2010 NORTHEASTERN NATURALIST 17(2):175–188 Lotic Odonata of the Connecticut River in New Hampshire and Vermont Pamela D. Hunt1,*, Michael Blust2, and Fred Morrison3 Abstract - Several riverine species in the insect order Odonata (dragonflies and damselflies) are recognized as being of conservation concern in the Northeast. Along the Connecticut River, most data on these species have come from the southern portion of the river that passes through Connecticut and Massachusetts, while the northern portion has been poorly sampled until recently. In this paper, we summarize recent surveys along the Vermont–New Hampshire stretch of the river and place these in the context of known distributional data for the river as a whole. Our focus is on species typical of large rivers, with a particular focus on members of the family Gomphidae (clubtails). Also included is information on the first Vermont or New Hampshire records of three species—Enallagma antennatum, Enallagma durum, and Stylurus amnicola—and the first upper river records for several other species. Introduction The Connecticut River runs 650 km from its source near the Canadian border in New Hampshire to its mouth at Essex, CT on Long Island Sound. In the northern portion between Vermont and New Hampshire, the river passes through many stages: starting as a small headwater stream, passing through several impoundments created by hydroelectric dams interspersed with free-flowing sections, and eventually entering Massachusetts as a large slow-moving, but still impounded, river with a broad floodplain. Concerns over water pollution, altered hydrology, and potentially damaging shoreline development have driven conservation focus on the river in Vermont and New Hampshire (Francis and Mulligan 1997, New Hampshire Fish and Game Department 2005, Vermont Fish and Wildlife Department 2005), including efforts to inventory its aquatic biota. Many species of Odonata are specialists on large rivers such as the Connecticut River. These include several listed as threatened or endangered in Connecticut (Connecticut Department of Environmental Protection 2004, Wagner and Thomas 1999) and Massachusetts (Massachusetts Division of Fisheries and Wildlife 2005). Although neither New Hampshire nor Vermont has any state-listed riverine Odonata, many are recognized as species of conservation concern (Hunt 2006a; M. Blust and B. Pfeiffer, unpubl. data). Despite increased knowledge of Odonata distributions in New England, there was a noticeable absence of data from the Connecticut River in Vermont and 1Audubon Society of New Hampshire, 3 Silk Farm Road, Concord, NH 03301. 2Green Mountain College, One College Circle, Poultney, VT 05764. 3A Natural Focus, 56 Montague Road, Westhampton, MA 01027. *Corresponding author - phunt@ nhaudubon.org. 176 Northeastern Naturalist Vol. 17, No. 2 New Hampshire. Of the 33 species considered typical of rivers and included in this paper, White and Morse (1973) list only 16 records of 13 species from New Hampshire towns along the Connecticut River, and perhaps only three of these records are positively associated with the river. Current efforts to compile historic and museum records for Vermont (M. Blust, unpubl. data) have yet to turn up any records of river species prior to 1975. We conducted surveys of riverine Odonata along the Connecticut River in New Hampshire and Vermont between 2005 and 2007, focusing entirely on the river’s main stem, and supplement the results of these surveys with historic information where available. Where appropriate, the conservation status of species of regional concern is discussed. Study Area and Methods We divided the Connecticut River of Vermont and New Hampshire into six segments based largely on the extent the river has been altered by dams (Fig. 1). For consistency, all locations are referenced to towns on the New Hampshire side (see DeLorme 2001 for more details). The Southern segment is dominated by the impoundments of the Vernon and Bellows Falls dams, and extends 94 km from the Massachusetts border to Claremont. This stretch of the river is characterized by numerous flooded backwaters and river mouths and varies in width from 150 to 250 m. Immediately to the north, the Macrosite segment extends north to the mouth of the Ompompanoosuc River in Hanover, and has been identified as an area of high biological diversity (“Connecticut River Rapids Macrosite”; Francis and Mulligan 1997). The Macrosite currently includes 34 km of free-flowing river and 12 km of impoundment above the Wilder Dam. The Central segment continues north from the Ompompanoosuc for 73 km to the Dodge Falls Dam in Bath, and for much of this distance is impounded by the Wilder Dam. The Reservoir segment consists largely of the impoundments of four dams (Dodge Falls, McIndoe Falls, Comerford, and Moore), and stretches for 47 km, including an area known as “Fifteen-Mile Falls” prior to dam construction. The Northern segment begins at the head of the Moore Reservoir in Dalton, and continues unobstructed for 52 km to the mouth of the Nulhegan River in North Stratford. From North Stratford, the Headwaters segment continues for 106 km to the source of the Connecticut River in Pittsburg. This portion of the river more closely resembles a tributary stream (NH Fish and Game Department 2005) and is less likely to host Odonata typical of larger rivers. The Headwaters segment has not received any recent survey effort, and due to a general lack of specific locality data for older records, it will not be considered further in this paper. During the summers of 2005 and 2006, we conducted regular surveys at 13 sites located within five river segments (Table 1). At most sites, surveys were conducted multiple times during the peak flight season (June–August) of riverine Odonata, although extremely high water levels in the late spring of 2006 precluded early June sampling at several sites. With the exception of 2010 P.D. Hunt, M. Blust, and F. Morrison 177 Figure 1.The Connecticut River in New Hampshire and Vermont, showing study sites and the approximate boundaries of the river segments discussed in the text. Lengths of segments are as follows: Headwaters (106 km), Northern (52 km), Reservoir (47 km), Central (73 km), Macrosite (46 km), and Southern (94 km). 178 Northeastern Naturalist Vol. 17, No. 2 larval surveys in the Northern segment (Blust 2006), the majority of records reported here are from systematic searches for exuviae in predetermined sections of river banks (e.g., Hunt 2006b, McLain et al. 2006). Exuviae are the shed skins left behind when odonates emerge from the water and transform into flying adults. They are frequently identifiable to species and provide a means of assessing species presence and relative abundance even when adults are not present or difficult to capture. Adult odonates were also recorded during survey visits to all sites. We also compiled additional records from several other sources. New Hampshire records prior to 1973 were obtained from White and Morse (1973). Additional records from 1973 to 2007 were obtained by examining specimens in the collections at the Universities of Vermont and New Hampshire and by querying other Odonata specialists in the Northeast. Specific records attributable to authors, but outside of the surveys listed in Table 1, are indicated by initials. Unless otherwise noted, specimens are retained in the authors’ personal collections. In the summaries below, we use four categories to describe relative abundance. These terms apply to both the numbers of individuals encountered (including exuviae) and the frequency of occurrence, as follows: abundant (found on all visits during its flight period, reserved for species that regularly outnumber all other taxa), common (easily found within its flight period, all records represent multiple individuals), uncommon (may not be detected on all visits, usually only a few individuals at a time), and rare (only 1 or 2 records for a site in a season). Table 1. Sites visited by the authors and mentioned in the text. These are listed from north to south and referenced to the appropriate river segment. For more details, see the appropriate reports (Blust 2006, Hunt 2006b, McLain et al. 2006). Segment/site name Latitude Longitude Surveyor Dates North Maidstone Bends 44.5694 71.5491 Blust May 30, June 21, July 11–12, Aug 9, Oct 16 (2006) Reservoir McIndoe Falls 44.2817 72.0596 Hunt July 23, 2006 Central Woodsville 44.1531 72.0411 Hunt June 30, July 23 (2006) Bedell Bridge 44.0454 72.0741 Hunt June 30, July 23 (2006) Macrosite Wilder Dam 43.6758 72.3018 Hunt July 1, 8, 30; Aug 13 (2006) West Lebanon 43.6361 72.3258 Hunt July 1, 8; Aug 13 (2006) Plainfield/Cornish 43.5272 72.3906 Hunt July 7, 30; Aug 13 (2006) South N Charlestown 43.2600 72.4288 Hunt June 15; July 7, 24; Aug 14 (2006) Herrick’s Cove 43.1826 72.4433 Hunt June 15, July 24, Aug 14 (2006) North Walpole 43.1439 72.4540 Hunt June 15; July 7, 24; Aug 14 (2006) Brattleboro/Chesterfield 42.8770 72.5539 Hunt June 11, 2007 Above Vernon Dam 42.8175 72.5415 Hunt July 21, 2005 Below Vernon Dam 42.7664 72.5068 Morrison June 10, 14; 22; July 5–6 (2005) 2010 P.D. Hunt, M. Blust, and F. Morrison 179 Results and Discussion We found 33 species of riverine Odonata in our study area on the Connecticut River in Vermont and New Hampshire (Table 2). Five of these species are more typical of small streams and are not discussed further. A summary of the distribution of the 28 species of riverine Odonata from our study area in Vermont and New Hampshire follows. Family Coenagrionidae (pond damsels) Argia apicalis (Say) (Blue-fronted Dancer). The only record for the region is of a male collected in Vernon, VT on 23 July 2005 (MB). The species is widespread but local along the river in Massachusetts (Nikula et al. 2003), and is thought to be expanding its range in the Northeast (D. Wagner, University of Connecticut, Storrs, CT, pers. comm.). Argia moesta (Hagen) (Powdered Dancer). This species is widespread on the Connecticut River south of Lebanon, NH. The only record north of Lebanon is a single teneral female seen in Haverhill, NH on 23 July 2006, suggesting that the species reaches its northern range limit along the river somewhere in the Central segment. Enallagma antennatum (Say) (Rainbow Bluet). This species was not recorded in New Hampshire until 8 July 2006, when several adults were observed immediately above Wilder Dam. Oviposition was observed at this site on 30 July. The species has been expanding its range eastward. It was first found in Vermont in 1995, and is now known from 5 counties in that state. A voucher specimen has been deposited at the University of New Hampshire. Enallagma carunculatum Morse (Tule Bluet). Prior to 2005, the only record of this species for the Connecticut River was from Lebanon, NH in 1972, although it was known from small ponds in Hartford, VT and Hanover, NH. In 2005–2006, it was recorded from Hinsdale (M. Blust), Lebanon (two sites, P. Hunt), and McIndoe Falls, NH and was relatively common at the latter location. Based upon these data, the species may be widely distributed, albeit local, in the study area. The Connecticut River is near the eastern edge of the species’ range in southern New England, although it does occur farther east in northern Maine and into the Maritime Provinces of Canada (Donnelly 2004b). It is listed as special concern in Massachusetts, although it may be more common than previously believed. Enallagma civile (Hagen) (Familiar Bluet). Although this species is not restricted solely to riverine habitats, it is frequently found on sparsely vegetated stretches of the Southern and Macrosite segments of the Connecticut River. Enallagma durum (Hagen) (Big Bluet). This species was not recorded on the Connecticut River in Vermont and New Hampshire until 21 July 2005, when several individuals were found on the New Hampshire side of the Vernon Dam impoundment in Hinsdale. Voucher specimens of a male and female have been deposited at the University of New Hampshire. Enallagma exsulans (Hagen) (Stream Bluet). This species was widespread along the river north to the Reservoir segment, and increased in abundance from south to north. 180 Northeastern Naturalist Vol. 17, No. 2 Table 2. Relative abundance of river-specialist Odonata along the Connecticut River in New Hampshire and Vermont. Distribution along the river is presented in terms of five of the river segments: N = Northern, R = Reservoir, C = Central, M = Macrosite, and S = Southern. Abundance categories are: A (abundant), C (common), U (uncommon), and R (rare). Abundance codes in lower case refer to records not obtained by one of the authors of this paper. Species indicated by an asterisk are not typical of larger rivers, but are included here to provide a more comprehensive overview of the lotic taxa present in each segment of the river under consideration. County distributions for New Hampshire and Vermont are from north to south as follows. NH: Co = Coos, Gr = Grafton, Su = Sullivan, and Ch = Cheshire; VT: Es = Essex, Ca = Caledonia, Or = Orange, Ws = Windsor, and Wh = Windham. Although the main stem of the river is technically entirely within New Hampshire, odonate records from the main stem of the Connecticut River are listed twice in this table, with one county on either side of the river. River segment Family/scientific name Common name N R C M S NH counties VT counties Calopterygidae Calopteryx aequabilis * River Jewelwing C R Co, Su Es, Ws Calopteryx maculata * Ebony Jewelwing R Co Es Coenagrionidae Argia apicalis Blue-fronted Dancer R Ch Wh Argia fumipennis * Variable Dancer R Gr Or Argia moesta Powdered Dancer R A C Gr, Su, Ch Or, Ws, Wh Enallagma antenatum Rainbow Bluet U Gr Ws Enallagma carunculatum Tule Bluet U U U Gr, Ch Ca, Ws, Wh Enallagma civile Familiar Bluet U U Gr, Su, Ch Ws, Wh Enallagma durum Big Bluet U Ch Enallagma exsulans Stream Bluet U C A A Gr, Su, Ch Ca, Or, Ws, Wh Aeshnidae Basiaeschna janata * Springtime Darner R Gr Ws Boyeria vinosa Fawn Darner U U U U Gr, Su, Ch Ca, Or, Ws, Wh 2010 P.D. Hunt, M. Blust, and F. Morrison 181 Table 2, continued. River segment Family/scientific name Common name N R C M S NH counties VT counties Gomphidae Dromogomphus spinosus Black-shouldered Spinylegs r U U C Co, Gr, Su, Ch Es, Or, Ws, Wh Gomphus abbreviatus Spine-crowned Clubtail C Ch Wh Gomphus descriptus Harpoon Clubtail U Co Es Gomphus quadricolor Rapids Clubtail R Ch Wh Gomphus vastus Cobra Clubtail R U U Gr, Su, Ch Or, Ws, Wh Gomphus ventricosus Skillet Clubtail r Ch Wh Hagenius brevistylus Dragonhunter R R R U Co, Gr, Su, Ch Es, Or, Ws, Wh Lanthus sp. Pygmy clubtail species R Co Es Ophiogomphus rupinsulensis Rusty Snaketail U U Gr, Su, Ch Ws, Wh Progomphus obscurus Common Sanddragon R Su Ws Stylogomphus albistylus Eastern Least Clubtail R Co, Ch Es, Wh Stylurus amnicola Riverine Clubtail U Su, Ch Ws, Wh Stylurus scudderi Zebra Clubtail A U U U R Co, Gr, Su, Ch Es, Ca, Or, Ws,Wh Stylurus spiniceps Arrow Clubtail U C A A Co, Gr, Su, Ch Es, Or, Ws, Wh Cordulegastridae Cordulegaster diastatops * Delta-spotted Spiketail R Co Es Cordulegaster maculata Twin-spotted Spiketail U R Co, Gr Es, Or Macromiidae Didymops transversa Stream Cruiser R R R Co, Gr, Su, Ch Or, Ws, Wh Macromia illinoiensis Swift River Cruiser U U C C Co, Gr, Su, Ch Es,Or,Ws,Wh Cordulidae Epitheca princeps Prince Baskettail R r U U Co, Gr, Su, Ch Es, Ws, Wh Helocordulia uhleri Uhler's Sundragon R R Co, Gr, Ch Ws, Wh Neurocordulia yamaskanensis Stygian Shadowdragon U C A Gr, Su, Ch Or, Ws, Wh 182 Northeastern Naturalist Vol. 17, No. 2 Family Aeshnidae (darners) Boyeria vinosa Say (Fawn Darner). Exuviae of this species were found in all segments except the Northern. It is more typical of smaller forested streams and probably doesn’t reach high densities along most of the Connecticut River. Family Gomphidae (clubtails) Dromogomphus spinosus Selys (Black-shouldered Spinylegs). This species is typical of medium to large rivers, as well as lakes and large ponds. While widespread in the Southern, Macrosite, and Central segments, it was not found in either the Northern or Reservoir segments, where suitable habitat occurs. There is a single unverified exuviae record for the Northern segment from Stratford, NH (fide M. Blust). Gomphus abbreviatus Hagen (Spine-crowned Clubtail). The only records of this species for the region are from the Southern segment: a single exuviae below Vernon Dam on 10 June 2005 and 14 exuviae above Brattleboro, VT on 11 June 2007. It is widespread and often abundant in portions of the lower Merrimack River watershed of New Hampshire, and thus might be expected to occur elsewhere on the Connecticut River. Although listed as endangered in Massachusetts, this status predates recent surveys that have found it more abundant than previously believed. Gomphus descriptus Banks (Harpoon Clubtail). The only records from the upper Connecticut River are from the Northern segment. This species is usually found in smaller coldwater rivers and streams, and would thus only be expected in this segment or farther upstream. It is listed as threatened in Connecticut. Gomphus quadricolor Walsh (Rapids Clubtail). The only record for the upper Connecticut River is an exuviae collected in the Southern segment above the mouth of the West River in Brattleboro on 11 June 2007. Elsewhere in the Northeast, it is known to occur on a wide range of rivers, including the Connecticut River or its tributaries in Massachusetts (Harper and Loose 2006), in northwestern Connecticut (Wagner and Thomas 1999), and the Merrimack drainage of New Hampshire (White and Morse 1973; P. Hunt, unpubl. data). More detailed surveys are needed to determine its status on the upper Connecticut River. It is listed as threatened in both Connecticut and Massachusetts, and as special concern in New Hampshire. Gomphus vastus Walsh (Cobra Clubtail). This large-river specialist appears to reach its northern range limit along the Connecticut River in the Central segment, where a single exuviae was collected from Haverhill on 23 July 2006. To the south, it occurs in higher numbers, but was unrecorded in the free-flowing section of the Macrosite segment below Wilder Dam. It is often the most abundant gomphid in sections of the river in Massachusetts (McLain et al. 2004, 2006). It is listed as special concern in both Connecticut and Massachusetts. Gomphus ventricosus Walsh (Skillet Clubtail). There are only three records of this species for the region, all in the Southern segment: Hinsdale 2010 P.D. Hunt, M. Blust, and F. Morrison 183 4 July 1939 (J. Bailey, in Kormondy 1960), and two unverified exuviae from Chesterfield and Walpole, NH in 1994 (no dates, fide M. Blust). Regular exuviae searches in Hinsdale and Walpole in 2005 and 2006, respectively, failed to detect the species, but this may be a result of the species’ tendency to emerge in microhabitats (e.g., tree root tangles) that are difficult to search (D. Wagner, pers. comm.). It becomes more common in Massachusetts (McLain et al. 2004, 2006) and Connecticut (Wagner and Thomas 1999), but is never abundant. It is listed as special concern in Connecticut, Massachusetts, and New Hampshire. Hagenius brevistylus Selys (Dragonhunter). This species appeared uncommon along the Connecticut River in comparison to its relative abundance on other rivers in New Hampshire. Given that it can be common on large lakes, its apparent absence from the Reservoir segment is likely a result of our undersampling of this area. Lanthus species (probably L. parvulus Selys [Northern Pygmy Clubtail]). The only record of this genus for the Connecticut River is a larva from the Northern segment at Maidstone Bends on 16 Oct 2007. This species and the closely related L. vernalis are found in both small forested streams and shallow reaches of large cold rivers, and either could occur elsewhere along the northernmost segments of the Connecticut River. Their near absence from this study is thus likely a result of both limited survey work and limited suitable habitat. Ophiogomphus rupinsulensis Walsh (Rusty Snaketail). Unlike other members of this genus, this species often occurs on larger rivers, usually at sites with rapidly moving water. As such, it was the only species of Ophiogomphus recorded from the study area. Progomphus obscurus (Rambur) (Common Sanddragon). A female captured on Hart Island, Plainfield, NH (Macrosite segment) on 14 August 2006 appears to be the only record of the species for the entire Connecticut River Watershed (Donnelly 2004a, Wagner and Thomas 1999), and further investigation into the species’ population status in the Macrosite is warranted. It is locally common in sand-bottomed rivers and streams in the Merrimack River drainage of New Hampshire, with the closest site to Hart Island being over 60 km to the southeast. The species has yet to be recorded in Vermont. It is listed as special concern in Connecticut, where it is restricted to the southeastern corner of the state (Wagner and Thomas 1999). Stylogomphus albistylus (Hagen in Selys) (Eastern Least Clubtail). The only record from the survey area is an exuviae below Vernon Dam on 6 July 2005. This clubtail is typical of riffles on streams and smaller rivers, and would not be expected to occur regularly on most of the Connecticut River. Stylurus amnicola (Walsh) (Riverine Clubtail). This species was not recorded in Vermont and New Hampshire until 2005, when exuviae were found both above and below Vernon Dam on 21 June and 5 July, respectively. Our surveys in 2006 documented it north through the rest of the Southern segment to the mouth of the Black River in Springfield, VT. It is listed as 184 Northeastern Naturalist Vol. 17, No. 2 threatened in Connecticut, endangered in Massachusetts, and special concern in New Hampshire. Stylurus scudderi (Selys) (Zebra Clubtail). Although abundant in the Northern segment, this species declines in abundance to the south. This pattern reflects its preference for smaller streams or rivers with sandy substrates. In Connecticut and Massachusetts, it is considered uncommon to rare, and is listed as endangered in Connecticut (although the latter status may reflect undersampling). Stylurus spiniceps (Walsh) (Arrow Clubtail). This species increases in abundance from north to south, and the Northern segment may represent its northern range limit. It was not recorded in the Reservoir segment, but likely occurs there. It is often the most abundant species in exuviae searches in the Macrosite and Southern segments. It remains abundant south of the border in Massachusetts, but is generally less common than G. vastus in areas studied to date (McLain et al. 2006). It is listed as threatened in Massachusetts. Family Cordulegastridae (spiketails) Cordulegaster maculata Selys (Twin-spotted Spiketail). This species is typical of small forested streams, and was not expected in most of the study area. It was uncommon in the Northern segment, and a single exuviae from the Central segment in Haverhill was in a section of river with extensive braiding that presumably was similar to its preferred habitat. Family Macromiidae (cruisers) Didymops transversa (Say) (Stream Cruiser). The only exuviae record from the study area was at Wilder Dam in 2006. Adults have been observed in the Central, Macrosite, and Southern segments. Because larvae can move considerable distances prior to eclosure, exuviae were likely underrepresented on our surveys. Macromia illinoiensis Walsh (Swift River Cruiser). This species is typical of large rivers and lakes, and occurred throughout the study area with the exception of the Reservoir segment (likely a sampling artifact). It is one of the more frequently seen adult dragonflies over slow-moving stretches of the Connecticut River. Family Cordulidae (emeralds) Epitheca princeps Hagen (Prince Baskettail). Although adults have been seen over the river in most segments, exuviae were most common above Wilder and Bellows Falls dams in the Macrosite and Southern segments. This pattern is typical for the species, which generally breeds in relatively still waters. Helocordulia uhleri (Selys) (Uhler’s Sundragon). The only records are of single adults in Lebanon and Walpole. No exuviae were found during 2005–2006 surveys, and more thorough sampling—including of larvae—is needed to determine the species’ actual distribution along the upper Connecticut River. 2010 P.D. Hunt, M. Blust, and F. Morrison 185 Neurocordulia yamaskanensis (Provancher) (Stygian Shadowdragon). Although found north through the Central segment, this species was most common immediately above the Wilder, Bellows Falls, and Vernon dams, a pattern similar to that seen in E. princeps. It should be looked for near the large dams in the Reservoir segment as well. Conservation and future needs Odonata are often considered to be useful as environmental indicators for water quality and microhabitat condition. All four states along the Connecticut River have undertaken reviews of their riverine Odonata fauna (Connecticut Department of Environmental Protection 2004, Hunt 2006a, Massachusetts Division of Fisheries and Wildlife 2005, Vermont Fish and Wildlife Department 2005, Wagner and Thomas 1999; M. Blust and B. Pfeiffer, unpubl. data), and in some cases, specific studies of human impacts have been initiated (McLain et al. 2004, 2006). Such impacts are highly varied, given that the Connecticut River valley has been occupied by people of European descent for close to 300 years. The principal threat facing the Odonata of the Connecticut River is habitat loss or degradation, primarily in the form of altered flows. The Connecticut River is a heavily modified waterway, with six large dams along its main stem and several smaller dams in the northern portions. Dams affect odonate habitat in several ways. Sedimentation increases in the quiet waters of impoundments, converting substrates from cobble or sand to mud or silt. Such changes can lead to faunal shifts as species typical of one substrate give way to those typical of another. For example, species such as S. spiniceps prefer silt, while G. vastus prefers gravel (McLain et al. 2004, 2006). Dams also eliminate upstream riffle habitats such as those preferred by some clubtail species (e.g., Ophiogomphus). Extensive impounding can also result in thermal conditions that are more suitable for species typical of lentic habitats, such as members of the families Aeshnidae and Libellulidae. At the same time that dams affect sedimentation, they act to interrupt a river’s flow regime and thus the natural patterns of erosion and deposition that maintain habitat above and below impounded sections. Where erosion threatens roads or buildings with undermining, banks are often stabilized with the addition of rip-rap (medium to large boulders or pieces of concrete) or vertical concrete walls. In extreme cases, entire sections of river can be so “armored”, resulting in a net loss of natural bank habitat, and possibly the removal of sediment from the river bed. This scouring is the opposite of sedimentation, and can result in stretches where the bed is dominated by gravel and provides little habitat for burrowing odonate larvae. Altered flows can also eliminate natural sandbars and the habitats they provide. Less well documented are threats related to adult survival during or after emergence (McLain et al. 2004, 2006). Emerging soft-bodied adults are particularly vulnerable to damage from boat wakes and changing water levels related to dam operations, but the magnitude of this mortality and its effects at the population level remain to be quantified. Upland forested 186 Northeastern Naturalist Vol. 17, No. 2 riparian buffers may provide critical habitat for recently-emerged adults, and their loss or degradation may similarly affect survival (Massachusetts Division of Fisheries and Wildlife 2005, Vermont Fish and Wildlife Department 2005), although data are again lacking. Not generally listed among threats to Odonata is pollution, including agricultural and road run-off, point-source contaminants, and atmospheric deposition, but none of these factors has received research attention. Finally, the potential effects of climate change on riverine Odonata are completely speculative, but include altered rainfall (and thus river flows) and increased temperature, both of which could alter species’ distribution and abundance along the river. Of the 28 species whose distributions along the Connecticut River were discussed in detail above, several have been identified as species of regional conservation concern (Hunt 2006a, Massachusetts Division of Fisheries and Wildlife 2005, Vermont Fish and Wildlife Department 2005). Of these, G. vastus, S. scudderi, and S. spiniceps are widespread and often abundant, and despite habitat changes and the risks associated with emergence, are probably secure at the scale of a given river segment. Stylurus amnicola is much more local in distribution, and given the high levels of conservation concern in Connecticut and Massachusetts, also bears watching in the northern two states. The rarest three species along the Vermont–New Hampshire portion of the Connecticut River are G. quadricolor, G. ventricosus, and P. obscurus. The two Gomphus may reach their northern range limits in the Southern segment, and like S. amnicola, are species of conservation concern in southern New England. More data are needed on their distributions and potential threats within the study area. Finally, although P. obscurus is not rare in New Hampshire, its discovery in the Macrosite may indicate a unique habitat type in this segment worthy of further investigation and protection. Although our knowledge of the Odonata of the upper Connecticut River has increased dramatically in recent years, there is still much to be learned. Perusal of the species summaries and Table 2 indicates that the northern stretches of the river are particularly in need of additional survey work, and the northernmost Headwaters segment lacks detailed data entirely. More biotic inventories in the Northern, Reservoir, and Central segments would help refine northern range limits of several species typical of more southerly sections of the river. In the Macrosite segment, discovery of Progomphus obscurus points to the need for more detailed surveys of this biologically rich and relatively unimpacted portion of the river. While the Southern segment has been better surveyed than the others, additional species could still be discovered in this area. Among these are Hetaerina americana (Fabricius) (American Rubyspot), Neurocordulia obsoleta (Say) (Umber Shadowdragon), and Gomphus fraternus Say (Midland Clubtail). A better knowledge of these species’ distributions along the Connecticut River has heuristic value and may inform future conservation decisions. Finally, the fact that three rare species (G. quadricolor, G. ventricosus, and S. amnicola) have already been documented in this segment illustrates the need for additional survey 2010 P.D. Hunt, M. Blust, and F. Morrison 187 work. Such surveys should include larval sampling, more systematic exuviae surveys (especially earlier in the season), and standardization of effort over space and time. Only then will we have baseline data against which to measure future changes and potentially link them with changes in habitat or other environmental factors. Acknowledgments The authors would particularly like to thank those entities which have funded their surveys along the Connecticut River. Pamela Hunt’s work in 2006 was funded by a grant to the New England Institute for Landscape Ecology through the Connecticut River Joint Commissions Partnership Program. Michael Blust’s surveys in the Maidstone Bends were funded by the Vermont Chapter of The Nature Conservancy. Work along the Connecticut River in northern Massachusetts by Fred Morrison and colleagues has been funded by Northeast Generation Services, the Massachusetts Environmental Trust Fund, and the Franklin Land Trust. The authors also wish to thank Hal White, Bryan Pfeiffer, David Wagner, Michael Thomas, Glen Mittelhauser, Paul Brunelle, and three anonymous reviewers for comments on the manuscript and the several volunteers who have occasionally aided in dragonfly searches along the river. Literature Cited Blust, M.H. 2006. Odonate survey of Maidstone Bends, 2006 Season. Report to Vermont Chapter of The Nature Conservancy. Green Mountain College, Poultney, VT. 8 pp. Connecticut Department of Environmental Protection. 2004. Connecticut’s threatened, endangered, and special concern species 2004. Connecticut Department of Environmental Protection, Hartford. Available online at http://www.ct.gov/dep/ lib/dep/wildlife/pdf_files/nongame/ETS04.pdf. Accessed 10 February 2007. DeLorme. 2001. New Hampshire Atlas and Gazetteer. DeLorme, Yarmouth, ME. Donnelly, T.W. 2004a. Distribution of North American Odonata. Part I: Aeshnidae, Petaluridae, Gomphidae, Cordulegastridae. Bulletin of American Odonatology 7:61–90. Donnelly, T.W. 2004b. Distribution of North American Odonata. Part III: Calopterygidae, Lestidae, Coenagrionidae, Protoneuridae, Platystictidae; with data sources and bibliography. Parts I–III. Bulletin of American Odonatology 8:33–99. Francis, S.F., and A.D. Mulligan (Eds.). 1997. Connecticut River Corridor Management Plan. Connecticut River Joint Commissions, Charlestown, NH. 124 pp. Harper, L., and J. Loose. 2006. The distribution of rare riverine dragonflies in Massachusetts: Where to make new discoveries! Ode News 8(1):6–11. Hunt, P.D. 2006a. Assessment of the conservation status of Odonata in New Hampshire. Report to the New Hampshire Fish and Game Department. Audubon Society of New Hampshire, Concord, NH. 17 pp. Hunt, P.D. 2006b. Odonata of the Connecticut River mainstem. Report to Connecticut River Joint Commissions. New England Institute for Landscape Ecology, Canaan, NH. 19 pp. Kormondy, E.J. 1960. North American records of Anisopterous Odonata. Entomological News 71:121–130. 188 Northeastern Naturalist Vol. 17, No. 2 Massachusetts Division of Fisheries and Wildlife. 2005. Massachusetts comprehensive wildlife conservation strategy. Massachusetts Division of Fisheries and Wildlife, Boston, MA. 751 pp. McLain, D., F. Morrison, and L. Sanders. 2004. Dragonfly population dynamics, effects of bank stabilization, and ecology of nymphs in the Turners Falls pool of the Connecticut River: 2004 field season. Report to The Massachusetts Environmental Trust Fund, Franklin Land Trust, and Northeast Generation Services. A Natural Focus, Westhampton, MA. 21 pp. McLain, D., F. Morrison, and L. Sanders. 2006. Bank stabilization and dragonfly emergence, population dynamics, and larval ecology in the Turners Falls pool of the Connecticut River: 2005 field season. Report to Northeast Generation Services, Massachusetts Environmental Trust Fund, and Franklin Land Trust. A Natural Focus, Westhampton, MA. 20 pp. New Hampshire Fish and Game Department. 2005. New Hampshire wildlife action plan. New Hampshire Fish and Game Department, Concord, NH. 1183 pp. Nikula, B., J.L. Loose, and M.R. Burne. 2003. A Field Guide to the Dragonflies and Damselflies of Massachusetts. Massachusetts Division of Fisheries and Wildlife, Natural Heritage and Endangered Species Program, Westborough, MA. 197 pp. Vermont Fish and Wildlife Department. 2005. Vermont wildlife action plan. Vermont Fish and Wildlife Department, Waterbury, VT. 1084 pp. Wagner, D.L., and M.C. Thomas. 1999. The Odonata fauna of Connecticut. Bulletin of American Odonatology 5:59–85. White III, H.B., and W.J. Morse. 1973. Odonata (dragonflies) of New Hampshire: An annotated list. NH Agricultural Experiment Station Research Report 30, Durham, NH. 46 pp.