Lestes disjunctus Selys and L. forcipatus Rambur (Odonata: Lestidae): Some Solutions for Identification

2004·5.2 MB·English

by  John P Simaika

J. Entomol. Soc. Brit.Columbia 101,December2004 131
Lestes disjunctus Selys and L.forcipatus Rambur
(Odonata: Lestidae): Some Solutions for Identification
JOHN P. SIMAIKA* and ROBERT A. CANNINGS^^
ABSTRACT
Five species of the damselfly genus Lestes live in British Columbia, Canada, and of
these, Lestesforcipatus Rambur and L. disjunctus Selys are the most similar and most
difficultto separate morphologically. Females canbereadily distinguishedby the size of
the ovipositor, but males are difficult to separate. In British Columbia, L. disjunctus is
the more common, widespread and familiar species. Before 1998, L. forcipatus speci-
mens were mistaken forthose ofL. disjunctus because the former isprimarily an eastern
North American species and because most Lestes species are usually identified using
male characters. With the discovery that L.forcipatus is part ofthe western fauna, an
evaluation ofthe relative status ofthe two species in British Columbia is necessary. The
best method for separating the two species uses the length ofthe anterior lamina (part of
the secondary genitalia) as a unique character or as part ofratios using other measure-
ments. In addition, in at least western North America, L.forcipatus males are more pru-
inescent than those ofL. disjunctus, especially on the thorax. Identification using the
pruinescence pattern was tested in the field and is recommended as a simple and accu-
rate method forwestern North America. Soaking Odonata specimens in acetone, a com-
mon technique used to preserve colours, damages surface pruinescence and should not
be used to preserve mature, pruinescent adults, including those ofLestes species. To
identify L disjunctus and L.forcipatus males treated in acetone, it may be necessary to
calculate ratios based on various character measurements. Future research should inves-
tigate spatial and temporal differences between the species, as well as modes ofinter-
specific communication.
KeyWords: Odonata,Lestesforcipatus, Lestes disjunctus, identification, British Co-
lumbia, pruinescence, acetone, anteriorlamina.
INTRODUCTION
Five species of the damselfly genus standing water habitats with abundant
Lestes (Odonata: Zygoptera: Lestidae) aquatic vegetation and, in southern BC,
occur in British Columbia (BC), Canada: adults fly from mid-June to mid-October
L. congener Hagen (Spotted Spreadwing), (Cannings 2002).
L. disjunctus Selys (Northern Spreadwing), L. forcipatus is generally much less
L. dryas Kirby (Emerald Spreadwing), L. common than L. disjunctus, although it is
forcipatus Rambur (Sweetflag Spread- as abundant in some cold fen habitats, and
wing), and L. ungiuculatus Hagen (Lyre- both species often occurat the same site. L.
tipped Spreadwing). L. disjunctus is the forcipatus does not range as far north as L.
most common, widespread and familiar disjunctus and, although not known from
Lestes species in the province, and one of much ofBC's north, it has been collected
the most abundant odonates in Canada, in the southeastern Yukon. In the western
ranging as far north as the Arctic treeline Canadian Cordillera, it is most common in
(Cannings 2002). It inhabits many types of sedge fens (Cannings 2002). Walker
'#402-1063 Foul BayRoad, Victoria, British Columbia, Canada V8S4J3
^RoyalBritish ColumbiaMuseum, 675 Belleville Street, Victoria, BritishColumbia, Canada V8W9W2
^Authortowhomcorrespondenceshouldbesent
132 J.Entomol. Soc.Brit.Columbia 101,December2004
(1953) described L. forcipatus habitat in females. In L. forcipatus females the ovi-
Ontario as "ponds, both temporary and positor valves reach the tips ofthe cerci; in
permanent, marshy lakes, and slow, weedy L. disjunctus they do not (Walker 1953,
streams". In BC L. forcipatus has been Cannings 2002) (Fig. 1).
collected from mid-June to mid-September Lestes species are usually brown, black,
(Cannings 2002). metallic green or bronze above and mostly
L. forcipatus was not reported in BC pale below; especially in males, the head,
until 1998, when it was first collected in thorax, base and tip of abdomen become
the Rocky Mountain Trench north of pruinescentbluish white with age. Pruines-
Golden and subsequently found in many cence (pruinosity) is a waxy substance
other localities in the southeastern part of produced by the hypodermis in many
the province. However, it probably has groups of Odonata and excreted on the
long been a resident ofthe province; it was cuticular surface through porous canals
long overlooked because of its close re- (Gorb 1994). Pruinescence is implicated in
semblance to L. disjunctus (Ramsay and thermal regulation in dragonflies (Garrison
Cannings 2000). Before 1998, L. forci- 1976, Paulson 1983) and is thought to play
patus was not known west of Saskatche- a role in species recognition and intraspeci-
wan (Walker 1953, Westfall and May fic communication ~ indeed, the patterns
1996), and hadjust recently been found in ofpruinescence in males may be aresult of
Washington State, the first record west of sexual selection (Jacobs 1955, Corbet
Montana (Ramsay and Cannings 2000). 1999). Therefore, pruinescence patterns
The species is now known from seven might offer good species identification
counties in that state and one in Idaho characters, especially in males.
(Paulson 2004). By 1999 L.forcipatus had The object of this project was to find
been discovered at several other BC loca- novel and definitive distinguishing charac-
tions farther south and west, and by 2000 teristics between males of L. disjunctus
had been collected on Vancouver Island. and L. forcipatus, building on the studies
Some ofour old museum specimens ofL. of workers in eastern North America.
disjunctus from many regions ofthe prov- Thus, we hope (a) to distinguish males in
ince have been re-identified as L. forci- the absence of associated females; (b) to
patus, indicating that museum collections identify, with relative ease, the species in
across western Canada probably contain the field, (c) to correct any misidentifica-
many misidentified specimens. tions of specimens in BC museum collec-
Males ofL. disjunctus andL.forcipatus tions; and (d) to establish accurate distribu-
are difficuh to separate, although numer- tions for both species in BC. The first part
ous characters have been employed in ofthe present work measures certain struc-
identification (Walker 1953, Westfall and tures ofthe male genitalia to find the best
May 1996, Catling 2002, Donnelly 2003). features to separate the species. The sec-
The usual methodofdistinguishing the two ond part quantifies the degree of pruines-
species and confirming their presence at a cence ofadultmales ofeach species.
location is through identification of the
MATERIALS AND METHODS
Specimens. We measured 50 male L. dis- rowed from the Spencer Entomological
junctus and 45 male L. forcipatus speci- Museum, UBC, Vancouver, and the Slater
mens from localities in BC and Alberta Museum ofNatural History, Tacoma, WA.
(two L. disjunctus only from Alberta) and A list ofthe specimens and their collection
RBCM
from Washington and Maine in the United data is on file at the and is avail-
States. Eighty-four of the specimens were able on request. Most specimens were in
from the Royal British Columbia Museum copula or in tandem, except for three L.
(RBCM), Victoria; the others were bor- forcipatus and one L. disjunctus; thus, the
J. Entomol. Soc. Brit.Columbia 101,December2004 133
Figure 1. Lateral view ofapex offemale abdomen. Top, Lestes disjunctus; Bottom, L.forci-
patus. OV =ovipositor.
identities of almost all males were con- sperm vesicle (MS) (penis vesicle of Cat-
firmedusingthe associated females. ling (2002) andDonnelly (2003)).
Measurements. During examination, each 6. Length ofpenis shaft (PS).
specimen was held by the base of the 7. Length ofsperm vesicle (SV).
wings using a small padded alligator clamp Apexofabdominalsegment 10 (Fig. 4):
soldered to a #7 insect pin. The pin was 8. Height of apical hood (HT). This
inserted into a cork mount, and the speci- structure is a triangular projection on the
men held in a standardized measuringposi- dorsal apex ofabdominal segment 10. The
tion. Specimens were examined at lOOx apex ofthe abdomen was viewed end-on.
magnification and measurements were 9. Width ofbase ofapical hood (HL).
made to 0.01 mm. 10. Width of the abdomen (WA). The
Thirteen characters were measured; greatest width of the abdomen measured
terminology follows Westfall and May when the apex ofthe abdomen was viewed
(1996) andDonnelly (2003). end-on.
Cercus (Fig. 2): Other:
1. Distance from base ofapical tooth to 11. Length of abdominal segment 2
base ofbasal tooth (AB). (52) Measured in lateral view.
.
2. Distance from apex ofcercus to base 12. Length of abdominal segment 3
ofapical tooth (AC). (53) Measured in lateral view.
.
3. Distance from swelling at medial 13. Width of head (HD). The distance
base ofcercus to base ofbasal tooth (BB) between the extreme lateral edges of the
(not figured). eyes, measureddorsally.
Secondary genitalia (abdominal segment We analysed the difference between
2) (Fig. 3): species for each character measured using
4. Length of the anterior lamina a z-test after checking for uniformity of
(anterior hamule) (AL). Walker (1952, variance, using the MS Excel Data Analy-
1953) did not explain how to measure the sis Tool (Stinson andDodge 2004).
lamina, but Catling (2002) and Donnelly Pruinescence.
(2003) prefer to measure the ventral length Pterothorax (Fig. 5). We compared the
ofthe hamule from where it appears from extent ofpruinescence on the head, ptero-
above stemite 1. He notes, however, that thorax (fused mesothorax and metathorax),
specimens show different degrees ofbend- and abdominal segments 1 to 10 between
ing in abdominal segments 1 and 2 and males of the two species. Pterothoracic
thus there is no good reference for the ha- pruinescence was divided into several
mule base. We measured the blade of the value categories, as follows: absent = 0,
laminaonly. low lateral (below interpleural suture) = 1,
5. Length of membranous shield of mid lateral (below midline of mese-
134 J.Entomol. Soc.Brit.Columbia 101,December2004
S10
Figure 2. Dorsal view ofmale primary genitalia. Left, Lestes disjunctus; Right, L.forcipatus.
AB = distance between base ofapical tooth and base ofbasal tooth ofcercus, AC = distance
between base ofapical tooth and apex ofcercus, AT apical tooth ofcercus, BT = basal tooth
Figure 3. Ventral view ofmale secondary genitalia. AL = length ofblade ofanterior lamina,
MS = length ofmembranous shield ofsperm vesicle, PS = lengthofpenis shaft, SV = lengthof
spermvesicle, S2 = abdominal segment2, S3 = abdominal segment 3.
Figure 4. Diagrammatic apical view ofabdominal segment 10 ofmd\QLestes disjunctus. HT
height ofapical hood, HL width ofbase ofapical hood, WA=widthofabdomen.
.
J. Entomol. Soc. Brit.Columbia 101,Dhcembhr2004 135
ABC
Figure 5. Lateral view ofthorax ofLestes disjunctus male. Stippling represents coverage of
pruinescence: A, low lateral; B, midlateral; C, complete lateral.
pimeron)= 2, complete lateral (below me- recorded.
sepistemal stripe) = 3, lateral+dorsal Abdominal segment 2 (Fig. 6). We exam-
(complete lateral plus mesepistemal stripe ined segment 2 for presence or absence of
and dorsal midline) = 4, complete thorax a rectangular patch free of pruinescence
(ventral + lateral + dorsal) = 5, Because and covering about the apical one-third of
specimens had been treated in acetone and the tergite. If the patch was present, we
the pruinescence patterns were thus dam- assigned a value of 1; ifabsent, a value of
aged, both sides of the pterothorax were 0.
compared, and the more pruinescent side
RESULTS
Measurements. Nine measurements failed was not significantly different. All com-
to show a significant difference between parisons were inconclusive except for
the species: AC, BB, HT, HD, MS, PS, S3, those of the pterothorax and abdominal
SV and WA. segment 2.
Table 1 summarizes the seven measure- Pterothorax (Table 2). In all specimens of
ments that we consider important to this both species, the pterothorax was pruines-
study; these include the mean, standard cent. In L. forcipatus, it was completely
deviation, range and significance values pruinose (covered ventrally, laterally and
for z-tests. In Table 1 the z-test values for dorsally) 72.5% (n = 40) of the time; L.
WA
the HD, S3, and are not significant. disjunctus was never completely pruines-
However, tests of the same measurements cent, and never covered dorsally. L. forci-
in character ratios show significant differ- patus was covered completely laterally and
encesbetween the species. dorsally in 20% of specimens but never
Ratios of character measurements are showed only low lateral or mid lateral pru-
often useftil in preventing individual size inescence. Ofthe 40 specimens measured,
variation from obscuring the value of a three (7.5%) had only the lateral area com-
character when comparing species varia- pletely covered. L. disjunctus was com-
tion. Analysis showed that several charac- pletely covered laterally 60.0% (n = 30),
ter ratios calculated were not useful in mid laterally 30%, and low laterally 10%
separating the two species: AB/AC, AB/ ofthe time.
AL, AC/AL, AC/HD, BB/HD, HL/AL, Abdominal segment 2. Segment 2 in all L.
HT/HD, HT/HL, HT/WA, SV/AL, SV/S2, forcipatus specimens had a distinct dorsal
MS/AL, MS/HD, MS/S2, PS/HD, S3/HD, bare patch. In L. disjunctus an indistinct,
SV/HD, WA/HD. The significant character different sort of patch was present 23.1%
ratios forbothL.forcipatus andL. disjunc- (n = 26) ofthe time. It was both asymmet-
tus are summarized in Table 1 rical and lightly pruinescent. The average±
Pruinescence. The head and abdominal SD patch size (n = 28) inL.forcipatus was
segment 1 were pruinescent in all speci- 0.67 ± 0.13 mm long, by 0.50 ±0.11 mm
mens; the pruinescence on segments 3-10 wide.
136 J.Entomol. Soc.Brit.Columbia 101,December2004
Figure 6. Lateral view ofthorax andabdominal segments 1 and2 offullypruinescentmales.
Left,Lestes disjunctus; Right,L.forcipatus. PR=prothorax, PT=pterothorax, S2 = abdomi-
nal segment2.
Table 1.
Summary statistics ofmeasurements and ratios made for Lestesforcipatus and L. disjunctus.
Two-tailedz-tests were usedto compare charactersbetween species. Distances measuredare as
follows: AB = apical tooth ofcercus to basal tooth ofcercus, AL = length ofblade ofanterior
hamule, HD = width ofhead, HL = width ofbase ofapical hood, S2 = lateral length of ab-
dominal segment2, S3 = lateral length ofabdominal segment 3, WA=widthofabdomen.
Character L. forcipatus L. disjunctus z-test results
p
Distance Mean SD Range n Mean SD Range n z
(Z<=z)
AB 0.51 0.04 0.47 -0.60 42 0.45 0.05 0.33-0.53 46 6.10 0.00
AL 0.91 0.09 0.67- 1.20 44 0.72 0.06 0.60- 1.00 50 11.37 0.00
HD 4.73 0.16 4.33 - 5.00 39 4.77 0.19 4.33 - 5.07 42 -0.89 0.37
HL 0.53 0.06 0.40 -0.67 41 0.44 0.05 0.33-0.53 45 7.42 0.00
82 2.57 0.16 2.27-2.87 44 2.40 0.13 2.13-2.67 50 5.47 0.00
S3 4.32 0.24 3.87-4.87 44 4.34 0.26 3.93 - 5.00 50 -0.41 0.68
WA
1.27 0.09 1.13 - 1.53 41 1.27 0.10 1.00-1.40 45 0.16 0.87
P
Ratio Mean SD Range n Mean SD Range n z
(Z<=z)
AB/HD 0.11 0.01 0.10-0.13 37 0.09 0.01 0.07-0.11 40 6.65 0.00
AL/HD 0.19 0.02 0.14-0.24 39 0.15 0.01 0.14-0.20 42 11.99 0.00
AL/S2 0.35 0.03 0.27 -0.42 44 0.30 0.03 0.25-0.39 49 8.27 0.00
AL/S3 0.21 0.02 0.14-0.26 44 0.17 0.01 0.14-0.22 49 11.08 0.00
HL/HD 0.11 0.01 0.08-0.14 36 0.09 0.01 0.07-0.11 38 7.50 0.00
HL/WA 0.42 0.04 0.30-0.50 41 0.35 0.04 0.28 -0.47 45 7.53 0.00
S2/HD 0.55 0.03 0.48 - 0.59 39 0.51 0.02 0. 47 - 0.54 42 7.74 0.00
S2/S3 0.60 0.03 0.52 -0.64 44 0.55 0.02 0.50 -0.60 49 8.57 0.00
DISCUSSION
Measurements. The search for diagnostic tinction between the two species is by
characters to differentiate L. forcipatus Garman (1917), who illustrated the longer
males from those ofL. disjunctus is not a ovipositor in L. forcipatus. Montgomery
new one. According to Donnelly (2003), (1941) noted the widespread confusion
the earliest paper that demonstrates a dis- between the species and cited four diag-
J. Entomol. Soc. Brit.Columbia 101,Dhchmbbr2004 137
Table 2.
Percentage ofspecimens ofLestesforcipatus and L. disjimctus displaying selected patterns of
pruinescence on thepterothorax. Figs. 5 and 6 illustrate thepatterns.
L.forcipatus L. disjunctus
% %
Pruinositypattern (n=40) (n=30)
Lateral-ventral-dorsal 72.5 0
Dorso-lateral 20 0
Complete lateral 7.5 60
Mid lateral 0 30
Low lateral 0 10
Absent 0 0
nostic differences: (1) distance between that it is best to use a combination ofchar-
teeth ofcercus; (2) width ofapical hood of acters for identification. In western North
abdominal segment 10; (3) width of base America, at least, both morphology and the
of membranous shield of sperm vesicle pattern of pruinescence should be consid-
(penis vesicle), and (4) shape ofpenis (see ered. A short review of useful characters
Donnelly 2003). When Walker (1952) re- and characterratios follows:
viewed Montgomery's findings, he re- 1. Anteriorlamina (AL). Rather than meas-
jected the shape of the penis, but retained uring the whole length of the lamina
the other three characters. Walker (1952) (including the stalk), we measured the ex-
added the relative lengths of abdominal panded apical blade-like part only. The
segments 2 and 3 and the length of the lamina in L. forcipatus is longer (mean =
anterior lamina (see Donnelly 2003). 0.91 mm) than that ofL. disjunctus (mean
Westfall and May (1996) also base their = 0.72 mm). The ranges of the lengths of
separation of the species on the relative the AL overlap in the two species, but the
lengths ofabdominal segments 2 and 3, but length in L. disjimctus does not exceed 1
added the distance between the tip of the mm, while that of L. forcipatus reaches
basal tooth and the swelling near its base, 1.20 mm. We found the lamina to be sig-
the shape ofthe membranous shield ofthe nificantly different in three character ratios
sperm vesicle and the relative size of the - those using the head width, the length of
cereal teeth. segment 2 and the length ofsegment 3.
Donnelly's (2003) findings are different 2. Base of apical tooth to base of basal
again. He stressed the use of the anterior tooth (AB). AB is a good identification
lamina length, the distance between the character as a simple measurement or as a
apical and basal teeth on the cercus, the ratio with head width (Table 1). The dis-
shape ofthe paraproct and the apical hood tance between the teeth is longer in L.for-
width on segment 10. He preferred not to cipatus than in L. disjunctus; this result is
use the membranous shield, the relative supported by Donnelly (2003). Although
lengths of abdominal segments 2 and 3, there is some overlap in the measurements
and the distance from the basal swelling of of the two species {L. disjunctus, 0.33 -
the cercus to the tip ofthe basal tooth. Cat- 0.53 mm; L. forcipatus,OAl - 0.60 mm),
ling's (2002) usefiil study ofOntario mate- the character is useful when used in con-
rial concluded that the best characters were junction with others.
the relative heights ofthe apical and basal 1. Width oftheapicalhood(HL)
teeth of the cercus and the relative extent The ranges ofapical hood widths over-
of pale and dark pigment (not pruines- lapped in the two species ~ L. disjunctus
cence) on the thorax. (0.33 - 0.53 mm) and L.forcipatus (0.40 -
Our findings support the conclusion 0.67 mm). The HL is generally greater in
138 J.Entomol. Soc.Brit.Columbia 101,December2004
L. forcipatiis, which gives the apical hood segments 3, 6, and 7. Patterns on abdomi-
the wide, low appearance (as opposed to nal segments other than segment 2 are not
the pinched shape in L. disjunctus) that is useful in identification because they are
often used to distinguish the species almost identical in shape, intensity and
(Donnelly, 2003, Lam 2004). Based on our frequency of occurrence in both species.
data, this is a generalization and is not reli- Abdominal segment 2 however, is reliable
able for differentiating the species. The HL in differentiatingL. disjunctus andL.forci-
is useful when used in ratios usingthehead patus (Table 2.). Although 23% ofL. dis-
andabdomen. junctus appear to have a clear patch at the
4. Width ofthe head (HD). There was no apex of this segment it has, upon closer
significant difference between the species inspection, not a clearly defined rectangu-
in the width of the head. We used the lar shape but an asymmetrical shape with
measurement to calculateratios. some pruinscence throughout. There was
5. Laterallengths ofabdominalsegments 2 little individual variation in the position of
and 3. There was a significant difference pruinescence in either L. forcipatus or L.
between the length of segment 2 in both disjunctus.
species; however, the ranges overlapped Conclusions. Even with careful analysis of
considerably. Segment 3 was not different each character, a specimen lacking pru-
between species but the relative lengths of inescence is difficult to identify. As a gen-
segments 2 and 3 were significant. eral rule, a specimen with longer or wider
6. Width ofthe abdomen (WA). This meas- measurements than the average L. disjunc-
urement is significant only when used in a tus specimen should be regarded as a po-
ratio with measurements of the apical tential L. forcipatus. The most worthwhile
hood. Comparing species using this char- characters to choose for identification are
acter is difficult as the ranges overlap the AB (the distance between the base of
greatly. the apical tooth and the base of the basal
Pruinescence. The literature from eastern tooth of the circus), AL (the length ofthe
North America, where L. forcipatus has blade ofthe anterior lamina), HL (the basal
been studied for decades, does not mention width ofthe apical hood), and S2 (the lat-
pruinescence as a basis for separating L. eral length of abdominal segment 2). In
disjunctus and L.forcipatus (Walker 1952, each, the mean distance is higher in L.for-
1953, Westfall and May 1996, Catling cipatus and, although ranges overlap con-
2002, Donnelly 2003, Lam 2004). In that siderably, the range exceeds that ofL. dis-
region, pruinescence patterns are appar- junctus.
ently different from those in northwestern The most useful ratios are the above
North America and are oflittle use in spe- measurements divided by the head width
cies identification. On the other hand, as (AB/HD, AL/HD and HL/HD). In AB/HD
was originally noted in Washington State the ranges ofthe two species overlap mini-
by Dennis Paulson, (D.R. Paulson, Slater mally compared to those of the other sig-
Museum, University of Puget Sound, Ta- nificant ratios. In the remaining two ratios
coma; pers. comm.), in far western North the range ofL. forcipatus far exceeds that
America, pruinescence in mature individu- oiL. disjunctus.
als seems a good character for separating In our study, any specimen with pru-
the species. It has the advantage of being inescence on the dorsum ofthe pterothorax
easy to use in the fieldwithout evenhaving (mesepistemal stripe plus midline) is L.
to capture the specimen. Further study of forcipatus, and the species showed this
these patterns over the whole range ofthe trait in over 90% of the specimens exam-
two species is required. ined. L. forcipatus never had only low lat-
Maturity is accompanied by pruines- eral or mid lateral pruinescence, a common
cence on abdominal segments 2, 8, 9, and pattern in L. disjunctus, and showed com-
10, and to a lesser degree on abdominal plete lateral coverage (without any dorsal
J. Entomol. Soc. Brit.Columbia 101,Dechmbhr2004 139
pruinescence) only 7.5% ofthe time, com- ment 2 was L. forcipatus. The segment in
pared to 60% of L. disjimctus specimens. L. disjimctus was usually completely pru-
Any specimen with a strongly differenti- inescent; about a quarter ofthe time it was
ated, symmetrical, pruinescent-free patch marked with an irregular, lightly pruines-
apically on the dorsum of abdominal seg- centpatch.
ACKNOWLEDGEMENTS
Dennis Paulson (Slater Museum of mens. Dennis Paulson gave useful advice
Natural History, Tacoma), Karen Needham concerning pruinescence patterns. Richard
and Rex Kenner (Spencer Entomological Ring read an early draft ofthe manuscript.
Museum, UBC, Vancouver) loaned speci-
REFERENCES
Cannings, R.A. 2002. Introducing the Dragonflies ofBritish Columbia and the Yukon. Royal British Co-
lumbiaMuseum,Victoria, BC.
Catling, P.M. 2002. An evaluation ofsome characters separating maletestes disjimctus andtestesforci-
patusinOntario.OntarioOdonata3: 51-58.
Corbet, P.S. 1999. Dragonflies: BehaviourandEcologyofOdonata. CornellUniversityPress. Ithaca,NY.
Donnelly, T.W. 2003.testesdisjunctus,forcipatusandaustralis: aconfusing complexofNorthAmerican
damselflies. Argia 15(3): 10-13.
Garman, P. 1917. The Zygoptera or damsel-flies ofIllinois. Bulletin ofthe Illinois State Laboratory of
Natural History 12(4): 411-587.
Garrison, R.W. 1976. Multivariate analysis of geographic variation in tibeUula luctuosa Burmeister
(Odonata: Libellulidae). Pan-Pacific Entomologist52: 181-203.
Gorb, S.N. 1994. Scanningelectronmicroscopyofpruinosity inOdonata. Odonatologica24(2): 225-228.
Jacobs,M.E. 1955. Studiesonterritorialismandsexual selectionindragonflies. Ecology36: 566-586.
Lam,E. 2004. DamselfliesoftheNortheast. Biodiversity Books,ForestHills,NY.
Montgomery, B.E. 1941. RecordsofIndianadragonflies. X. 1937-1940. Proceedingsofthe IndianaAcad-
emyofSciences50: 229-241.
Paulson, D.R. 1983. A new species of dragonfly, Gomphus (Gomphurus) lynnae spec, nov., from the
Yakima River, Washington, with notes on pruinosity in Gomphidae (Anisoptera). Odonatologica 12
(1): 59-70.
Paulson, D.R. 2004. Western US Odonata range maps: testesforcipatus. http://www.ups.edu/biology/
museum/westemOD.html
Ramsay, L.R. and S.G. Cannings. 2000. Dragonflies at Risk in British Columbia. Vol. I, pp. 89-93. In L.
Darling (ed.). At Risk: Proceedings ofa Conference on the Biology and Management ofSpecies and
Habitats at Risk, Kamloops, BC, 15-19 February 1999. Ministry ofEnvironment, Lands and Parks,
Victoria, BC.
Stinson, C. and M. Dodge. 2004. Microsoft Office Excel 2003: Inside Out. Microsoft Press. Redmond,
WA.
Walker, E.M. 1952. testes disjunctus ^nd forcipatus complex (Odonata: Lestidae). Transactions ofthe
AmericanEntomological Society88: 59-74.
Walker, E.M. 1953. The Odonata ofCanada and Alaska. Vol. 1. University ofToronto Press, Toronto,
ON.
Westfall,M.J.,Jr.,andM.L. May. 1996. DamselfliesofNorthAmerica. Scientific Publishers, Inc. Gaines-
ville, FL.
J.Entomol. Soc.Brit.Columbia 101,December2004