nena masthead
NENA Home Staff & Editors For Readers For Authors

Seasonal Abundance of Tabanus calens and other Tabanidae (Diptera) near Snake Mountain, Addison County, Vermont
Jeffrey V. Freeman

Northeastern Naturalist, Volume 24, Issue 2 (2017): 137–151

Full-text pdf (Accessible only to subscribers. To subscribe click here.)

 

Access Journal Content

Open access browsing of table of contents and abstract pages. Full text pdfs available for download for subscribers.



Current Issue: Vol. 30 (3)
NENA 30(3)

Check out NENA's latest Monograph:

Monograph 22
NENA monograph 22

All Regular Issues

Monographs

Special Issues

 

submit

 

subscribe

 

JSTOR logoClarivate logoWeb of science logoBioOne logo EbscoHOST logoProQuest logo

Northeastern Naturalist Vol. 24, No. 2 J.V. Freeman 2017 137 2017 NORTHEASTERN NATURALIST 24(2):137–151 Seasonal Abundance of Tabanus calens and other Tabanidae (Diptera) near Snake Mountain, Addison County, Vermont Jeffrey V. Freeman* Abstract - Weekly collections, from May to September over 3 years from one 2-tier box trap in Addison, VT, totaled more than 29,000 tabanids representing 44 species and included 417 Tabanus calens. Seasonal succession was evident, with T. calens most numerous in the upper trap in August. This very large horse fly was easily collected around a bait horse with an aerial net. Other abundant horse fly species were Hybomitra lasiophthalma, T. quinquevittatus, T. lineola, and T. sackeni. The most abundant deer flies were Chrysops univittatus and C. vittatus. Most deer flies showed a preference for the upper trap, whereas horse flies showed a mixed preference. Tabanus quinquevittatus made up 60% of the collected horseflies in the trap. The trap collected 41% less horse flies (73% less T. calens) and 24% less deer flies with the trap in 2013 to compared to in 2015. Both the presence of T. calens and the use of the 2-tier box trap in Vermont at one site for 3 whole seasons were new events. Tabanus calens was not listed in the 1990 checklist of Vermont tabanids. These results are presented in the context of trap modifications, range extension, and polarized light. Introduction I know of 2 previous collections of Tabanus calens L. (Fig. 1), one that I collected at the Mt. Independence Historic site in Orwell in 2000, and another above Hemenway Road in Bridport in August of 2006 (Graham 2006). In 2012, the author received a photograph of a dead T. calens that led to planning a 3-year study of T. calens and other tabanids at a location near Snake Mountain (Mountain Road Extension) in Addison, VT, utilizing a 2-tier box trap (Fig. 2). The objectives were to find out if a trap of this design would detect the presence of this large horse fly and other Tabanidae, to characterize seasonal abundance of tabanid species at 1 location, and to explore the possibility of lowering a local tabanid pest population. The importance of T. calens is that it attacks people as well as horses. It is one of the largest of horse flies; one can hear this fly approaching. I have talked to people who recall when it was not present in the area, so it appears to be a non-endemic pest. This study sought to learn more about T. calens and other tabanids during 3 whole seasons at 1 location. There have been several other tabanid surveys in the northeastern US. Bartlett et al. (2002) found 55 species of tabanids by using canopy traps set up at 20 locations in Rhode Island. There were 24 horse fly species including T. americanus Forster, the largest North American tabanid, but no T. calens in their statewide sampling. Thompson (1969b) reported 37 species from the Great Swamp National Wildlife Refuge (GSNWR) in Morris County, NJ, but T. calens was there in only small numbers in the upper Passaic River watershed. *Adjunct Professor, Biology, University of Vermont, Burlington, VT 05405; redmug34@ gmail.com. Manuscript Editor: Daniel Pavuk Northeastern Naturalist 138 J.V. Freeman 2017 Vol. 24, No. 2 Tabanus calens was present near the slow-flowing Passaic River 6 km to the south at the southern tip of Morris County in Millington (Freeman 1960). Tallamy et al. (1976) collected 50 species of tabanids, including T. calens, as they compared net vs. Malaise trap at 5 stations near Deer Lake, Boonton, NJ, in northern Morris County. Smith (1899) in his Insects of New Jersey, written when there were still an abundance of farms, horses, and pastures in the region, listed T. giganteus (T. calens now) from Caldwell in the lower Passaic River valley in Essex County. Pratt and Pratt (1986) collected 29 species in 5 genera—Stonemyia (2), Goniops (1), Chrysops (18), Hybomitra (7), and Tabanus (1)—near Laurel Lake, Jacksonville, Windham County, near Route 100 in southern Vermont. They used nets only and emphasized male tabanid visits to flowers. Knutson et al. (1954) collected selectively at state parks and wildlife refuges in Vermont and found 20 deer fly species out of 37 species known in New England then. There are difficulties in matching some of the species reported in Vermont by Johnson (1925) due to changes of species names and vagueness of locations reported. Nielsen (1990) prepared faunal lists of Tabanidae in Vermont, but T. calens was not listed. Nielsen’s listing included notations about locations, larval habitat, and range of dates in season for 31 species. Graham (2010) compiled a list of 72 species of tabanids in Vermont. Gojmerac and Devenport (1971), concerned about how deer flies affect people at Kegonsa State Park, collected 27 tabanid species in 2 seasons by nets and Manitoba traps at this location 33 km (20 mi) south of Madison, WI, adjacent to a managed wetland. The distribution map for T. calens shown by Pechuman et al. (1983) includes southern Wisconsin, but Gojmerac and Devenport did not report any T. calens collected by net or trap. At Lake Lansing Park-North in southern Michigan, Strickler Figure 1. A female Tabanus calens with a quarter dollar coin for scale. This specimen was among hundreds collected at Addison, VT, in a 2-tier box trap in August. Smoky wing, unicolorous abdomen, and large size help to identify this species. (Photograph © J.V. Freeman). Northeastern Naturalist Vol. 24, No. 2 J.V. Freeman 2017 139 and Walker (1993) in a study using nets and traps over 2 full seasons, collected no T. calens even though all the lower peninsula of Michigan is shown as within the Figure 2. The 2-tier box trap at Addison, VT. The author (left) and Mike Kennedy (right), who added the white PVC down pipe from the collecting bottle up at the high corner. The upper trap increases the effectiveness of this trap. As an improvement to this design, the clear plastic “cone” visible on the screen wire of the upper trap should be replaced with a tapered screen wire tube. The distances between the upper and lower traps and from the lower trap to the ground are each 0.6 m (2 ft). Photograph © D. Laramie. Northeastern Naturalist 140 J.V. Freeman 2017 Vol. 24, No. 2 range of T. calens (Pechuman et al. 1983). As in the Rhode Island study, at Lake Lansing they collected the even larger T. americanus. Roberts and Dicke (1958) reported T. calens in Adams and Sauk counties not far from the lower Wisconsin River. Philip (1952) showed why the earliest name for Tabanus giganteus Degeer 1776 should be T. calens L. 1758. Today’s accepted nomenclature follows Burger (1995) for names of North American Tabanidae. Johnson (1925) reported T. giganteus, the junior synonym, in Massachusetts and Connecticut but not among 19 tabanid species in Vermont. Blickle (1954) listed no T. calens among 71 species of tabanids in New Hampshire. For more than 50 years, box traps have been used to study and reduce populations of salt marsh greenhead horse flies (T. nigrovittatus group) (Jamnback and Wall 1959). Hansens et al. (1971) continued this research in New Jersey and Wall and Doane (1980) reviewed work on Cape Cod using hundreds of 0.6-m (2-ft) cubical box traps with the attractant Octenol (Hayes et al. 1993). Field Site Description We placed the trap in a 0.25-ha (0.5-acre) horse pasture above Mountain Road Extension in Addison, VT, and downhill from the house and barn. To the north was mixed woodland with Fraxinus americana L. (White Ash), Quercus rubra L. (Red Oak), (Ostrya virginiana (Mill.) K. Koch (Hophornbeam), Betula alleghaniensis Britton (Yellow Birch), (Carya ovata (Mill.) K.Koch) (Shagbark Hickory), Acer saccharum Marsh. (Sugar Maple), Tilia americana L. (Basswood), Fagus grandifolia Ehrh. (American Beech), and Pinus strobus L. (Eastern White Pine). Thompson (1996) refers to this type of woodland composition as mesic transition hardwood forest. Going uphill to the north and east is Snake Mountain and the state Wildlife Management Area sharing that name (Fig. 3). To the east was more woodland of similar composition and wet pasture nearby. Just west was a 0.2-ha (0.5-ac) pond with a 5-m2 stand of Typha latifolia L. (Broad Leaf Cattail). Water from the pond flows under the road and through a place with Matteuccia struthiopteris (L.) Todaro (Ostrich Fern) and Caltha palustris L. (Marsh Marigold) under Salix sp. (willow). More cattail grows in the ditch along the road. Open grassy fields farther to the west suggest conditions suitable as habitat for T. quinquevittatus as described by Teskey (1969). The trap location was 44°02'24"N, 73°16'41"W at an elevation of 108 m (354 ft). The elevation of Snake Mountain is 772 m (1288 ft). Methods We used a modified version of the 2-tier box trap described by French and Hagan (1995) constructed from 2 full sheets of plywood, each 12 mm (1/2 in) thick x 1.2 m (4 ft) wide x 2.43 m (8 ft) long. To our knowledge, this is the first use of this kind of rigid, durable, all-weather trap for this purpose in Vermont. The material was cut to provide for 2 rectangular sides (1.2 m x 0.56 m) and 2 high-corner sides as follows—on the left side, from the top down: 56 cm (22 in), 77 cm (30 in), 56 cm (22 in), and 56 cm (22 in); and on the right edge, from the top down: 56 cm (22 Northeastern Naturalist Vol. 24, No. 2 J.V. Freeman 2017 141 in), 56 cm (22 in), 77 cm (30 in), and 56 cm (22 in). The middle of the 3 cuts was angled. The use of black paint outside the top and bottom traps seemed appropriate to maximize polarized light reflection. Figure 2 shows the constructed 2-tier trap on location in Addison. The distances from the lower trap to the ground and between the upper and lower traps were each 0.6 m (2 ft). Figure 3. Study area near Mountain Road Extension south of Snake Mountain in Addison, VT. The circle marks the trap location north of the road. This location is east of Vt. Rte. 22A and 3.2 km (2 mi) southwest of the confluence of the muddy Lemon Fair River and Otter Creek in Weybridge. (Prepared by K. Partlow). Northeastern Naturalist 142 J.V. Freeman 2017 Vol. 24, No. 2 The uprights at each corner were pressure-treated “2 x 4” studs ripped in half lengthwise ~3.8 cm (1.5 in) square. An extension at the high corner of the upper trap allowed attaching the collecting bottle with 2 bands of electrical tape. Collecting bottles were 1.75-L (1.8-qt) inverted clear plastic juice containers with 3.8-cm (1.5-in) openings to allow for the removal of the catch from the trap. The aluminum mosquito screen across the top of each trap was stapled to a wooden strip rather than to the plywood itself. Near the high corner was an opening leading through an inverted plastic cone (Fig. 2, upper trap) made from the top portion of a juice bottle, and then up a wire-screen tube to the inverted juice bottle as the collecting container. A bundle of strips of a product called Hot Shot®, No-Pest® Strip (active ingredient: Dichlorvos [2, 2 divinyl diethyl phosphate]; Spectrum Group, St. Louis, MO) provided the lethal atmosphere. An adhesive-sealant product called Goop® (Eclectic Products, Pineville, LA) helped to close seams and cracks. An improvement would be to make a flared or tapered wire-screen tube that allowed the largest tabanids traction to walk up the tube with no plastic cone. Under the screen top and bottom were spherical black targets made from a 60-cm-diameter (24-in) exercise ball tightly covered with a shiny black plastic garbage bag as a visual attractant. We used the chemical attractant Octenol only in 2015 and 2016, in a form designed for mosquito traps (EPA Registration No. 72563-5) because the liquid-reagent form as used by Hayes et al. (1993) was not available for this purpose. The horse owner who participated in our study used a small hand net with an opening of about 570 cm2 (former badminton racket), about half the area of the usual aerial net (1100 cm2), for collecting while riding on or standing near the horse. We used the keys of Pechuman (1981), Teskey (1990), and Thomas (2011) to identify all specimens to species. Burger (1995) provided our current taxonomic catalog. We could evaluate the numbers in the catch in the upper and lower traps by inspection or, where appropriate, with the paired sample Wilcoxon signed rank test calculator (Strangroom 2017). Growing awareness of the utilization of linearly polarized light (PL) in nature has moved beyond an understanding of navigation and orientation of honey bees and ants through neuroanatomical studies to how invertebrates and vertebrates use PL for finding food, water, and mates. I used polarizing sunglasses in testing for PL coming from horse fly traps. PL is one of the visual signals emitted by black 2-tier box traps and available to host-seeking tabanids (Cronin et al. 2003). Results Horse flies, Tabanus and Hybomitra Comparing trap collections from 3 seasons with grand totals of horse flies, Tabanus and Hybomitra, there is an evident decline (10,576 to 9656 to 6174; Table 1). Combining the trap catch and hand-net removals of T. calens (670 in 2013, 346 in 2014, 241 in 2015, and 92 in 2016) still resulted in a decline over the study years. The trap collected 297 T. calens (287 in the upper trap, 10 in the lower trap; Table 1) mostly in August 2013. By trap only, there were 42 collected in 2014 (down 86%). Northeastern Naturalist Vol. 24, No. 2 J.V. Freeman 2017 143 Combined captures from the trap and netting declined from 670 in 2013 to 346 in 2014 (48% decrease). In 2015, this total capture was 64% lower than 2013 (670 to 241) and 48% below 2014 (346 to 241). Table 1 lists the species within each genus alphabetically for each of the 3 collecting seasons. In 2015, the trap collected 78 T. calens, 85% more than during 2014 (42). Although we collected fewer T. calens than some other horse flies, this species remained a focus in this first whole-season study of our 2-tier box trap in Vermont. The preference for the upper or lower trap for some species was evident. Together T. quinquevittatus (60%) and H. lasiophthalma (26%) made up 86% of all horse flies collected over the 3 seasons. Of 24 species of horse flies, there were 6 species that appeared in just 1 of the 3 years: T. atratus, H. affinis, H. cincta, H. illota, H. lurida, and H. trispila (Table 1). This box trap collected 446 T. calens in 4 years (297 + 42 + 78 + 29) and hand-netting by the cooperator both near the horse and while riding the horse produced 903 (373 + 304 + 163 + 63) more T. calens for a Table 1. A 3-year summary of Tabanus and Hybomitra species collected in a 2-tier box trap at the Mountain Road Extension site south if Snake Mountain in Addison, VT, with abundance and seasonal occurrence of species. See text for results of special netting and removals of T. calens. 2013 2014 2015 3-yr Seasonal Species Upper Lower Upper Lower Upper Lower total range T. atratus Fabr. 1 1 0 0 0 0 2 7/21–9/4 T. calens L 287 10 42 0 76 2 417 7/31–9/3 T. lineola Fabr. 137 177 86 267 150 156 973 5/24–8/15 T. marginalis Fabr. 0 1 4 28 6 9 48 6/1–8/22 T. novaescotiae Macq. 7 1 12 2 8 1 31 8/2–9/3 T. pumilus Macq. 151 88 251 250 46 26 812 6/27–8/16 T. quinquevittatus Wied. 5566 1289 3949 1473 2972 592 15,841 6/16–9/7 T. reinwardtii Wied 2 1 2 1 4 1 11 6/21–8/7 T. sackeni Fairch. 43 263 28 123 20 143 620 7/20–8/31 T. sagax O. S. 0 1 0 2 1 1 5 8/3–8/9 T. superjumentarius Whit. 0 0 2 0 1 1 4 6/22–7/28 Tabunus totals 6194 1832 4376 2146 3284 931 18764 Combined totals 8026 6522 4216 H. affinis (Kirby) 0 0 0 0 1 0 1 7/26–8/1 H. aurilimba (Stone) 7 0 19 26 18 2 72 6/15–8/9 H. cincta (Fabr.) 0 0 1 0 0 0 1 7/2 H. epistates (O. S.) 76 127 65 67 46 16 397 6/8–8/17 H. hinei (Johns) 0 0 6 6 3 1 16 5/3–5/9 H. illota (O. S.) 0 0 0 0 3 0 3 6/14–6/20 H. lasiophthalma (Macq.) 1132 1180 1613 1240 1194 597 6956 5/25–7/18 H. lurida (Fallen) 0 0 0 0 12 14 26 5/10–5/16 H. microcephala (O. S.) 3 3 1 3 0 0 10 7/18–8/2 H. pechumani T. & T. 0 2 2 3 11 0 18 6/21–8/2 H. sodalis (Will.) 15 5 44 20 18 2 104 7/5–8/7 H. trepida (McD.) 0 0 15 3 12 1 31 6/29–8/16 H. trispila (Wied.) 0 0 0 0 5 2 7 7/5–7/11 Hybomitra totals 1233 1317 1766 1368 1323 635 7642 Combined totals 2550 3134 1958 Northeastern Naturalist 144 J.V. Freeman 2017 Vol. 24, No. 2 total removal of 1349 female T. calens. Trap collecting continued in 2016, but only totals of T. calens for that year are included here. Collections of T. sackeni contrast with those of T. calens in their consistent preferences for the lower trap. Both species showed crepuscular host seeking. In sheer bulk, the collections of H. lasiophthalma in May through June into July and T. quinquevittatus from mid June through August dominated the tabanid catch. The trap preference for the upper trap by T. quinquevittatus was significant as denoted by the Wilcoxon paired sample signed rank test (P < 0.0001, n = 26). There was no significant preference for H. lasiophthalma (P = 0.085, n = 19). Mostly black muscomorph flies made up most of the rest of the trap catch. Comparing the first-year (2013) total catch (8011 flies) with the third-year (2015) total (4183 flies) for the 5 most abundant Tabanus species (T. calens, T.lineola, T. pumilus, T. quinquevittatus, and T. sackeni; Table 1) showed a 47% reduction. T. calens (297 to 78) was down 73% for the same periods. The netting initiative by the horse owner/cooperator removed 903 T. calens females or twice the catch of the 2-tier trap (446) in 4 seasons. The trap works all the time. Netting rates ranged from 4 to 74 per hour and averaged 31 per hour (n = 13). Tabanus quinquevittatus was down 48%, H. lasiophthalma down 22%, T. sackeni down 46%, and T. lineola down by only 2% in 2013 compared to 2015. This 3-year record of tabanid species collected at Addison perhaps indicates that trapping can reduce populations of horse flies locally, though it is possible other factors were responsible for the decrease in number of horse flies we caught over those years. For example, it should be noted that 2015 started cooler and much rainier and overall had less hot weather than 2013, which might have affected fly populations and/or trapping success in the area. Deer flies, Chrysops The 6 most abundant Chrysops species overall by trap (C. univittatus, C, vittatus, C. callidus, C. geminatus. C. macquarti, and C. aberrans; Table 2) included C. callidus, which is difficult to catch by net. Chrysops callidus does not bother people but was numerous in the upper box trap. The other deer flies can be bothersome to people even in small numbers. The 20 species of deer flies in Table 2 include 6 that appeared in just one of the 3 years: C. lateralis, C. mitis, C. moechus, C. carbonarius, C. montanus, and C. cuclux. Although the total deer fly catch in 2014 was up by 36% over 2013, comparing 2015 to 2013 (521 in 2015 vs. 691 in 2013) showed a reduction of 24% (Table 2). A major part of the increase in 2014 was due to C. callidus, but only 3 of the first 10 species decreased (C. vittatus, C. sackeni, and C. cincticornis) from 2013 to 2014. People in Addison commented on increased deer fly activity in 2014, and our catch results confirm their perception. The design of this 2-tier box trap aimed to exploit the around-the-head behavior of deer flies affecting people. This trap enabled detection of single specimens of certain deer fly species as well as more-abundant species, though our intended emphasis was on T. calens and other horse flies. Northeastern Naturalist Vol. 24, No. 2 J.V. Freeman 2017 145 Discussion For horse flies, Tabanus and Hybomitra, our annual capture totals started out (2013) at 10,576. In 2015, they totaled 6162, a 41% reduction from the 2013 total. In these 3 years, hand-netting resulted in 1.27 to 5.76 times the number of T. calens that we collected in the trap based on daily trap collections and timed nettingsessions that were carefully bagged separately. Net-collecting sessions ranged from 30 to 132 min, and capture rates averaged 31 T. calens per hour. The 2-tier trap also successfully collected T. calens, and a very good day might yield 22 T. calens (158 in a week) in the trap. At Addison, the total deer fly collection in 2014 was 927, up 34% over 2013 (Table 2) driven by C. callidus (+197) and 5 other species: C. univittatus (+31), C. geminatus (+40), C. indus (+12), C. calvus (+24) and C. aberrans (+37). Local perception in 2014 was that “deer flies are up”. There was a period of rainy weather from mid-June to mid-July 2015 with below-normal temperatures and about double the normal rainfall during this time, as displayed graphically by Schultz et al. (2015) from NOAA-USWS data at Burlington (BTV). Such wet and cool weather might be expected to reduce deer and horse fly activity. However, comparing totals in this period for deer flies and horse flies in each of the years of trapping, there was no clear pattern except lower numbers in 2015 than in 2014. Perhaps the use Table 2. Deer flies, Chrysops, collected by a 2-tier box trap at the Mountain Road Extension site south of Snake Mountain in Addison, VT, in 2013 to 2015, arranged in descending order based on first-year (2013) totals. 2013 2014 2015 3-yr Seasonal Species Upper Lower Upper Lower Upper Lower total range C. univittatus Macq. 228 25 252 32 124 12 673 6/22–8/30 C. vittatus Wied. 170 2 111 2 103 4 392 6/22–8/20 C. sackeni Hine 71 3 0 0 9 0 83 6/27–7/20 C. geminatus Wied. 56 2 96 2 67 0 223 6/29–8/3 C. aberrans Philip 35 4 66 10 58 0 173 6/29–8/16 C. macquarti Philip 20 2 54 2 13 1 92 6/15–8/23 C. callidus O.S. 17 2 204 12 58 3 296 5/25–7/19 C. calvus P. & T. 16 1 40 1 22 0 80 5/21–6/29 C. cincticornis Walk. 15 1 9 2 3 0 30 6/8–6/28 C. indus O.S. 7 2 18 3 9 1 40 5/25–7/19 C. niger Macq. 6 0 0 0 4 0 10 5/24–7/3 C. lateralis Wied. 2 0 0 0 0 0 2 7/14–7/27 C. frigidus O.S. 1 0 0 2 1 1 5 6/22–7/31 C. mitis O.S. 1 0 0 0 0 0 1 5/18–5/25 C. moechus O.S. 1 0 0 0 0 0 1 7/14–7/20 C. carbonarius Walk. 1 0 0 0 0 0 1 5/28 C. montanus O.S. 0 0 3 1 0 0 4 6/29–7/19 C. cuclux Whit. 0 0 0 0 1 0 1 5/24–5/30 C. ater Macq. 0 0 5 0 26 0 31 5/10–6/7 C. venus Philip 0 0 1 0 1 0 2 6/15–6/21 Totals 647 44 859 69 499 22 2140 Combined totals 691 928 521 Northeastern Naturalist 146 J.V. Freeman 2017 Vol. 24, No. 2 of Octenol in the trap was making a difference. The role of the 2-tier box trap in reducing deer and horse fly populations remains unclear, since we did not conduct an experiment with and without trapping, but our device certainly proved useful in sampling the local tabanid populations. Such traps exploit horse flies’ use of polarized light to seek a blood meal (Horvath et al. 2008). It is important to note that G.A. Vale (Chicot State Park, LA, 1996 pers. comm.) mentioned to the author that our traps might well be collecting only a small part of the local tabanid pest population. Further reinforcing this point, Vale and Hargrove (1979) calculated trap efficiencies as being only about 10 percent. Smith et al. (1970) listed 31 tabanid species over 3 years at Algonquin Park, ON, Canada, and included host preference and habitat preferences of some species. Host preferences for H. lasiophthalma and H. epistates were clearly for cervids such as Odocoileus virginianus (Zimmermann) (White-tailed Deer) or Alces alces (L. (Moose) and not for humans. Thompson (1969a) collected C. callidus in a Manitoba trap (243) and by net (only 12) in New Jersey. It remains for others to discover the host preference of C. callidus, as it did not appear at Algonquin Park. The species does not bother people and prefered the upper trap in our study. At Addison, the annual trap catch totals of H. epistates went down (203 to 132 to 62). On the other hand, numbers of H. lasiophthalma (2312 to 2853 to 1791) went up in 2014, and this species certainly affects horses. Tabanus calens in Addison is well known as a pest to gardeners in August. Range extension seems likely for 2 horse fly species. Nielsen (1990) did not list T. calens or H. hinei in the faunal list in his Horse Flies of Vermont. Wright et al. (1986) used 150 km (90 mi) as their criterion for range extensions, and 16 of 61 species they caught in Oklahoma were collected beyond what was previously known by that distance or more. Addison, VT, is about 225 km (135 mi) north of Albany, NY. Pechuman (1981) mapped collections of T. calens near Albany, others in the lower Hudson Valley, and a small number in the Finger Lakes area. Matthysse et al. (1974), near Ithaca, NY, used a group of canopy traps that yielded thousands of T. quinquevittatus and H. lasiophthalma, similar to the present study, but just 3 T. calens in the second season, even with the use of dry ice as a CO2 attractant. The range of T. calens appears to have extended into Vermont in Orwell (J.V. Freeman, unpubl. data), Bridport (Graham 2010), and, from our study, Addison. Our trap detected a total of 16 specimens of H. hinei in this study (12 in 2014, 4 in 2015) but none initially in 2013. Pechuman et al. (1983) showed the distribution of H. hinei to be scattered and disjunct in the Midwest, with a coastal occurrence from South Carolina to New Hampshire. Modern planetary research evaluating reflected light as albedo images from Saturn’s moons, planets, or asteroids such as Burleigh et al. (2010) cited Umov’s (1905) work and led to the realization that the amount of polarized light coming back from an object is the inverse of its albedo (whiteness) and increases with increasing phase angle or the angle between the incident and the reflected light. Turning a polarizing filter 90 degrees can help detect polarization. Können (1985) described and tested many common examples of polarized light in nature Northeastern Naturalist Vol. 24, No. 2 J.V. Freeman 2017 147 and their importance. Human color vision mainly uses wavelength, intensity, contrast, detail, and motion. Smith and Butler (1991) found, however, that the microstructure of the compound eyes of 17 species of tabanids had the capability of polarization vision. Horvath et al. (2008) demonstrated with 27 species that tabanids of both sexes used polarization vision in making choices looking forward and down. Shiny black plastic was most attractive. Freeman (2011) illustrated the difference between images of a canopy trap (Catts 1970) with shiny black plastic by turning a polarizer 90 degrees; this polarization might catch the attention of an approaching tabanid fly. A review of polarization vision by Cronin et al. (2003) explains the process down to microstructure in the microvilli of the rhabdom in the dipteran photoreceptor. Pechuman (1981) and others have used the 4-sided pyramidal canopy trap designed by Catts (1970) with its lower part made of shiny black plastic leading up to clear plastic of the upper part. Our trap design was also designed to make use of polarized light to attract tabanids. This 2-tier trap by design emits the visual signal of a large black non-plant object with the sensory signal of warmth when the absorbed sunlight becomes heat. It emits a polarized light image available to photoreceptors with polarization vision. With an added attractant like Octenol, there is the olfactory signal. Each of these signals mimics the presence of a large, warm, dark-haired mammal as a source of blood. Reflection of daylight from dark hair is partly polarized as well. The trap catch in 2015 was lower than 2014 for the majority of horse fly species (17 down, 7 up) and deer fly species (15 down, 5 up) even with our using Octenol in 2015 but not in 2014. Overall the trend was downward. In contrast, from 2014 to 2015, T. calens went from 42 to 78 in the trap, T. sackeni from 151 to 163, H. trispila from 0 to 7, and C. ater from 5 to 26. In 2016, the selective trap catch was 29 and the net catch was 63 for T. calens. Taking into account the total captures (trap + net) in 2015 compared to the total in 2016 (241 to 92) provides evidence of a further 61% reduction of this local population. Trap-only captures (78 to 29) represented a 62% reduction from 2015 to 2016. This reduction relates to the third objective of this study. If these (anautogenous?) trapped female T. calens had not already laid a batch of eggs as they sought a blood meal approaching a trap mimic of a large mammal, the trapping effort itself might help to explain the reduction in population of T. calens seen here. The trap catch of T. calens in 2014 (42) was 85% less than in 2013 (297). It was up 87% in 2015 (78) compared to 2014 but still far below 2013. The hand-netting occurred about 60 m (200 ft) away from the trap and might have influenced the trap catch. The cooperator expressed firm belief in the benefit her horse experienced compared to previous years with no trapping or hand-netting. The value and convenience of a tame bait animal is that it remains calm even with the net in use and allows reliable and safe access to a host animal. Two trap modifications from French and Hagan's (1995) design were the clear plastic exit near the high corner of each trap (Fig. 2) and the extension tube coming down from the upper trap. The down tube was securely attached to the cap of the upper juice bottle and allowed emptying without a ladder. In addition, there should Northeastern Naturalist 148 J.V. Freeman 2017 Vol. 24, No. 2 be a continuous tapered wire-screen tube to allow large tabanids to both walk and fly up into the inverted juice bottle. French and Hagan (1995) used flat black for the outside of the lower trap but offwhite outside and inside for the upper trap without explanation as to the potential advantage of that color arrangement. Vale and Hargrove (1979) made estimates of the efficiency of their 1-m tall tsetse fly trap that also caught tabanids. They had a thin-wire electric grid that could kill flies that visited but did not enter this trap and there were also observers behind one-way glass. At Addison, the cause for aggregation in trapping was, as mentioned above, the warmth and polarized light emitted by the structure of the trap itself and, in 2015 and 2016, the use of Octenol as an olefactory signal. The lure used in the hand-netting portion of our study was a large mammalian host. Our study demonstrated several benefits to trapping and netting flies. We were able to gain a sense of abundance and seasonal occurrence of each species, even less common ones for which we trapped only single specimens such as C. moechus or C. carbonarius or C. cuclux. For most species, we were also able to observe the preference for upper or lower trap specific to this trap, which can indicate if a fly is more likely to attempt to feed on the head of a person or animal or the lower portions of the body. To what extent trapping and netting are effective at reducing the local populations of horse flies should be the focus of future r esearch. A net and Malaise trap study at Lansing Lake Park-North (Strickler and Walker 1993) clearly showed the greater effectiveness of capturing C. vittatus and C. univittatus by net compared to trapping, but they collected just a single T. americanus and no T. calens and in 1 year of effort, despite the fact that all of the Michigan lower peninsula is considered part of the range of T. calens (Pechuman et al. 1983). Thus, it seems the search needs to continue for the combination of habitat components that might successfully predict the presence and abundance of T. calens similar to what we found at Addison, VT, with the 2-tier box trap. This single 2-tier box trap has provided another tool for exploring populations of Tabanus calens and other tabanids in Vermont. We placed voucher specimens in the Zadock Thompson Natural History Museum insect collection at the University of Vermont (UVM) in Burlington, VT. These new additions will improve the representation in that collection. We now have a set of specimens of species collected at this 1 location, some idea of relative abundance, and some indication that this trap might be effective in reducing the population of pest species. While daily collections can increase the precision, weekly collections allowed an accurate broad representation along with indications of abundance and preference for upper or lower traps through a whole season. As a follow-up to this study, more general collecting is needed in southwest Vermont in August and near sunset. While the apparent reduction in the pest population over our study period, whether the result of our trapping efforts and/or other factors, is noteworthy, the reality is that just one of these large horse flies can harass a person, a dog, or a horse. Similarly with deer flies, while our third-year capture total Northeastern Naturalist Vol. 24, No. 2 J.V. Freeman 2017 149 at 521 was 24% below the first-year total of 691, single deer flies or small numbers can find, follow, and cause distress to a person. French and Hagan (1995) did not determine any acceptable level of population control near the almost endless supply of deer flies from a tidal salt marsh adjacent to St. Simon’s Island, GA. They operated 6 pairs of traps that allowed experimental comparisons. However, apparently the removal of so many deer flies was satisfying to golfers in t he area. Acknowledgments I wish to thank Deb Laramie and Mike Kennedy as interested cooperators; Deb got the frozen tabanid flies from the 2-tier box trap in Addison by courier to Rutland each week. She took time off specifically to expand the net collections in August. I thank S.R. Alm, S. Schutz, D.W. Tallamy, and A.W. Thomas, earlier reviewers of this paper, for their suggestions, and most recently J.F. Blount for his careful editing. Clarifying translation of the Umov paper relating to polarized light by Charles Ferguson and Christian Heller is much appreciated. I thank John F. Burger at the University of New Hampshire for checking species identifications. For help along the way in connecting people, I want to recognize Trish Hanson of the Vermont Department of Forests, Parks, and Recreation. Kim Partlow, also with the Vermont Department of Forests, Parks, and Recreation, helped create Figure 3. I thank Carol Freeman for storing many bags of well-wrapped frozen flytrap samples in her freezer. Literature Cited Bartlett, K., S.R. Alm, R. Lebrun, and H. Ginsburg. 2002. The horse and deer flies (Diptera: Tabanidae) of Rhode Island. Annals of Entomological Society of America 95(5):547–551. Blickle, R.L. 1954. Tabanidae of New Hampshire. Psyche 61(2):74–80. Burger, J.F. 1995. Catalog of Tabanidae (Diptera) of North America North of Mexico. Contributions on Entomology, International. Vol.1, No. 1. 100 pp. Burleigh, K.J., P. Helfenstein, B. Carcich, J. Veverka, and others. 2010. Linear polarization and albedo reconnaissance for regolith texture on Saturn’s moon Iapetus. 41st Lunar and Planetary Conference. Available online at http://www.lpi.usra.edu/meetings/lpsc2010/ pdf/2607.pdf. Catts, E.P. 1970. A canopy trap for collecting Tabanidae. Mosquito News 30(3):472–474. Cronin, T.W., N. Shashar, R.L. Caldwell, J. Marshall, A.G. Cheroske, and T.-H. Chiou. 2003. Polarization vision and its role in biological signaling. Integrative and Comparative Biology 43(4):549–558. Freeman, J.V. 1960. Observations on the seasonal distribution, biology, and behavior of some species of horse flies and deer flies (Diptera: Tabanidae) in Morris County, New Jersey. M.Sc. Thesis. Department of Entomology, Rutgers-the State University of New Jersey, New Brunswick, NJ. 101 pp. Freeman, J.V. 2011. Tabanidae and other Diptera on Camel’s Hump, Vermont: Ecological observations. ZooKeys 147:559–576. French, F.E., and D.V Hagan. 1995. Two-tier box-trap catches Chrysops atlanticus and C. fuliginosus (Diptera: Tabanidae) on Cape Cod, Massachusetts. Journal of Medical Entomology 32(2):197–200. Gojmerac, W.L., and E.C. Devenport. 1971. Tabanidae (Dipters) of Kegonsa State Park, Wisconsin: Distribution and seasonal occurrence as determined by trapping and netting. Mosquito News 31(4):572–575. Northeastern Naturalist 150 J.V. Freeman 2017 Vol. 24, No. 2 Graham, A.C. 2010. Unpublished compilation of lists of species by Nielsen, Graham, Pratt, Pratt, and Freeman. Vermont State Entomologist. Available from the author or from alan.c.graham@gmail.com. Hansens, E.J., E.M. Bosler, and J.W. Robinson. 1971. Use of traps for study and control of salt marsh greenhead flies. Journal of Economic Entomology 64:14 81–1486. Hayes, R.O., O.W. Doane Jr., G. Sakolsky, and S. Berrick. 1993. Evaluation of attractants in traps for greenhead fly (Diptera: Tabanidae) collections on a Cape Cod, Massachusetts, salt marsh. Journal of the American Mosquito Control Association 9(4):436–440. Horvath, G., J. Majer, L. Horvath, I. Szivak, and G. Kriska. 2008. Ventral polarization vision in tabanids: Horseflies and deer flies (Diptera: Tabanidae) are attracted to horizontally polarized light. Naturwissenschaften 95:1093–1100. Jamnback, H. and W. J. Wall 1959. The common salt marsh Tabanidae of Long Island, New York. N.Y State Museum Bulletin 375:1–177. Johnson, C.W. 1925. List of Diptera of New England. Fauna of New England 15. Occasional Papers of the Boston Society of Natural History VII. Boston, MA. 326 pp. Knutson, H.E., I. Coher, F.R. Lisciotto, and J.C. Kuschke. 1954. Notes on Chrysops, or deer flies (Tabanidae, Diptera) of New England. Mosquito News 14(4):205–21 1. Können, G.P. 1985. Polarized Light in Nature. Cambridge University Press, London, UK. Matthysse, J.G., D.E. Mock, and H.R. Netherton. 1974. Seasonal flight period and relative abundance of Tabanidae (Diptera) near Ithaca, New York. Annals of the Entomological Society of America 67(2):158–166. Nielsen, G.R. 1990. Horse flies: Diptera: Tabanidae. Vermont Faunal List VFL-DI-2 UVM Extension Service, Plant and Soil Science Department, Burlington, VT. 10 pp. Pechuman, L.L. 1981. The horse flies of New York (Diptera, Tabanidae). Cornell Univeristy Agricultural Experiment Station, Ithaca, NY. 72 pp. Pechuman, L.L., D.W. Webb, and H.J. Teskey. 1983. The Diptera, or true flies, of Illinois. I. Tabanidae. Illinois Natural History Survey Bulletin Vol. 33, No. 1. 121 pp. Philip, C.B. 1952. The Linnaean and Degeerian species of American Tabanidae. Annals of Entomological Society of America 45(2):310–314. Pratt, G.K., and H.D. Pratt. 1986. Notes on deer flies and horse flies (Diptera: Tabanidae) from southern Vermont. Journal of American Mosquito Control Association 2(3):365–367 Roberts, R.L., and R.J. Dicke. 1958. Wisconsin Tabanidae. Transactions of the Wisconsin Academy of Sciences, Arts, and Letters 47:23–42. Schultz, B., T. Hanson, S. Wilmot, J. Halman, K. Decker, and T. Greaves. 2015. Forest insect and disease conditions in Vermont. Agency of Natural Resources, Department of Forests, Parks and Recreation, Montpelier, VT. Smith, C., and J.F. Butler. 1991. Ultrastructure of the Tabanidae compound eye: Unusual features for Diptera. Journal of Insect Physiology 37(4):287–296. Smith, J.B. 1899. Catalog of insects of New Jersey, Tabanidae. Pp 639–642, In Insects of New Jersey, Twenty Seventh Annual Report of the State Board of Agriculture. MacCrellish and Quigley, State Printers, Trenton, NJ. Smith, S.M., D.M. Davies, and V. Golini. 1970. A contribution to the bionomics of the Tabanidae (Diptera) of Algonquin Park, Ontario: Seasonal distribution, habitat preferences, and biting records. Canadian Entomologist 102:1461–1473. Stangroom, J. 2017. Wilcoxon signed-rank test calculator. Available online at http://www. socscistatistic.com/tests/signedranks. Accessed 10 May 2017. Strickler, J.D., and E.D. Walker. 1993. Seasonal abundance and species diversity of adult Tabanidae (Diptera) at Lake Lansing Park-North, Michigan. Great Lakes Entomologist 26(2):107–112. Northeastern Naturalist Vol. 24, No. 2 J.V. Freeman 2017 151 Tallamy, D.W., E.J. Hansens, and R.F. Denno. 1976. A comparison of Malaise trapping and aerial netting for sampling of a horsefly and deerfly community. Environmental Entomology 5:788–792. Taylor, P.D., and S.M. Smith. 1989. Activities and physiological states of male and female Tabanus sackeni. Medical and Veterinary Entomology 3:203–212. Teskey, H.J. 1969. Larvae and pupae of some Eastern North American Tabanidae (Diptera). Entomological Society of Canada Memoir 63. Winnipeg, MB, Canada. 147 pp Teskey, H.J. 1990. Horse Flies and Deer Flies of Canada and Alaska Diptera: Tabanidae. The Insects and Arachnids of Canada Part 16. Publ. 1838. Agriculture Canada, Ottawa, ON, Canada. 381 pp. Thomas, A.W. 2011. Canadian Tabanidae east of the Rocky Mountains 2: A photographic key to the genera and species of Tabaninae (Diptera: Tabanidae) Canadian Journal of Arthropod Identification No.13. Available online at http://www.biology.ualberta.ca/bsc/ ejournal/t_13/t_13.html, doi: 10.3752/cjai2011.13. Accessed 9 May 2017. Thompson, E. 1996. Natural communities of Vermont uplands and wetlands. Nongame and Natural Heritage Program, Vermont Department of Fish and Wildlife, Waterbury, VT. 34 pp. Thompson, P. H. 1969a. Collecting methods for Tabanidae (Diptera). Annals of the Entomological Society of America 62(1):50-57. Thompson, P.H. 1969b. Abundance and seasonal distribution of the Tabanidae of the Great Swamp, New Jersey. Part II. The second year. Annals of the Entomological Society of America 62(6):1429–1433. Umov, N.A. 1905. Chromatische depolarization durch Lichterstreuung. Physik Zeitschrift 6:674–676 Vale, G.A., and J.W. Hargrove. 1979. A method of studying the efficiency of traps for tsetse flies (Diptera: Glossinidae) and other insects. Bulletin of Entomological Research 69:183–193. Wall, W.A., and J.O.W. Doane. 1980. Large-scale use of box traps to study and control salt marsh greenhead flies (Diptera: Tabanidae) on Cape Cod, Massachusetts. Environmental Entomology 9:371–375. Wright, R.E., R.K. Whittle, and L.L. Pechuman. 1986. Range extensions, new state records, and annotated checklist for Oklahoma Tabanidae (Diptera). Journal of Kansas Entomological Society 59(2):235–245.