New species of springtails in the Proisotoma genus complex from Vermont and New York, USA with descriptive notes on Ballistura alpa Christiansen & Bellinger 1980 (Hexapoda, Collembola, Isotomidae).

Abstract Three new species of Isotomidae springtails are described from the Lake Champlain Basin (Vermont and New York, USA), Lake Willoughby and Greater Averril Pond in Vermont. Subisotoma joycei sp. n. and Scutisotoma champi sp. n. were collected in sandy beaches whereas Ballistura rossi sp. n. was found only in a constructed wetland built and managed by the University of Vermont. Scutisotoma champi sp. n. was found in Lakes Champlain and Willoughby, and Greater Averril Pond and is probably present in most lakes and large ponds in the area. Subisotoma joycei sp. n. was found only along the southern and eastern coast of South Hero, and the mainland coast facing eastern South Hero. Ballistura alpa is redescribed and transferred to the genus Pachyotoma based on the absence of tibiotarsal seta B4/B5, the presence of secondary cuticular granules, 4 prelabral setae, a full complement of guard setae on labial papilla E and in having a bifurcate outer maxillary lobe with 4 sublobal setae.


Introduction
The springtail fauna of Vermont is poorly known. The most comprehensive list includes only 57 species (Bellinger 1982). Most of the springtail collections studied by Bellinger (1982) were made by Ross and Joyce Bell, and their students at the University of Vermont in the highlands around Camel's Hump. To date the fauna of the Vermont side of the Lake Champlain basin remains mostly unexplored.
As part of a general survey of the springtail fauna of sandy beaches on the Vermont side of Lake Champlain four species in the family Isotomidae belonging to the Proisotoma genus complex were collected. Further analysis showed that the samples included Proisotoma minuta (Tullberg, 1871) and three undescribed species. In the present contribution we describe the new species and provide additions to the description of Ballistura alpa Christiansen & Bellinger, 1980.

Methods
Most of the terminology used in the descriptions follows Potapov (2001), but we retain the use of dorsal and ventral to identify the faces of the furcula (posterior and anterior, respectively, in Potapov 2001). Postlabial chaetotaxy (other than the number of setae along the ventral groove) has not been used to diagnose species in this group of Isotomidae and its utility is uncertain. However, the general organization of the chaetotaxy in the forms described here is the same as in Lepidocyrtini entomobrids (Soto-Adames 2010), suggesting it may be useful to distinguish species among members of the Proisotoma genera complex. The postlabial chaetotaxy (Fig. 3) is described following the nomenclature and column delimitation conventions of Soto-Adames (2010). The columns of setae are designated as (I)nner, (C)entral, (E)xternal, and (L)ateral. The level of intraspecific variation in the number of setae in each column differs among species. The 5 individuals of Subisotoma joycei sp. n. studied showed variation in the number of setae in all postlabial columns, but no variation in number of setae was observed in 19 individuals of Scutisotoma champi sp. n. Columns I and E are most stable. Column C is usually not well organized and is probably best described as a field of setae; this column also shows the largest amount of intraspecific variation in the species examined .
The number of postlabial setae and tergal sensilla are often variable and the formula is presented in the format x (y) where x is the mode and y represents other setae number observed. If there is more than one mode the formula is expressed as x/z (y). Separation between thoracic and abdominal tagma is represented by //, segments within tagma are separated by a semicolon (;). Number of setae are given for half a tergite (i.e., formula 10//101 should be understood as 10//101 + 10//101) Abbreviations used in the descriptions that follow are: Ant., Th., Abd., L and PAO for antenna, thorax, abdomen, leg and postantennal organ.
The types of the new species are deposited in the Insect Collection of the Natural History Survey, at the University of Illinois, Champaign, IL, USA. The slides of P. alpa are deposited in the Zadock Thompson Natural History Collection at the University of Vermont, Burlington, Vermont, USA.
All collections were made by extracting a plug of sand or soil using a commercial bulb planter. Sand/soil plugs were placed in plastic bags in the field and transported to the laboratory. In the laboratory each plug was washed in water, the water was filtered using commercial coffee maker filters, and the content of the filters was extracted in Berlese funnels (with 15 watts light bulbs) for three days. All sand plugs were moist, as completely dry sand cannot be extracted with a bulb planter. Most samples included only sand, but others contain variable amounts of surface plant debris and the species collected could be either in the sand or on the surface plant debris. Etymology. The new species is dedicated to Ross Bell in celebration of his contributions to our understanding of the entomological fauna of Vermont.
Remarks. Following Potapov (2001), Ballistura rossi sp. n. is most similar to B. hankoi (Stach), 1929 from which it can be distinguished (Table 1) by the number of dorsal manubrial setae (13 in rossei, 20 in hankoi), number of setae around the PAO (3 in rossei, 2 in hankoi), and number of dorsal setae on dens (11-12 in rossi, 10 in hankoi). Ballistura tuberculata (Stach), 1947 (if this is really different from B. hankoi) can be separated from B. rossi by the same characters of the furcula and PAO as B. hankoi, and by coloration (pale blue-grey in tuberculata, dark purple in rossi), size (largest individual of rossi is 0.5 mm whereas tuberculata reaches 0.9 mm) and, possibly, shape of the basal membrane of the mucro (wider at middle in rossi, wider on basal half in tuberculata). Other chaetotaxic characters may distinguish these three species, but practically nothing has been published about the chaetotaxy B. hankoi or B. tuberculata (Potapov 2001). Christiansen and Bellinger (1998) reported B. tuberculata from Indiana and Nova Scotia, but whereas the relatively large size of those individuals (up to 0.8 mm) support the determination as tuberculata, the relative size of the OPA and shape of the mucronal membrane suggest similarities to hankoi. The specimens from Vermont fit the general description provided by Christiansen and Bellinger (1998) for B. tuberculata, except for the larger number of dorsal setae on the dens and the smaller size of the Vermont specimens (Table 1).
Remarks. Three individuals show variation in the number and size of eyes. In the small juvenile eyes G and H are small, barely rising above the cuticle; one male is blind; and in one female all eyes in one patch are subequal, but on the other patch eye E is less than half the size of F. The axial setae are often disorganized and the number of setae in a column is open to interpretation. The number of tergal sensilla is variable. Most individuals have an asymmetric number of sensilla, and two individuals lack the microsensilla of Abd. 1 on one side. Tenent hair B7 on metathoracic legs often appears acuminate instead of capitate. Two individuals have 3+4 tenacular teeth.
The generic placement of the new species is problematic. It better fits in the genus Subisotoma (Table 2), but it is unique among species currently assigned to that genus  in having more than 8+8 tergal sensilla on each segment, by the significantly larger number of dental setae (most Subisotoma species have 4 or fewer dorsal and 1-2 ventral setae, whereas S. joycei has 18-20 dorsal and 4-6 ventral setae), and by having a well developed furcula with mucro exhibiting a wide lamella and clearly separated from the dens. Subisotoma joycei is similar to species in the genus Isotopenola Potapov, Babenko, Fjellberg & Greenslade, 2009 in the presence of sensillar polychaetosis on body terga, but differs from all forms in that genus by having smooth thoracic sterna, lacking an isolated field of setae on Abd. 3 sterna and in the number of guard setae on labial papilla E. The strong polychaetotic furcula in S. joycei resembles the condition in Ballistura, but the new species clearly differs from Ballistura in maxillary palp structure, sensillar and microsensillar formulae, presence of a full complement of setae in tibiotarsal whorl B, and dens sculpturing ( Table 2). The new species is most similar to Ballistura ewingi James, 1933, sensu Folsom (1937 from which it differs in aspects of color pattern (trunk ventrally white in ewingi, dark purple brown in joycei), the number of tenent hairs (2-3 on all legs in ewingi, 1, 2, 2 in joycei), number of distal seta on ventral tube (4 on ewingi, 11 in joycei), and number of tenacular teeth (2 in ewingi, 3-4 in joycei). The new species may be the same as the Pennsylvania specimens preliminarily assigned to B. ewingi by Christiansen and Bell- (Folsom, 1937, Fig. 128) ?
14-16? (Folsom, 1937, Fig. 136) Ventral Setae on Dens 4-6 6 5 at least 2 (Folsom, 1937, Fig. 136) inger (1998), although this form also seems to have considerably fewer distal setae on the ventral tube than S. joycei sp. n. (Table 3). Ballistura ewingi has been described as having smooth dens, and probably does not belong in Ballistura, which Potapov (2001) restricts to species with tuberculate dens. Important characters needed to determine the appropriate generic placement of B. ewingi remain undescribed and require the study of fresh material. The new species is also similar to B. excavata Folsom, 1937, but the two species are easily separated by body color, eye number, shape of tenent hairs and unguiculus, and structure of the dens (Table 3). Etymology. The species is named after 'Champ' the denizen monster of Lake Champlain, were the new species seems to be most abundant.
Scutisotoma champi sp. n. is the most common species found in sandy beaches on the northern 2/3 of Lake Champlain as well as Lake Willoughby and Greater Averill Pond, and it is likely present in most if not all lakes and large ponds in northern Vermont and southern Quebec. The species was collected in apparently healthy beaches (e.g., Pearl Bay, locality H), as well as on highly disturbed, strongly compacted beaches (e.g., Colchester Beach, locality C). The species is most abundant in beaches with aquatic plant litter, but it is also found in sand in beaches without visible surface plant remains.
Remarks. This species was originally placed in the genus Ballistura, but the presence of a complete whorl B on pro-and mesothoracic legs and four prelabral setae excludes it from that genus (Potapov 2001). The presence of secondary granules on the cuticle in combination with a well developed furcula and the insertion of the medial and almost all other tergal sensilla on Th. 2-Abd. 3 on the posterior row of setae place this species in the genus Pachyotoma. In the original description of the species Christiansen and Bellinger (1980) indicate the presence of 5 distal setae on the ventral tube and a toothless unguis. The individuals from Vermont have 7 distal setae on the ventral tube and one distinct inner tooth on all unguis and may represent a distinct species.