,27] concern the Siberian Arctic. No records were found from the European Arctic ( Figure 1). Overall, 23 proturan species from two orders (Eosentomata and Acerentomata) and 14 genera were identified from both the North American and Siberian Arctic (17 and 6 species, respectively). The order Eosentomata is represented by only four species, which all belong to the genus Eosentomon. Acerentomata are considerably more diverse, with 19 species belonging to 13 genera (eight Nipponentominae and five Acerentominae). Moreover, two records undetermined to species level, both reported as "Acerentulus sp. (Family Acerentomidae)" [23], were listed from Northern Canada. Unfortunately, Weber [23] did not provide any descriptions and whereabouts of these collections are unknown, so they could not be re-examined. It should also be remembered that, since that time, Acerentulus has been split into several genera. Thus, at present, these records can be assigned, at best, to Acerentomidae only. Acerentulus has been split into several genera. Thus, at present, these records can be assigned, at best, to Acerentomidae only. Species richness for particular sites ranged from one to four even though the majority of sites (18) hosted only single species and only three sites (i.e., Chena Ridge and Denali National Park in Alaska, and Turukhansk in Siberia) had four species. However, it should be borne in mind that these estimates are mostly based on historical data taken from the literature and they might be biased by the sampling effort. Nearly all species reported from Alaska and the Canadian Arctic (i.e., 13 of 17 species) are known only from their type localities and only four were also found in other sites across this region. On the other hand, among Siberian species, only Nienna chukotka Shrubovych, 2019 is known exclusively from its type locality. The other five species seem to have broader distributions across Siberia and records of Yamatentomon yamato (Imadaté and Yosii, 1956), reported in the present paper from the northernmost locality, span a wide latitudinal range across the Northeastern Palearctic. It is also remarkable that none of the species have been found in both regions.   Species richness for particular sites ranged from one to four even though the majority of sites (18) hosted only single species and only three sites (i.e., Chena Ridge and Denali National Park in Alaska, and Turukhansk in Siberia) had four species. However, it should be borne in mind that these estimates are mostly based on historical data taken from the literature and they might be biased by the sampling effort. Nearly all species reported from Alaska and the Canadian Arctic (i.e., 13 of 17 species) are known only from their type localities and only four were also found in other sites across this region. On the other hand, among Siberian species, only Nienna chukotka Shrubovych, 2019 is known exclusively from its type locality. The other five species seem to have broader distributions across Siberia and records of Yamatentomon yamato (Imadaté and Yosii, 1956), reported in the present paper from the northernmost locality, span a wide latitudinal range across the Northeastern Palearctic. It is also remarkable that none of the species have been found in both regions.

Remarks:
The specimens are identical with more southern specimens (see References [18,33]) in body length, chaetotaxy, and porotaxy. The sensilla pattern on the foretarsus is similar, with the very long sensillum b surpassing the seta γ4 insertion point. Unlike southern specimens, foretarsal sensillum a is shorter. Its apex reaches to the sensillum t2 insertion, and does not reach the base of sensillum d. This small character difference is not enough to justify a description of a new species.

Mastodonentomon Shrubovych, New Genus
Characterized by the presence of four pairs of A-setae on meso-and metanota (A1, A2, A3, A4), setae M2 and M3 on metanotum, six pairs of A-setae on tergites I-III (additional setae A1' present), five pairs of A-setae on tergites IV-VII, filiform foretarsal sensillum t1, sensillum b' missing, position of foretarsal sensillum b proximal to c insertion, d situated closer to e than to c, sensillum a' in distal position, close to c' insertion, well-developed maxillary, and labial palps with terminal tuft of setae, posterior position of P3 setae on tergites II-VI, well-developed striate band on segment VIII with distinct parallel-sided striae, 4/2 setae on sternite VIII, and presence of two setae of equal length on abdominal legs II and III.

Remarks:
The new genus Mastodonentomon is placed into subfamily Nipponentominae due to possessing a distinct calyx with large vesicle and racemose appendices on its surface, presence of four pairs of A-setae on metanotum, presence of 4/2 setae on sternite VIII, well-developed striate band on segment VIII, and two setae of equal length on abdominal legs II and III [5]. The new genus is similar to Nipponentomon and Imadateiella in having four pairs of A-setae on metanotum and filiform foretarsal sensillum t1. The genus differs in the presence of five pairs of A-setae on tergite VII and in the position of foretarsal sensillum a', which is placed far distally, close to c' insertion, as in some Acerentomon spp. [34,35]. The positions of sensillum a' on the foretarsus clearly separates most of the genera within Nipponentominae [19]. This new genus differs from all other Acerentomidae in the presence of four pairs of A-setae on the mesonotum (A1, A2, A3, A4) and three pairs of M-setae (M1, M2, M3) on the metanotum. Other Protura have only two or three pairs of A-setae on mesonotum and one pair of M-setae on the metanotum. These characters are important at the generic level and justify the erection of Mastodonentomon n. gen. within the subfamily Nipponentominae. (Nosek, 1977), New Combination ( Figure 2A-P, Table 2) syn. Nipponentomon macleani Nosek, 1977 Material Examined: Holotype female (no. 75801), collected in Dryas on exposed cliff, Mastodon Dome at Eagle Creek, 1200 m elev., Alaska, 1.VIII.1976, coll. A. Fjellberg. Paratype male collected together with the holotype. The slides are mounted in Swan's medium and stored in the National Museum of Natural History (Smithsonian Institution) in Washington, USA.

Mastodonentomon macleani
Diagnosis: Large species with a body length of more than 1500 µm. Setae A1 present on mesonotum, metanotum, and metasternum, setae M2 and M3 on metanotum, six pairs of A-setae on tergites I-III, five pairs of A-setae on tergite VII, five A-setae on sternites I-VI and three A-setae on sternite VII. Additional cephalic seta d6 very long; sensilla f and c' the longest on foretarsus. Pores al and sl present on meso-and metanota, group of sc pores present on mesosternum, two separate sc pores present on metasternum, psl present on tergites I-VII, pore spm present on sternites I-VII, a pair of sal pores present on sternite I, and a pair of spsm pores on sternites VI-VII.
Re-description: The maxillary and labial palpi, canal of the maxillary gland, arrangement of the foretarsal sensilla, setae of the third abdominal legs, striate band and comb on abdominal segment VIII, female squama genitalis were all accurately depicted by Nosek [11] to which the reader is referred for illustrations.
Chaetotaxy formula given in Table 2. Setae on nota differing in length ( Figure 2F). Pronotal seta 1 three times longer than seta 2. Meso-and metanota with seta A1. Seta P1a on meso-and metanota long, setiform, seta P2a short, half length of seta P1a; P2a situated midway between P2 and P3. Length ratio of mesonotal setae P1:P1a:P2 as 1.6:1:1.8. Mesonotum with a pair of single sl and al pores and metanotum with a pair of doubled sl pores and a pair of al pores ( Figure 2F,G). Prosternum without pores, mesosternum with a group of three closely placed sc pores, metasternum with a pair of separately placed sc pores ( Figure 2H-J). Seta A2 on thoracic sterna and P4 on tergite I short, setiform ( Figure 2I,J,M).
Abdominal segment VIII with a distinct striate band. Tergite and sternite each with anterior, irregular, transverse row of small teeth (see Reference [11]: Figure 7). Pore psm with several accompanying teeth. Posterior margin of sternite VIII and laterotergites smooth. Comb VIII with 9-10 long teeth (see Reference [11]: Figure 6G). Sternites IX-X with central setae half the length of lateral setae (see Reference [11]: Figure 6H). Tergite IX with minute serrations in the central part of the hind margin ( Figure 2O), tergites X-XI, and sternite X with distinct serrations on the entire posterior margin ( Figure 2P). Sternite IX with serrations on the lateral part of the hind margin, sternite X with serrations on the entire posterior margin (see Reference [11]: Figure 6H). Dorsal lobe of tergite XII with a median pore and a ventral lobe with 1+1 anterolateral pores.
Etymology: The generic name is taken from the general locality (Mastodon Dome) where the specimens were collected.
Remarks: In this re-description of Nipponentomon macleani, morphological characters are added to the previous description of the species, such as chaetotaxy of head, length of seta on head, nota and foretarsus, and porotaxy on tergites and sternites. According to Nosek [11] tergite I has 6 A-setae and margin ( Figure 2P). Sternite IX with serrations on the lateral part of the hind margin, sternite X with serrations on the entire posterior margin (see Reference [11]: Figure 6H). Dorsal lobe of tergite XII with a median pore and a ventral lobe with 1+1 anterolateral pores.

Discussion
The proturan fauna of the Arctic regions is not rich but is quite diverse, represented by two orders (Eosentomata and Acerentomata), 14 genera and 23 species ( Table 1). The order Sinentomata has not been recorded from this region, but only five species are known at present [4]. Considering over 800 species of Protura known worldwide [1,4,18], the proportion of the global proturan diversity found in the Arctic is very low (less than 3%). Species richness recorded for particular sites was also very low and the majority of sites hosted only single species. Nevertheless, nearly 20% of 76 genera known worldwide [5] have been found in the region. Members of the widespread genus Eosentomon, which are very abundant and diverse in the Americas and Europe [36][37][38], are poorly represented, with only four known species, and restricted in distribution to Alaska and Northern Canada (Table 1). However, the lack of Eosentomon in the Siberian Arctic is not surprising because species of this genus are generally rare and not numerous in neighboring regions [18,39]. Acerentomidae is very diverse in the Arctic at the generic level and comprise eight genera of Nipponentominae and five of Acerentominae. The genus Verrucoentomon is most diverse in the Arctic fauna. Its members have usually been collected in mountainous regions, such as V. montanum (Martynova, 1970) and V. rafalskii Szeptycki, 1997. Such apparently relict species may have been able to survive on non-glaciated mountaintops [18,40]. The subfamilies Berberentulinae and Acerellinae apparently are absent from the Arctic regions. Berberentulinae is rich in species and widely distributed. Acerellinae is restricted to four species know only from Europe [4]. Members of these subfamilies are characterized by a reduction of some morphological characters, e.g., shape of labial palps, striate band, number of setae on body segments, and sensilliform accessory setae on the body. The acerentomids on the northern edge of proturan distribution possess well-developed labial palps with a terminal tuft of setae, well-developed striate band on segment VIII (except Fjellbergella and Tuxenentulus spp.), and a larger number of setae on the body than species with a more southern distribution.
In general, Protura can be found wherever decaying organic matter and sufficient moisture levels are available [1,8], and often are restricted to habitats characterized by luxuriant vegetation growth [7]. It is believed that proturans are fungivorous and some species may feed preferentially on ectomycorrhizal fungi [41]. It is not surprising, therefore, that proturans have rarely been collected in tundra biome, which marks the northern limit of their distribution. They have never been collected in the taiga biome in the European part of Russia, but proturans occur in Siberian taiga and tundra biotopes [42].
Alaska, Yukon, and Siberia are a part of the Beringian region that was mostly ice-free during the Pleistocene glaciations [32] and, therefore, proturans are found further north there than in the rest of the Holarctic. This suggests the importance of historical biogeography in the distribution of proturan species, which was a hypothesis highlighted by Tuxen [43]. Furthermore, records suggest pronounced biogeographical differences in distribution of proturan species within the Beringian region. For instance, no species were found in both the North American and Siberian Arctic. This disparity implies the existence of a strong intercontinental disjunction in proturan distributions between these two Arctic regions. Moreover, the majority of species seem to have regionally restricted distributions or are known only from their type localities. This pattern is particularly pronounced for the proturan fauna in North American Arctic, but also among Siberian species. Only Yamatentomon yamato is widely distributed over a broad latitudinal range extending far beyond the Arctic region (see Reference [18]). While substantial northward migration and post-glacial recolonization from the south is apparent from datasets for some previously glaciated regions, e.g., Southern Canada [44] or Southern Fennoscandia [4], it appears that further north in the Arctic regions distribution of proturans is still limited to the areas of Beringia that were unglaciated during the Pleistocene (Figure 1). Moreover, the restricted distributions and morphological differences among some taxa, when compared to species with a more southern distribution (as indicated above), imply a relict origin and a potentially very high level of endemism in proturan fauna of that region. This hypothesis is consistent with distribution patterns reported for some other arthropod taxa, particularly those with limited dispersal capabilities (see Reference [45]), which were presumably eliminated or prevented from dispersing by glaciers elsewhere, but for which Beringia acted as a refugium. However, the available data on proturan occurrence are fragmentary and restricted to very few specific collection localities. Many areas have not been sampled at all. Thus, the apparent restricted distributions may, in part, reflect the limited sampling intensity and location of specific collection sites rather than the real distribution of these animals. While our work provides both a benchmark for the region and the foundation for future research, considerably greater survey efforts are necessary to elucidate the true diversity of proturans and biogeographical patterns of their distribution in the Arctic regions.

Conclusions
Although proturans are often believed to be absent in polar regions, their occurrence above the Arctic Circle was noted more than 70 years ago with the northernmost records exceeding 72 • N. While early works focused on documenting and describing new species, primarily in Alaska and the Canadian Arctic, recent advances in proturan taxonomy have facilitated re-descriptions of several taxa and documented the occurrence of proturans throughout Northern Siberia. Accordingly, in the current work, we re-describe Nipponentomon macleani, supplementing the original description with new characters, including head chaetotaxy, seta lengths, and porotaxy, and, on the basis of several unique characters, establish a new genus, Mastodonentomon, for this species. Nonetheless, the Arctic literature dataset contains only 35 records representing 23 species, i.e., less than 3% of the global proturan diversity. In contrast, generic diversity is relatively high but dominated by Acerentomidae. Another striking feature of the collected dataset is that occurrence of proturans in the Arctic regions appears to be limited to regions of Beringia that were glacier-free during the Pleistocene. Emerging patterns in species distribution and their distinct morphology also imply a relict origin and a potentially high level of endemism within the proturan fauna of that region. It is clear, however, that proturans in the Arctic regions have been studied very selectively. Hence, consideration of the importance of biogeographical history and the concept of endemism is itself strongly hampered by the limited extent of available data, which highlights the need for considerably greater survey efforts. Still, this work provides a strong baseline needed to facilitate further studies addressing diversity and biogeography of these little-known arthropods in the Arctic regions and elsewhere.