The Behavior of the Seaside Dragonlet, Erythrodiplax berenice
(Odonata: Libellulidae), in a Maine Salt Marsh
W. Herbert Wilson, Jr.*
Abstract - The behavior of the marine dragonfly, Erythrodiplax berenice (Seaside Dragonlet),
was studied over two summers at Weskeag Marsh, South Thomaston, ME. These dragonflies
are lethargic, spending over 99% of their time perched on the culm of a salt marsh plant. No
evidence of territorial behavior was found. Females oviposit while in tandem in algal mats on
the surface of salt-water pannes. These dragonflies perch preferentially into the wind, presumably
to aid in providing lift on take-off.
The Seaside Dragonlet, Erythrodiplax berenice (Drury), is a widespread libellulid
dragonfly associated with marine habitats. In North America, this species is a
denizen of salt marshes and other supratidal habitats from Nova Scotia southward to
Florida along the Atlantic Ocean and from Florida to extreme southern Texas on the
Gulf of Mexico (Needham et al. 2000, Nikula et al. 2003). Disjunct populations are
found in saline lakes in Texas and New Mexico. Beyond North America, the Seaside
Dragonlet is found along the Atlantic coast from Mexico south to Venezuela, along
the Pacific coast from Baja California to Oaxaca, Mexico and on some Caribbean
islands (Dunkle 1989, 2000; Shortess 1930).
The Seaside Dragonlet is perhaps most notable for the fact it is the only truly
marine odonate (Dunson 1980, Kelts 1977). Nymphs live in salt-water pools with
salinities as high as 48‰ (Dunson 1980).
Despite the broad geographic distribution of the Seaside Dragonlet, there is a
dearth of information on adult behavior and reproductive ecology. In this contribution,
I describe the results of observations of a Seaside Dragonlet population in a
Maine salt marsh during the summers of 2001 and 2002.
The data for this study were collected in Weskeag Marsh in South Thomaston,
ME (44.04693°N, 69.08454°W). The particular area studied was an arm of Weskeag
Marsh on the north side of Buttermilk Lane, extending for about 2 km in a southwesterly
to northeasterly direction. The width of this portion of the marsh varies from 100
to 300 m and is bounded by upland forest on either side. Marsh Creek runs in a sinuous
fashion through the marsh, flowing into the Weskeag River south of Buttermilk
Lane. The marsh is dotted with salt pannes.
Most of the marsh is covered with Spartina patens (Ait.) Muhl. (Salt Marsh Hay)
although S. alterniflora Loisel. (Smooth Cordgrass) is found along the banks of Marsh
Creek. A number of other halophytes are present at low densities (e.g., Salicornia maritima
Wolff & Jefferies [Slender Glasswort], Limonium carolinianum (Walt.) Britt.
[Carolina Sea-lavender] and Distichlis spicata (L.) Greene [Seashore Saltgrass]).
Preliminary visits to Weskeag Marsh in the summer of 2000 indicated that adult
Seaside Dragonlets do not appear in significant numbers until early July. In 2001
and 2002, I visited Weskeag Marsh on 18 different days. Observations were made
between 0800 and 1700, when the dragonlets are most active. Temperatures ranged
from 14 to 27 °C. Skies were clear or partly cloudy on each observation date. Although
I did not attempt to do daily censuses, population densities were clearly the
highest in the first and second weeks of July.
The primary means of data collection was following a focal individual until the
individual flew out of sight or my patience waned because of lack of activity. Most
individuals were monitored between 15 and 60 minutes.
Notes of the Northeastern Nat u ral ist, Issue 15/3, 2008
465
466 Northeastern Naturalist Notes Vol. 15, No. 3
I described any behavioral trait (perching height, perching site, flight distance,
interaction with conspecifics and other insects) into a hand-held tape-recorder. Observations
were made through 8X Bushnell Elite close-focusing binoculars from a
distance of 3–4 meters.
On five dates in 2002, I made opportunistic observations on the orientation of
perched Seaside Dragonlets with respect to the wind direction. On each occasion, the
wind was either from the northeast or southwest so that a considerable fetch resulted.
Wind speeds varied from 8 to 16 km per hour on each day.
When I found a perched Seaside Dragonlet, I recorded the quadrant in which
its head was directed relative to the wind. For a southwesterly wind, a dragonfly
with its head oriented between south and west was scored as upwind. Similarly, a
dragonfly oriented beween north and east was scored as downwind. Dragonflies oriented
beween west and north or between east and south were scored as lateral to the
wind. Some observations were made from upwind, others downwind and yet others
lateral, eliminating any possible bias in direction of my presence on the orientation
of the dragonflies. Two sets of observations were made in the morning, one at noon,
and two in the mid-afternoon. Thus, orientation with respect to the sun was not a
confounding variable.
Oviposition occurred in the many salt pannes in the marsh. To describe oviposition
behavior, I stood near the edge of one or more salt pannes and made observations
with binoculars, recording data with the tape-recorder.
The adjective that best describes Seaside Dragonlet behavior in this study is lethargic.
These dragonflies spent the vast majority of their time perched. Most perched
on a culm of Spartina alterniflora at a height of 5–10 cm above the marsh surface,
although occasionally one perched on S. patens thatch that had accumulated on the
marsh surface.
Observations of focal individuals indicate that Seaside Dragonlets usually spent
more than 99% of their time perched (n = 37 [20 males, 17 females]). Of four randomly
drawn individuals, a female stayed perched for 55 minutes, a second female
made four 3-second flights in 36 minutes, a male made five 3-second flights in 10
minutes and 45 seconds, and a second male made four 3-second flights in 30 minutes
The proportions of the observation period these four individuals spent in flight are
0%, 0.1%, 1.4% and 0.7%. All flights were less than 30 cm in distance, and the dragonfly returned to its original perch in each case.
The sedentary nature of Seaside Dragonlets is striking when compared to other
dragonflies in Weskeag Marsh (the libellulids Libellula pulchella Drury [Twelvespotted
Skimmer], Plathemis lydia (Drury) [Common Whitetail], and Sympetrum
vicinum (Hagen) [Autumn Meadowhawk], as well as the corduliid Epitheca cynosura
(Say) [Common Baskettail]). A few comparable rates of perching time in libellulid
dragonflies are available in the literature. Campanella and Wolf (1974) showed male
Common Whitetails flew 36–68% of the time. Pezalla (1979) found that Twelve-spotted
Skimmers spent 64–82% of their time in flight (patrol flights or pursuit flights).
Fried and May (1983) showed male Pachydiplax longipennis (Burmeister) (Blue
Dashers) spent 24–56% of their time in flight.
The short flights made by focal individuals were not impelled by the presence
of a conspecific. These forays may have been survey flights or foraging flights (prey
items were not seen).
No evidence of territorial behavior by Seaside Dragonlets at Weskeag Marsh
was found. Males were frequently observed to be perched within 1 m of each other.
On 10 July 2001, four males were seen perched together adjacent to a pool within
2008 Northeastern Naturalist Notes 467
a 3-m2 area without interacting. On 1 July 2002, a male Seaside Dragonlet flew up
and chased a Common Baskettail that was flying over the marsh; the male did not
respond to another perched Seaside Dragonlet less than a meter away. I recorded two
brief aerial encounters of uncertain function of two Seaside Dragonlets on the same
date. Otherwise, aggressive interactions between males were not observed away from
oviposition sites.
Table 1 presents data on perching orientation of Seaside Dragonlets with respect
to the wind. If the perching orientations were randomly distributed, one would expect
a ratio of 1:1:2 for upwind, downwind, and lateral orientations. The upwind and
downwind orientations each include 90° of the compass, and the lateral orientation
includes 180° of the compass. The observed distribution is statistically different from
random by a chi-square test (χ2 = 19.32, P < 0.001).
Pairs of Seaside Dragonlets in tandem flew from the marsh to the pools dotting
the marsh. Despite diligent efforts, I was never able to discover where pairing of
males and females took place. Coupling presumably takes place on the marsh surface
before the tandems approach the salt panne for oviposition.
Each tandem flew over the surface of the pool with the female spanking the
water smartly with her abdomen. Oviposition usually occurred in mats of algae
growing on the surface of the pool. A pair in tandem would fly over the pool and
periodically fly slowly for periods of four to 48 seconds over an acceptable oviposition
site. The mean number of dips of the female’s abdomen per second was 2.1
(± 0.18 [s.d.]).
Tandems were often attacked by solitary males in flight over the surface of the
pool. I never saw a solitary male succeed in uncoupling a tandem, however.
Adult dragonflies can be classified behaviorally as perchers or fliers (Corbet
1999). Fliers, such as most members of the family Aeshnidae, spend much of their
time in flight. Perchers, including most libellulids, alternate between flying and
perching, often on the same perch. Seaside Dragonlets, libellulid dragonflies, are
clearly perchers. However, the time spent on a perch is extraordinarily long in Weskeag
Marsh with most individuals spending less than 1% of their time in flight.
With a length of only 28–35 mm (Dunkle 1989, 2000), Seaside Dragonlets are
weak fliers. Their reluctance to fly may be an adaptation to minimize predation from
birds, other odonates, spiders, and other predators. Seaside Dragonlets were the only
odonates that I saw captured in the many spider webs found at Weskeag Marsh, perhaps
an indication of its weak flying ability.
I found no evidence of territorial behavior of Seaside Dragonlets at Weskeag
Marsh. This report differs from the only information I can find in the literature on
adult behavior of Seaside Dragonlets: a single sentence in a field guide (Dunkle
1989) stating that Seaside Dragonlet males in Florida perch near pools and defend
about 5 m of shoreline.
Table 1. Orientation of Seaside Dragonlets relative to the wind direction on five days at Weskeag
Marsh, South Thomaston, ME.
Date Wind direction Upwind Downwind Lateral
7/8/02 SW 1 0 0
7/10/02 NE 4 0 0
7/22/02 NE 8 1 0
7/26/02 NE 6 3 2
8/13/02 SW 5 1 0
Total 24 5 2
468 Northeastern Naturalist Notes Vol. 15, No. 3
The lack of territoriality in the Maine population contrasted with territorial
populations further south suggests that a biogeographic study of Seaside Dragonlets
could be profitable. In Weskeag Marsh, the only aggressive interactions predictably
observed were solitary males attempting to break-up ovipositing tandems flying over
the salt pannes.
Seaside Dragonlets perch preferentially with their head into the dominant wind
direction (Table 1). A reasonable hypothesis for this orienting behavior is the generation
of lift when a Seaside Dragonlet takes flight. Odonate wings are fusiform in
cross-section and therefore can generate lift through the Bernoulli effect when facing
into the wind (Alexander 2002).
Acknowledgments. This research was supported by grants from the Division of
Natural Sciences at Colby College. Bets Brown, Blair Nikula, and three anonymous
reviewers made many helpful suggestions on earlier drafts of this manuscript.
Literature Cited
Alexander, D.E. (2002). Nature’s Flyers: Birds, Insects, and the Biomechanics of Flight. Johns
Hopkins University Press, Baltimore, MD. 358 pp.
Campanella, P.J., and L.L. Wolf. 1974. Temporal leks as a mating system in a temperate-zone
dragonfly (Odonata: Anisoptera). I: Plathemis lydia (Drury). Behaviour 51:47–87.
Corbet, P.S. 1999. Dragonflies: Behavior and Ecology of Odonata. Comstock Press, Cornell
University, Ithaca, NY. 829 pp.
Dunkle, S.W. 1989. Dragonflies of the Florida Peninsula, Bermuda and the Bahamas. Scientific
Publishers, Gainesville, fl. 155 pp.
Dunkle, S.W. 2000. Dragonflies through Binoculars: A Field Guide to Dragonflies of North
America. Oxford University Press, Oxford, UK. 266 pp.
Dunson, W.A. 1980. Adaptations of nymphs of a marine dragonfly, Erythrodiplax berenice, to
wide variations in salinity. Physiological Zoology 53:445–452.
Fried, C.S., and M.L. May. 1983. Energy expenditure and food intake of territorial male Pachydiplax
longipennis (Odonata: Libellulidae). Ecological Entomology 8:283–292.
Kelts, L.J. 1977. Ecology of two tidal marsh insects, Trichocorixa verticalis (Hemiptera) and
Erythrodiplax berenice (Odonata), in New Hampshire. Ph.D. Dissertation. University of
New Hampshire, Durham, NH.
Needham, J.G., M.J. Westfall, Jr., and M.L. May. 2000. Dragonflies of North America. Scientific Publishers, Gainesville, fl. 839 pp.
Nikula, B., J.L. Loose, and M.R. Burne. 2003. A Field Guide to the Dragonflies and Damselflies
of Massachusetts. Natural Heritage and Endangered Species, Program, Massachusetts Division
of Fisheries and Wildlife, Westborough, MA. 197 pp.
Pezalla, V.M. 1979. Behavioral ecology of the dragonfly Libellula pulchella Drury (Odonata:
Anisoptera). American Midland Naturalist 102:1–22.
Shortess, S.L. 1930. Variations in wing reticulation and size of Erythrodiplax berenice with
regard to geographical distribution. Transactions of the American Entomological Society
55:415–423.
*Department of Biology, Colby College, 5739 Mayflower Hill, Waterville, ME, 04901; whwilson@
colby.edu.