Title Page
 Table of Contents
 List of Tables
 List of Figures
 Description of study sites
 Materials and methods
 Reproductive behavior
 Reproductive isolation
 Summary and conclusions
 Biographical sketch

Title: Reproductive behavior and isolation of two sympatric coenagrionid damselflies in Florida /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098161/00001
 Material Information
Title: Reproductive behavior and isolation of two sympatric coenagrionid damselflies in Florida /
Physical Description: viii, 79 leaves : ill. ; 28 cm.
Language: English
Creator: Tennessen, Kenneth Joseph, 1946-
Publication Date: 1975
Copyright Date: 1975
Subject: Damselflies   ( lcsh )
Entomology and Nematology thesis Ph. D
Dissertations, Academic -- Entomology and Nematology -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 73-78.
Statement of Responsibility: by Kenneth Joseph Tennessen.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098161
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000408109
oclc - 02163137
notis - ACF4522


This item has the following downloads:

PDF ( 3 MBs ) ( PDF )

Table of Contents
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
    List of Tables
        Page iv
    List of Figures
        Page v
        Page vi
        Page vii
        Page viii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Description of study sites
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
    Materials and methods
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    Reproductive behavior
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
    Reproductive isolation
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
    Summary and conclusions
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
    Biographical sketch
        Page 79
        Page 80
        Page 81
        Page 82
Full Text








I sincerely thank the members of my supervisory committee for

their time and assistance during this study. Dr. Dale H. Habeck

served as committee chairman and provided advice and access to facili-

ties. Dr. Philip Callahan stimulated interest in insect communication

and sense receptors. Dr. Lewis Berner provided equipment for close-up

photography and offered advice and criticism. Dr. F. C. Johnson helped

in the choice of the species studied and made many helpful comments.

My inspiration and opportunity to study Odonata came through Dr. Minter

J. Westfall, Jr., who served as cochairman and unhesitantly provided

advice, materials, facilities and help with field work. He has been a

teacher and a friend.

Special thanks and appreciation are extended to T. C. Carlysle

for the scanning electron microscopy, histological preparations, and

advice. I thank Dr. James L. Nation for discussions of insect cells

and their functions. Discussions with Kenneth W. Knopf and his sug-

gestions are sincerely appreciated. I thank Mr. Paul Laessle, staff

artist of the Department of Zoology, for equipment and advice in prepa-

ration of figures.

To my wife, Sandi, I extend sincere appreciation and thanks for

encouragement, understanding, and the late dinners.



ACKNOWLEDGEMENTS .. . . . . . . ii

LIST OF TABLES ... . . . . . . iv

LIST OF FIGURES . . . . ......... v

ABSTRACT . . . . . . . vii

INTRODUCTION .. . . ... ... . . . 1


MATERIALS AND METHODS . . . ... . . . 18


The Pre-reproductive Period . . . . .. 23
The Reproductive Period . . .... .. 26



REFERENCES ....... . . . . .. . 73

BIOGRAPHICAL SKETCH . . . . . . . 79


Table Page

1 Maturation period (in days) of Enallagma pollutum
and E. signatum, determined by rearing and
marking-return experiments .. . . . ... .25

2 Number of Enallagma pollutum males attempting tan-
dem with simultaneously presented females with
straight or decurved abdominal postures . . .. .47

3 Response of Enallagma pollutum and E. signatum
males to 900 presentations of conspecific and
heterospecific females . . . . .. 52

4 Number of positive responses by males to females
of both species when presented simultaneously . . 54


Figure Page

1 Known distributions of Enallagma pollutum
and E. signatum in Florida - .. ...... 4

2 Map of Hatchet Creek drainage northeast of
Gainesville; study sites indicated by arrows .. 14

3 Study site on Hatchet Creek, 70 m west of
bridge on Hwy. S-225 . . . . ... 16

4 Study site on Hatchet Creek, 50 m north
of bridge on Hwy. 26 . . . . ... 16

5 Lateral view of male apical abdominal append-
ages (40X) (E. pollutum above, E. signatum
below) showing superior appendage shape used
for field identification . . . . 19

6 Seasonal change in onset of daily reproductive
activity in males of E. pollutum and E. signatum 28

7 Diurnal fluctuation in number of unpaired males
of E. pollutum and E. signatum in a 10 m stretch
of Hatchet Creek, 21 May 1975 . . . .. 30

8 Orientation of a confronting E. signatum male
toward a hovering male, and their ensuing
flight paths . . ... ... . ... . 32

9 Sequence of events in mating behavior of E.
pollutum and E. signatum: a) male landing on
female; b) male positioning abdominal appendages
on thorax; c) tandem flight to a perch; d) "abdo-
minal probing" by female; e) transfer of sperm;
f) copulation . .. . . . .. . 40

10 Lateral view of E. pollutum male appendages and
corresponding points of contact with female thorax 41

11 Lateral view of E. signatum male appendages and
corresponding points of contact with female thorax 41


Figure Page

12 Appendages of E. pollutum male clasping thorax
of female in tandem 28X) . . . . 42

13 Appendages of E. si natum male clasping thorax
of female in tandem 30X) . . . . ... .42

14 Perched unreceptive female of E. pollutum with
abdomen decurved and wings spread in response
to an approaching male . . . . ... .. .45

15 Schematic account of mating sequence in
E. pollutum and E. signatum . . .... ..... 49

16 Oblique section through male superior appendage
of E. pollutum; arrow indicates specialized se-
cretory cells (pulled away during fixation) be-
neath the thickened, unsclerotized cuticle (500X) 60

17 Scanning electron micrograph of thorax of an
E. pollutum female showing sensory receptors (10OX). .63

18 Scanning electron micrograph of thorax of an
E. signatum female showing sensory receptors (100X). 63

19 Scanning electron micrograph of hind pronotal
lobe (left side, posterior view) of an E. pollutum
female (215X) . . . . . . . 64

20 Scanning electron micrograph of hind pronotal
lobe (left side, posterior view) of an E. signatum
female (215X) . . . . .. . . 64

21 Scanning electron micrograph of E. pollutum
female mesostigmal plate (170X) .. . .. 65

22 Scanning electron micrograph of E. signatum
female mesostigmal plate (170X) . . . . 65

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy



Kenneth Joseph Tennessen

August, 1975

Chairman: D. H. Habeck
Cochairman: M. J. Westfall, Jr.
Major Department: Entomology and Nematology

The coenagrionid damselflies Enallagma pollutum (Hagen) and E.

signatum (Hagen) are sympatric throughout most of Florida and may be

called sibling species. Reproductive activity at water begins early

in the afternoon in spring and fall, but the species are semicrepuscular

in summer. Males usually appear at water before females, and E. pollu-

tum males usually before E. signatum. Behavior and location of indivi-

duals prior to their appearance at water are unknown. Males interact

vigorously in order to maintain a position at the water; prey capture

at water is uncommon. Males and probably females are capable of sex

recognition; the male's orange ninth abdominal segment is an important

sex-recognition character. Pairing occurs only at water, without court-

ship displays. The male flies directly to the female and grasps her

by the thorax with his legs. After clasping her with the abdominal

appendages, the pair flies in tandem to a perch. In response to the

female's "abdominal probing," the male transfers sperm, followed by

copulation which averages 20 minutes. Individuals may mate once in a

day, but rarely mate on two successive days. Oviposition usually occurs

in tandem shortly after copulation, at or below the water surface.

Enallagma pollutum and E. signatum are very effectively isolated.

Flight seasons and daily mating times are concurrent. Partial isola-

tion is gained by differences in microhabitat: E. pollutum males

usually perch whereas E. signatum males usually hover over open water;

E. pollutum females usually remain on or near vegetation, and although

E. signatum females are also found there, they usually fly toward open

water. These differences lessen the probability of interspecific

meetings. However, many exceptions occur and males attempt tandem with

females of either species; 25 percent of the pairings observed were

interspecific. Males are incapable of visually distinguishing females.

Isolation occurs during tandem, as non-conspecific males release their

grasp upon "refusal motions" by the female. Evidence indicates that

tactile stimuli, provided by the shape of the male superior appendages,

are releasers of mating behavior in females, and may isolate the

species; this proposed mechanism is ethological. Males with altered

appendage shape were unsuccessful in mating, and while the observations

are not as controlled as desired, the results suggest a tandem period

of species recognition. The female thoracic parts contacted by the

appendages bear numerous trichoid and other sensillae. The function

of specialized, secretary cells in the male superior appendages is

unknown; chemical isolation is unreported in Odonata, but the possi-

bilities should be investigated.

I propose that the nongenitalic mechanical isolation theory, in

which the male abdominal appendages, because of structure, are sup-

posedly unable to securely hold a heterospecific female, is untenable

for Coenagrionidae. Clearly, ecological and ethological isolation is



The basic premise of the biological species concept is that groups

of natural populations are reproductively isolated. Reproductive iso-

lation is maintained by means of isolating mechanisms, defined by Mayr

(1970) as biological properties of individuals which prevent the inter-

breeding of sympatric populations. Mayr classified isolating mechanisms

as premating or postmating. Premating isolation includes 1) seasonal

and habitat [ecological] differences, 2) ethological differences, by

way of visual, auditory or chemical stimuli, and 3) mechanical differ-

ences. Postmating isolation includes gametic and zygotic mortality

and hybrid inviability and sterility. Premating isolating mechanisms

are more efficient in terms of energy because there is no wastage of

gametes; they are highly susceptible to natural selection. Mayr points

out that the precise mechanisms preventing interbreeding of closely

related species are unknown for the majority of animals. Discovery of

these mechanisms requires knowledge of behavior and identification of

the stimuli important in mate and species recognition.

The early discovery of specific structural differences in insect

genitalia led to the idea of a "lock and key" mechanism in which the

genitalia of a male of one species fail to fit with the genitalia of

a female of another species, thus preventing interspecific hybridiza-

tion. This theory of mechanical isolation in insects has been largely


rejected, beginning with Jordan's work (1905) on Sphingidae. Reasons

for refuting the theory include: 1) in many insect groups possessing

species-specific male genitalic armatures, females are indistinguish-

able on the basis of genitalia; 2) certain species with different

genitalia can interbreed; and 3) altering male genitalia often fails

to prevent successful copulation and fertilization. Few studies

cited as examples of mechanical isolation have withstood critical

analysis. Evidence strongly supporting mechanical isolation in

Melanopline grasshoppers has been given by Cantrall and Cohn (1970)

and in shield-backed katydids by Rentz (1972). Paulson (1974) reviewed

Webb's study (1947) of polygyrid snails and gives reasons why it

should not be considered evidence of mechanical isolation. Paulson,

however, defends the mechanical isolation theory, arguing that contrary

to Mayr's (1970) and Dobzhansky's (1970) reasoning, there is not yet

sufficient evidence to justify considering the great array of variation

in insect genitalia nonadaptive. He cites two studies (Lorkovic, 1958;

Schick, 1965) as valid examples of mechanical isolation in arthropods,

and reviews the possibilities of the phenomenon occurring in the order


Dragonflies and damselflies are large, readily observed and

recognized insects, and are favorable subjects for behavior studies.

Because of the unique copulatory behavior of Odonata, a type of me-

chanical isolation other than incompatible genitalia is possible. The

male clasps the female by the head in Anisoptera (dragonflies) and by

the thorax in Zygoptera damselfliess) with his apical abdominal append-

ages, after which the female bends her abdomen forward to bring her

genitalia in apposition with the male's accessory genitalia of segment

2. This first point of contact by the male abdominal appendages is

nongenitalic and prerequisite for the copulatory position, and is

referred to as the tandem position. The male abdominal appendages are

variously modified in different species, and the corresponding points

of contact in the female are often specifically modified to receive

the male appendages. Corbet (1963) proposed that the precise inter-

locking of male appendages and female head or thorax acts as a release

for the initial stages of copulation. He states, however, that "The

way in which sexual isolation is achieved in any one species [of Odona-

ta] is not known" (p. 182).

The purpose of my study is to determine the type(s) of mechanisms)

isolating the coenagrionid damselflies Enallagma pollutum (Hagen) and

E. signatum (Hagen). These two species are sympatric and very similar

in appearance. Enallagma pollutum occurs throughout Florida, although

not commonly in the northwestern panhandle. Enallagma signatum, of

widespread distribution in the eastern United States, is common in

Florida south to Highlands County. Figure 1 shows the known distribu-

tions in Florida and localities where both species have been collected.

Breeding seasons coincide, lasting from March to November. Mating

pairs can often be found side by side; however, no interspecific copu-

lation nor hybrids are known. Both species are sexually dimorphic.

Adult males are orange and black; the abdomen is black dorsally and

pale orange laterally, except segment 9 is orange. Female E. signatum

are black and light green or blue, whereas female E. pollutum are black

and pale orange; the abdomen is black dorsally and grey or greyish

A 0

-' .

Figure 1. Known distributions of Enallagma pollutum and E. signatum
in Florida. o = pollutum; a = signatum.

orange laterally, except segment 10 is nearly entirely pale. There

appear to be no significant differences in habitat, flight season,

size, or color pattern.

In absence of such differences, some type of ethological or

mechanical difference appeared to be isolating the species. I posed

these questions: 1) what are the details of mating behavior and do

significant differences exist between the two species, 2) are there

sex and species recognition prior to tandem and what stimuli are

important, 3) do interspecific mating attempts occur, and 4) what

mechanism isolates these species, and if more than one is operative,

what is the relative importance of each? Interactions with other

species of damselflies were studied to enhance the understanding of

reproductive isolation in these insects. Status of the knowledge of

Odonata reproductive isolation is reviewed.

Williamson (1906) was probably the first to examine perspectively

the question of what isolating mechanisms are important in Odonata.

He generalized that species with visually distinctive colors or patterns

have similar male abdominal appendages, whereas those species with

similar color patterns have distinctive male appendages. He further

proposed that in the first group visual stimuli are used for conspeci-

fic mate recognition, while "harmonious relationships of parts" and

"complexity of the sexual act" [tactile stimuli] are important in the

The terms sex recognition and species recognition are not meant to
imply cognition by these insects; they refer to certain reactions by
individuals in response to specific stimuli received from other indi-

second group. Williamson considered the patterns used by taxonomists

as releasers for mating, and his implications correspond to ethologi-

cal and mechanical isolation. Despite the advanced alpha and beta

taxonomy of the group, for many years the phenomenon of Odonata re-

productive isolation and Williamson's proposals were all but neglected.

Evidence supporting habitat and seasonal isolation became available

(Kennedy, 1922, 1928), but its significance apparently went unrecog-


Johnson (1962a) was the first to review the problem since

Williamson's paper, calling attention to the importance of additional

mechanisms such as seasonal and ecological isolation, and emphasizing

areas where research was most needed. Three detailed studies on

Anisoptera were discussed, Jacobs (1955), Kormondy (1959) and Johnson

(1962b); the results tend to confirm Williamson's proposal. Studies

contributing to understanding isolation in Zygoptera were those of

Buchholtz (1951, 1955, 1956), Krieger and Krieger-Loibl (1958), Loibl

(1958), Bick and Bick (1961) and Johnson (1961, 1962c), also substan-

tiating Williamson's proposal.

Since Johnson's review, papers on Odonata reproductive behavior

have continued to appear, but often without evidence or even specula-

tion on the isolation of closely related species. Those studies con-

tributing observation or experimental proof for proposing isolating

mechanisms are relatively few. For clarity of their discussion, I am

describing male territoriality, as this behavior is important in

acquiring a mate. Many definitions of and requisites for territorial

behavior in Odonata have been proposed (Moore, 1952, 1957; Jacobs,

1955; Kormondy, 1961; and St. Quentin, 1964). I have followed the

definition of Noble (1939), that is, "any defended area," as it is

simple and implies no particular function. Males exhibiting terri-

toriality recognize and are aggressive toward conspecific males,

resulting in males occupying microareas which are defended and main-

tained by display (intraspecific only) or physical contact (intra-

and inter-specific). The main function of territoriality, clearly

recognized by Johnson (1964), is to prevent other males from inter-

fering with courtship, mating and oviposition. Not all Odonata are


The ability and mode of sex and species recognition determines

the type of isolating mechanism operative. Corbet (1963), in his

discussion of reproductive behavior, distinguished two main types of

visual recognition: 1) unilateral, in which the male recognizes the

female, and 2) reciprocal, in which each sex recognizes the opposite

sex. These types of recognition, however, are complicated by the

species recognition ability of the males. For instance, if males of

species A, besides merely recognizing females, can recognize females of

species A (i.e., attempt to copulate only with conspecific females),

then some type of ethological mechanism is functional. Possible

mechanisms include courtship displays, distinctive wing or body colors

or color patterns, or distinctive flight behavior, or some combination

of these factors. On the contrary, if males are unable to distinguish

females of species A from those of species B (plus others if present),

then attempts to copulate with heterospecific females can be expected;

it is then plausible that some type of contact ethological or mechanical

isolation is functional. There is no published mention of auditory

or chemical stimuli being used in odonate sex or species recognition.

Despite the growing literature, knowledge of reproductive iso-

lation in Anisoptera is still scanty. A study not discussed by Johnson

(1962a) is that of Ito (1960) on the libellulid Orthetrum albistylus

speciosum Uhler. The abdomen of a mature male is white dorsally, that

of a female black with yellow spots. Males are aggressive toward other

males, but females are seized and taken into copulation. Ito observed

one female with a "male-type" abdomen (white dorsum) that was seized

by a male without aggression. He concluded that sex recognition is

facilitated not only by color pattern but also by some behavioral

difference. The white abdomen may merely be a sign stimulus aiding

males to maintain their territories.

Corbet (1963) briefly reviewed the complexity found in the shape

of male appendages of Petaluridae, Aeshnidae, Gomphidae and some

Corduliidae, and the corresponding modifications in the female head

or thorax which they clasp. He proposed that, in addition to possibly

helping secure the tandem bond during erratic flight as suggested by

Walker (1912, 1915), these complex structures may act as mechanical

isolating mechanisms, but he lacked experimental proof. No experi-

mental studies on mechanical isolation in these families have since

appeared. Corbet also suggested mechanical isolation as a function

of the male oreillets, but this has remained uninvestigated. Neville

(1959a) observed an Anax male in copulation with an Aeshna female,

and proposed that intergeneric isolation is not effected by genitalic


I _

Visual isolation was found by Pajunen (1964) to be functional in

the libellulids Leucorrhinia dubia van der Linden and L. rubicunda

Linnaeus, sympatric species very similar in color pattern and behav-

ior. He concluded that isolation is based on "mutual species recog-

nition of the sexes," in which males recognize females by abdominal

size differences and females appear to recognize males by undeter-

mined sight stimuli prior to tandem attempts. He further suggested

that form discrimination may be important in other Anisoptera. Male

appendages of the two species are slightly different in measurements

but similar in structure and shape. Altering appendage shape did

not affect the males' ability to clasp females and successfully co-

pulate. Hence Pajunen concluded absence of mechanical isolation.

Currie (1963), in observing the libellulid Erythemis simplicicollis

(Say), found no evidence of courtship behavior. Mature males are

blue and females green, so visual sex recognition is certainly possi-

ble. Females are seized almost immediately upon arrival at a pond,

but only when in moving flight, indicating a behavioral stimulus as

suggested by Ito.

Watson (1966) proposed that mechanical isolation may operate

during copulation due to specific structural differences in genitalia.

He based his hypothesis on correlations between male hamule length

and female vulvar scale length in Indo-Australian species of

Trapezostigma. That the hamules contact the vulvar scale during

copulation, however, is unclarified. Johnson (1972) contested Watson's

proposal, presenting evidence that the hamules in the gomphid Hagenius

brevistylus Selys support the erected penis during sperm transfer and

copulation. He concluded that the functions of the genitalic struc-

tures in Anisoptera need to be reinvestigated. Unfortunately, stud-

ies of Anisoptera in which mechanical isolation is likely, especially

species of Aeshnidae and Gomphidae with complex and distinctive male

appendages, have not appeared.

Temporal isolation is apparently rare. Paulson (1973) presented

convincing evidence that the corduliids Epitheca sepia (Gloyd) and

E. stella (Williamson) are isolated by mating at different times of

the day.

Knowledge of reproductive isolation in Zygoptera is more advanced

than in Anisoptera, probably because damselflies are more amenable to

behavioral studies. Evidence of ethological and mechanical isolation

has been reported. Studies on Calopterygidae by Buchholtz (1951,

1955), Johnson (1961, 1962c), Bick and Sulzbach (1966), Heymer (1973)

and Waage (1973), have shown that courtship displays and wing color

patterns are important in sex and species recognition. Male append-

ages are similar in size and shape, but I found no report of inter-

specific mating attempts, indicating a high degree of efficiency of

visual isolation. Courtship behavior and other evidence of visual

isolation is not commonly found in other families. Tillyard (1913)

described the unusual male and female courtship behavior of Hemiphle-

bia mirabilis Selys; the intriguing problem of the complex behavior

and Fraser's (1957) phylogenetically ancestral placement is discussed

by Johnson (1962a). Consiglio (1974) briefly described courtship in

a species of Chlorocyphidae, but did not mention the tibial display

reported by Fraser (1949). Buchholtz (1956) found a simple male

courtship in two species of Platycnemididae. Males supposedly recog-

nize conspecific females by the color tone of light areas on the head

and thorax. Heymer (1966) described a simple male courtship in two

other platycnemidids.

The only mention of courtship in Coenagrionidae is by Patrick

and Lutz (1969), who state for Ischnura posita (Hagen) that "Court-

ship behavior was observed prior to mating", but no details are given.

Courtship is unknown also in Lestidae. The majority of coenagrionids

and lestids are clear-winged, and many pairs of sympatric species are

very similar in appearance. Loibl (1958) gave evidence that females

of Lestes recognize males by tactile stimulation of the male append-

ages which are distinctively shaped. Males attempt tandem with

heterospecific females (plus almost any approaching zygopteran), but

these females refuse to copulate. Furthermore, females refuse to

copulate with conspecific males having experimentally altered infe-

riorappendages, indicating mechanical or tactile isolation. Bick and

Bick (1961) and Bick and Hornuff (1965) observed no courtship in two

species of Lestes, and the former authors noted one mixed pair in


Krieger and Krieger-Loibl (1958) found that males of the

coenagrionid Ischnura elegans (van der Linden) respond sexually

toward females of I. pumilio (Charpentier) only a fourth as often as

to conspecific females. Size differences appear to be important in

visual recognition. Male appendages of the two species are different,

and females refuse copulation with males whose appendages have been

altered. Bick and Bick (1972) reported absence of courtship and

display in two Argia species, but did not speculate on reproductive

isolation. Bick and Bick (1963) and Bick and Hornuff (1966) found

no courtship or display in three species of the coenagrionid genus


Paulson (1974) observed and experimented with five species of

Argia and Enallagma in the western United States. He found that in

general males do not distinguish between different species of con-

generic females, but do distinguish and hence avoid non-congeneric

females, probably because of size differences. He proposed that in

most pairs of similar species some difficulties in fitting of male

appendages with the female thorax prevents successful mixed tandem,

and concluded that mechanical isolation is clearly operative.


The majority of observations and experiments were conducted at

two sites on the Hatchet Creek drainage, northeast of Gainesville,

Alachua County, Florida. Locations of these sites, near Hwy. S-225

and Hwy. 26, are shown in Figure 2. Hatchet Creek originates in a

swamp about 2.5 miles north of the city, drains northward for nearly

2 miles, then eastward for about 4.5 miles, and finally southward to

Newnan's Lake. The area is mostly flat, elevation gradually decreas-

ing from 170 feet at the northern origin to about 75 feet near New-

nan's Lake. The creek is mostly inaccessible because of swamp, lack

of roads, and the creek splitting into many small channels amongst

cypress and heavy vegetation. Depth and flow rate fluctuate greatly,

depending on rainfall. The creek is usually shallow and slow from

October to May, a period of relatively low rainfall. Greater depths

and high rates of flow persist from June to August, a period of

higher rainfall. During this latter period, the swamp is flushed

of much organic matter, greatly increasing the brown color of the


The study site near Hwy. S-225 (R 20 E T 9 S, Section 2,

southeast quadrant) is from 30 to 70 m southwest of the bridge. Maxi-

mum width is about 6 m, depth varies from 15 to 90 cm. The substrate

is mostly fine sand, although the edges consist of an overlying silt,












w w





organic ooze and detritus layer. At this site, the creek best fits

Beck's (1965) sand-bottomed stream type; in certain places, however,

it could be classified as a swamp-and-bog stream. The rate of flow

is usually moderate, at times slow; the water is slightly acid and

usually moderately colored. Vegetation is very plentiful (Figure 3),

but the species vary in extent and location with season and year. The

dominant submerged plants are the pondweed Potamogeton capillaceus,

and bladderwort, Utricularia inflata. Frogbit, Linnobium spongia,

floats thickly near the edges, and in 1974 covered the entire creek

for a 15 m stretch. Near the edges occur Juncus spp., Sagittaria sp.,

Polygonum sp. and pickerel weed, Pontedaria lanceolata. Panicum is

common along the north bank, with occasional Sparganium americanum

and Orontium aquaticum. In summer, duckweed, Lemna, covers much of

the water surface. Sphagnum moss is common upstream. Button Bush,

Cephalanthus occidentalis, and hazel alder, Alnus serrulata, grow

thickly along the banks, intertwined by the wild grape vine, Vitis

rotundifolia. At least 18 species of coenagrionid damselflies occur

here: Anomalagrion hastatum (Say), Argia fumipennis atra Gloyd,

A. sedula (Hagen), A. tibialis (Rambur), Enallagma cardenium Selys,

E. concisum Williamson, E. daeckii (Calvert), E. divagans Selys,

E. dubium Root, E. geminatum Kellicott, E. pollutum (Hagen), E.

signatum (Hagen), E. weewa Byers, Ischnura posita (Hagen), I. progna-

tha (Hagen), I. ramburii (Selys), Nehalennia integricollis Calvert,

and Telebasis byersi Westfall.

Figure 3. Study site on Hatchet Creek,
Hwy. S-225.

Figure 4. Study site on Hatchet Creek,
Hwy. 26.

70 m west of bridge on

50 m north of bridge on


The study site near Hwy. 26 (R 21 E T 9 S, Section 21, south-

east quadrant) is from 50 to 70 m north of the bridge. This site is

off the main flow of the creek and is pond-like, although there is

some flow, especially after heavy rains. This "pond" is from 4 to

20 m wide; depth varies from 20 to 120 cm, and fluctuates with rain-

fall more than at the Hwy. S-225 site. The substrate is similar to

that at Hwy. S-225, although there is a greater buildup of mud and

detritus. Vegetation is more sparse (Figure 4). Southern naiad,

Naias, dominates the narrow, upper end of the pond where the studies

were conducted; bladderwort is also present. The edges support

Eleocharis sp., Polygonum sp., Ludwigia sp. and some Potamogeton.

Button Bush is common along the banks. Twelve species of coenagrionids

are present: Argia tibialis, Enallagma cardenium, E. concisum, E.

divagans, E. doubledayi (Selys), E. dubium, E. pallidum Root, E.

pollutum, E. signatum, E. weewa, Ischnura posita and I. ramburii.

Observations were conducted also at the east shore of Newnan's

Lake, where sizeable populations of both species occur.


Males of Enallagma pollutum and E. signatum are impossible to

distinguish in the field without close examination. Capture was

necessary for positive identification after a particular behavioral

sequence was observed. Individuals were identified with a hand lens

based on the shape of the superior abdominal appendages, which are

short and subtriangular in E. pollutum and long and cylindrical in

E. signatum (Figure 5). Females can usually be identified by

thoracic color; I verified each determination by the location of the

dorsal prothoracic pits, which are near the middle in E. pollutum

and anterior in E. signatum.

Observations of behavior were conducted with aid of binoculars

from partly concealed positions along the bank. Certain inquiries

required close inspection, and at times the damselflies were dis-

turbed. However, these species are not particularly wary of a human

observer; perched E. pollutum can often be touched if approached

slowly. Abnormalities in behavior because of my presence were not

apparent and probably occurred infrequently.

Notes were usually recorded with a portable tape recorder.

Daily high and low temperatures were noted along with cloud conditions

and wind velocities. Close-up photographs for recording postures were

taken with a 50 mm Macro-Rokkor lens on a Minolta SRT 101 35 mm camera

Figure 5. Lateral view of male apical abdominal appendages (40X)
(E. pollutum above, E. signatum below) showing shape
of superior appendages used for field identification.

and high-speed Ektachrome film (ASA 160). An 8 mm Bolex movie camera

with a 30 mm macro-zoom lens was used to film certain behavioral se-

quences. A portable video tape recorder, the Video Rover 3400, was

found unsatisfactory because the images were small and unsharp, pre-

venting species identification; in addition, the equipment was cum-

bersome. Certain behavioral postures and patterns were sketched in

the field and later redrawn; others were drawn from still or movie


Enamel paints were used to mark individuals. Tenerals were

marked on the thoracic dorsum, mature individuals usually on the

thorax or abdomen. Marks were made just large enough to enable re-

cognition without capture or undue disturbance. In preliminary

trials, wing-marked individuals could not be found after only two

to four days at most, and this technique was abandoned. Marked

individuals otherwise appeared to be generally unaffected by the

presence of a mark, and were often seen successful in male inter-

actions and in securing a mate.

Duration of the pre-reproductive period was determined by

comparing captive individuals with mark-return data. Emerging

individuals were kept in screened cages at the study site and sup-

plied with small flying insects. Sexual maturity was assumed when

individuals turned from general blue to the orange color character-

istic of mating individuals. In mark-return experiments, newly-

emerged tenerals were characteristically marked each day and their

first return to the reproductive site as orange individuals was


In visual sex-recognition experiments, individuals were altered

in certain aspects and presented as models to individuals in their

natural habitat. For example, females without abdomens were presented

to males; a high percentage of positive responses (flight towards and

grasping the model) would indicate that the abdomen is unimportant for

recognition of a female. Unaltered females were presented to the same

males, whenever possible, as controls. These experiments are discussed

in more detail along with results in the reproductive behavior section.

The function of certain behavioral postures was determined by use of

models mounted and fixed so as to resemble the natural postures.

In species-recognition experiments, models attached to a 90 cm

black wire were presented to active males. An insect pin secured to

the tip of the wire was inserted into the venter of the model's thorax.

This method allowed presentation of models with a minimum of human

interference, and males were able to land on the model and attempt

tandem positioning. In the latter respect, the fishing-line techni-

ques of St. Quentin (1934) and Moore (1952) were impractical.

For examination of sense receptors and ultrastructure, the Kent

Cambridge Mark II A scanning electron microscope at the Insect Attrac-

tants, Behavior and Basic Biology Research Laboratory of the United

States Department of Agriculture in Gainesville was used, operated

by T. C. Carlysle. Specimens were treated in acetone and coated with

a 200 A layer of gold in a Denton DV-502 high vacuum evaporator

(5 x 10-2 torr). Micrographs were recorded on Polaroid PN/55 film.

Apices of male abdomens were prepared for histological section-

ing by fixing them in Bouin-Duboscq (alcoholic Bouin's) solution for

two days; they were washed and dehydrated in ethanol, and then placed

in methylbenzoate. Afterwards they were transferred to methylbenzoate-

benzene solutions, washed with benzene and embedded with paraffin, sec-

tioned and stained with Mallory's triple stain.

For statistical analysis, the data in Table 3 were arranged into

separate contingency tables for each species as follows:

Male response

pollutum a b

signatum c d

Chi-square was calculated by the following formula, with one degree of


X2 N(ad-bc)2
X a+F (c-dT a+c) (b+dT"

The data in Table 4 were considered two-cell tables and chi-square

calculated as ( E)2 with one degree of freedom.


Studies of the reproductive behavior of Enallagma pollutum and

E. signatum were aimed at discovering the stimuli important in mate-

acquiring, and developing a basis for better understanding the isolat-

ing mechanisms. My discussion will include two stages of adult life,

the pre-reproductive period and the reproductive period. Besides

the reports of Bick and Bick (1963) and Bick and Hornuff (1966),little

information is available for comparison with other enallagmas, but it

appears that E. pollutum and E. signatum are similar in many behavioral

aspects to the other species. Unless differences are specifically

noted, statements apply to both E. pollutum and E. signatum.

The Pre-reproductive Period

The initial phase of adult life, often called the "maturation

period," is a period of feeding and sexual inactivity. During this

time, which varies in duration, the gonads mature and body colors

change (Corbet, 1963). The data available indicate that the maturation

period in coenagrionid damselflies varies from about one to two weeks.

Argia moesta Hagen matures in seven to 14 days (Borror, 1934), and

Pyrrhosoma nymphula (Sulzer) in nine to 15 days (Corbet, 1952).

Newly emerged individuals of E. pollutum and E. signatum are

blue. I have assumed that the subsequent complete change to orange

in males is coincident with sexual maturity. Blue, general males

are rarely found near water after emergence, and were never seen

engaging in sexual activities. Females of E. pollutum become orange

upon maturity, although not as bright orange as males. Females of

E. signatum apparently are polymorphic, becoming green or blue upon

maturity; the majority (about 85 percent) are green. The data in

Table 1 indicate that the maturation period varies from seven to 17

days, and suggest that E. pollutum matures more quickly than E. signa-

tum and males more quickly than females. In males, at least two days

are required from the time the orange first appears until the full,

mature orange is acquired. The face and thoracic dorsum change first,

then the sides of the thorax and lastly abdominal segment 9. High

temperatures shorten the maturation period in Calopteryx (Buchholtz,

1951). My data were obtained in May during very warm weather and

therefore probably represent minimum maturation times.

The activities and location of individuals during the pre-

reproductive period were not entirely determined. The few individuals

found were in thick vegetation away from the water. Evidence that

they had been feeding was the presence of frass when kept captive.

Failure to find many such individuals and the general non-return of

marked, emerging individuals indicates that many leave the habitat

and never return, for reasons unknown. Corbet (1962) reports that

migrating dragonflies may be immature.

Table 1. Maturation period (in days) of Enallagma pollutum and
E. signatum, determined by rearing and marking-return
experiments. A dash indicates insufficient data.

Captive1 Emergence-return2

pollutum males

signatum males

pollutum females

signatum females


Taken emerging, caged and fed at
until attainment of mature color

study site, and time
(orange or green)

Individually marked upon emergence, released, and time
until first return to water as mature individuals

The Reproductive Period

Details of the location and behavior of males and females prior

to their daily appearance at the reproductive site are largely unknown.

Males were occasionally found perching on vegetation within 10 m of

the stream edge. Feeding probably occurs during this time; rarely

were individuals observed feeding at the water, even though prey was


The breeding populations of E. pollutum and E. signatum at the

study sites are not homogeneous in age. Emergence continues through-

out the breeding season, with new individuals continuously joining the

population. The age of nonmarked individuals could not be determined,

although a darkening of the orange color could be seen in recaptured

marked males. Orange males marked and released were seen for the last

time a maximum of 12 days later. Considering the maturation period

data, a maximum male adult life of approximately three to four weeks

is indicated. Density estimates were obtained by counting the males

present in a 10 m stretch of stream at hourly intervals. Maximum

densities, from 15 to 25 individuals of one species per 10 m, occurred

on warm, calm days from late spring through early fall. Usually less

than 15 individuals were encountered on cool or very windy days, and

in early spring and late fall. The number of females seen per 10 m

varied from zero to eight.

The time spent at water by the adults is referred to as the re-

productive activity period. No mating activity was observed away from

water. Mating apparently occurs only at aquatic habitats that are

suitable for oviposition. Because males arrive at water earlier in the

diel cycle than females, it appears they are capable of recognizing

areas containing suitable oviposition sites. Jacobs (1955) reported

that males of Perithemis tenera (Say) actually select the oviposition


Behavior of Single Males

Males are sexually active mainly during the later hours of the

day. Howe (1917) described E. signatum as semicrepuscular in New

England. Male activity periods of both species in Florida vary sea-

sonally in onset and duration (Figure 6). In spring, males are found

active at the breeding site from late morning or early afternoon

until very late afternoon. Throughout this season, onset is progres-

sively delayed. In summer, onset is relatively late, around 1600 hours

or later, activity being concentrated within a one- to two-hour period.

At this point, the species are actually semicrepuscular. In fall,

onset becomes gradually earlier again, reciprocating the spring curve.

Changes in photoperiod and light intensity are probable triggering

factors. Lutz and Pittman (1970) concluded that onset in E. signatum

is not in response to temperature, but to "time of day or perhaps de-

creasing light intensity." Cloud cover affects male activity, espe-

cially in E. signatum. Frequently I observed males suddenly appear

over the water when a cloud obscured the sun, their numbers greatly

diminishing when direct sunlight returned. Onset in E. signatum is

consistently later than in E. pollutum, varying from less than one-half

*. \





* /

o -

o o 0 0 0 0
0 0 0 0 0 0

(I'S'3) AVO iO 3L II








0 0

O -f-

5- *
2 0)

C 0)



hour to two hours later. On several days, a few E. signatum males were

found as early as males of E. pollutum.

During the middle of the flight season, the number of single

males at water usually increased sharply after the onset of activity,

reaching a peak between 1800 and 1900 hours and then dropping off

sharply at the end of the period. This typical diurnal fluctuation

is shown in Figure 7 for 21 May 1975, a nearly cloudless day with a

maximum temperature of 91F. Pairing began around 1730 and continued

until 1900, after which only ovipositing pairs were seen.

After their initial appearance at the water, E. pollutum males

spend most of their time perched on vegetation. Males of E. signatum

either perch near the bank or begin hovering over open water. Sex

recognition is apparent as attempts at tandem with other males were

not seen. The males' orange ninth abdominal segment is an important

sex-recognition character. Males with this segment painted black were

presented to active males; 20 percent took the altered males into tan-

dem, whereas unaltered males presented were not taken into tandem.

Females with an orange spot painted near the tip of the abdomen were

also presented to males; only 33 percent took them into tandem com-

pared to about 80 percent taking unaltered females. In another ex-

periment, the last two abdominal segments of a male and female were

interchanged, and the individuals presented to males; approximately

the same response percentages were obtained. Additional differences

enabling males to recognize sex are probably the brighter coloring of

males and their distinctive flight behavior.



4 N -

z kIw

D F=

ui Ic
ca- -

4- 00

0 0 -

W or
o oo
0 w

0 ui

o o



0 S-


S 3 -1 v o 839vqfN

Males maintain a position at the water, although it is transitory,

by vigorously interacting with other males. These interactions, espe-

cially those intraspecific, increase in frequency within the first hour

at water. Terms used in describing these interactions are defined as

follows. Confrontation refers to the movement of one male directly

toward another; the confronting individual is termed the aggressor.

Confrontation may result in attack. An attack is a rapid flight toward

another male which usually results in a close, vigorous encounter or

physical contact and one of the males being displaced.

Males of E. signatum hover 5 to 20 cm above the water away from

the creek bank, but can also be found perched near the bank. Distance

between hovering individuals depends on population density, and varies

from 30 to 60 cm or more during normal densities (12 to 20 individuals

per 10 m). A male flying into the vicinity (an intruder) of another

male may be confronted and attacked or may itself attack the hovering

male. The attacked male flies backwards and then upwards, attacking

and usually making contact with his aggressor who in turn flies back-

wards and upwards. This sequence is usually repeated one to three

times, resulting in a spiral displacement of the pair (Figure 8).

Many variations of this basic pattern occur. Often they come near a

third hovering male who may join the interaction; at times the attacked

hovering male will not react, and the attacker will simply depart.

Interactions usually last only two or three seconds, the individuals

returning to their original positions within five seconds; usually the

intruding male is displaced. No male mating swarm such as reported for

Enallagma cyathigerum (Charpentier) by Neville (1959b) was observed.



\ o--o

/ O


I ,4


C "O E
n/ E


I I ^w
r~0 S-

Males of E. pollutum are usually perched near the stream edge or

on pondweed or duckweed near the middle. They frequently venture out

over the water and vigorously interact with males of E. signatum.

Interactions between E. pollutum males are similar to those of E. sig-

natum; however, males sometimes perched on the same stem within 5 cm

of each other. Bick and Bick (1963) observed perched, unpaired males

of Enallagma civil (Hagen) spread their wings upon approach of another

male; they called this behavior "wing-warning" and interpreted it as a

threat display. The usual response of perched E. pollutum and E. sig-

natum males is flight toward the intruder; in the four instances of

wing-warning seen, the intruding males departed. I observed wing-

warning more frequently by males in tandem with ovipositing females.

Intruding males were warded away in six of the 18 cases observed; in

12 instances, the intruding male landed on the female and then soon


Bick and Bick (1963) considered Enallagma civil males territo-

rial according to Kormondy's (1961) requisites. Enallagma pollutum

and E. signatum appear to be territorial as defined in my Introduction.

However, the defended area is very transitory, and I believe that males

are merely maintaining a certain distance between themselves and other

males. The advantage of maintaining such a position at water is that

an individual is better able to detect and secure a female.

Behavior of Single Females

I observed the initial approach of females to the reproductive

site infrequently. It appears they fly directly to the stream from the

surrounding vegetation. Once at the water, females either perch or

fly, usually near the stream edge. Females remain perched for rela-

tively short periods, usually less than 10 to 15 minutes. Their sub-

sequent flights are usually short. Females of E. signatum eventually

fly away from the bank and toward open water, whereas females of E.

pollutum usually remain near the bank.

Females do not fly toward any particular male; males detect

females visually when they come within range or by finding them

perched on vegetation. I could not determine whether males actually

search for females in vegetation. Dead, dried females with wings

together were fixed to perches on vegetation; they were detected by

males after 10 to 50 minutes. Models with wings extended and moved

about in mock flight were usually detected almost immediately. Over

80 percent of the natural tandem attempts observed were with flying


In order to elicit a tandem attempt from a male, a female model

must possess a thorax, at least one pair of wings and a head. Although

models without abdomens were taken into tandem, the response was low

(about 25 percent). Males landing on a model without a head would

not attempt tandem.

It appears that females visually recognize males also. When

approached by males, an unreceptive female will decurve her abdomen

and fly evasively. Perched tandem females and perched unreceptive

single females also decurve their abdomen and spread their wings upon

intrusion by a male.

Mating Behavior

The following account of mating is based on observation of 68 pairs

achieving or already in precopulatory tandem, and over 100 pairs in

copulation. The sequence occurs when male and female are conspecific

and both are sexually responsive.

Immediately upon spotting a female, the male flies directly to

her. This flight is very fast, with no detectable display by either

sex. The male usually lands on the female's wings or thorax, crawls

forward and grasps her by the thorax with his legs (Figure 9a). His

fore legs hold onto the sides of her prothorax, the middle legs are

on the sides of the pterothorax with the claws gripping the ventro-

lateral margin, and the hind legs are near the base of the wings with

the claws gripping the ventrolateral margin. During this time, his

mouthparts are actively touching the top of the female's head. The

function of this "biting" is unknown, but may serve to sexually

stimulate the female.

Within two seconds after catching the female, the male arches

his abdomen down and forward, bringing the apex up to the anterior

part of her thorax (Figure 9b). The superior appendages are held

divergent, placed over the mesostigmal plates, and closed like opposing

jaws; the tips of the inferior appendages hook the posterior edge of

the prothoracic pits (Figures 10 and 11). The tip of the E. pollutum

superior appendage extends to the upper edge of the female's black

humeral stripe (Figure 12), whereas that of E. signatum extends to at

least the middle of this stripe (Figure 13). The tips of E. pollutum

superior appendages are 0.5 mm apart "at rest" and 0.8 mm apart when

held at maximum divergence; those of E. signatum are also 0.5 mm apart

at rest, but 1.08 mm apart at maximum divergence. The high mesa

tubercles of the female's mesostigmal plates fit mesally at the base

of the superior appendages. Both the superior and inferior appendages

may act as a pair of forceps, and may also act together as a pair of

forceps, effecting a very firm grip. The dorsoapical emargination of

the male's tenth abdominal segment fits over the middorsal carina (the

specimens in Figures 12 and 13 became slightly separated during trans-

port from field to laboratory).

Contrary to the interpretation of Needham and Heywood (1929) and

Koen (1937), female coenagrionids do not grip the male's appendages

between pro- and ptero-thorax as in a vise. Evidence includes: 1)

males fly in tandem with dead, dried females and release them after

failure to copulate; 2) males in tandem with submerged, ovipositing

females escape before becoming totally submerged themselves; 3) close

examination of tandem pairs with a O1X hand lens revealed that it is

physically impossible for the female thorax to grip or hold the male

appendages because the hind lobe of the prothorax is too small to

effectively press against the superior appendages, and the inferiors

are free of any structure. That the male grips the female concurs

with the findings of Buchholtz (1956) and Bick and Bick (1963).

After the female is securely held by the appendages, the male

releases the grasp with his legs, straightens his abdomen, and begins

to fly with her toward a perch (Figure 9c). This flight is character-

istically smooth but varies in duration and path. For example, the

flight may be straight to a perch, lasting only 10 to 15 seconds; in

others several turns and loops may occur, the flight lasting up to

30 seconds. The latter type often is a result of pursuit by single

males, especially during high male density. When pursued by a male,

the tandem female's abdomen is decurved, whereas it is otherwise

straight. Pursuing males sometimes land on the abdomen of a flying

or perched tandem female; she then sharply deflexes the abdomen,

moving it up and down. The extra male usually leaves within one to

three seconds. Highly persistent males were never observed to sepa-

rate a tandem pair such as reported by Johnson (1961) for the Calop-

terygid Hetaerina americana (Fabricius).

Upon reaching a suitable perch site, the male perches using all

six legs. This site is not within the area in which the male had

originally perched or hovered, but is usually near the stream edge.

The female usually does not perch. Almost immediately after landing

the male pulls the female upward and forward with his abdomen and the

female responds by swinging her abdomen forward from below and bringing

the tip near the male's anterior segments (Figure 9d). I have termed

this behavior "abdominal probing," and found no mention of it in the

literature. It appears to be a release for the male to transfer

sperm, evidenced by the fact that males taking dead presented females

into tandem did not transfer sperm, with one exception. Many of these

males repeatedly pulled the dead female upward and forward with the

abdomen for up to five minutes before finally releasing the appendages.

Attempts to induce sperm transfer in these males by touching the venter

of their anterior abdominal segments were unsuccessful.

After the female's "abdominal probing," the male transfers sperm

to his accessory genitalia on the second segment. Sperm transfer was

never observed in unpaired males, which agrees with the generalization

by Bick and Bick (1965). During sperm transfer, the female's head is

drawn close to the venter of the male's thorax (Figure 9e). Whether

the female receives releasing stimuli for copulation in this position

is unknown. Transfer of sperm requires 10 to 22 seconds.

Upon completion of sperm transfer, the male starts to slowly

straighten his abdomen, moving the female backward. Five to 12 seconds

later he arches his abdomen upward in short, abrupt movements, followed

by the female bending her abdomen forward and upward from beneath. The

male continues to arch his abdomen, drawing the tip of the female's

abdomen closer to his accessory genitalia. She simultaneously attempts

to place her genitalia in apposition to his. Occasionally the female

requires as many as 15 attempts to engage the genitalia, because the

tip of her abdomen passes by the sides of the male's narrow abdomen.

When the genitalia are engaged, the pair is in the copulatory position

("copulatory wheel" of Corbet, 1963, and shown in Figure 9f). If the

pair is perched vertically, the female's legs are usually drawn close

to the thorax and do not contact the substrate; if the perch is hori-

zontal, her middle and hind legs may hold onto the substrate with her

abdomen passing between them. The female's wings are held close to-

gether, the male's are half-spread.

During copulation, the female intermittently thrusts her abdomen

against the anterior segments of the male in almost rhythmic pulses.

The ventral spine of her eighth segment is pressed against the anterior

Figure 9. Sequence of events in mating behavior of Enallagma
pollutum and E. signatum: a) male landing on female;
b) male positioning abdominal appendages on thorax;
c) tandem flight to a perch; d) "abdominal probing"
by female; e) transfer of sperm; f) copulation.

'~b I

& /1

a K" b3 -

liii 1

ci d

Figure 10. Lateral view of E. pollutum male appendages and the
corresponding points of contact with female thorax.

Figure 11. Lateral view of E. signatum male appendages and the
corresponding points of contact with female thorax.

*i /


Figure 12. Appendages of E. ollutum male clasping thorax
of female in tandem (28X



Figure 13. Appendages of E. signatum male clasping thorax
of female in tandem (30X)

ventral lobe of the male's third segment, and is withdrawn and reinserted

with each thrust. The male flexes his abdomen at irregular intervals,

repeatedly altering his abdominal arch. Both individuals rub their eyes

and other parts of the head with their forelegs, similar to the movements

assumed to be cleaning in Platycnemis pennipes (Pallas) by Buchholtz

(1956). Time in copulation ranged from 15 to 24 minutes, with a mean

of 20.4 in E. pollutum (n=16) and 19.7 in E. signatum (n=12). No inter-

ruptions were seen as described by Bick and Bick (1963) for E. civil.

Release of the genitalia is abrupt, after which the female slowly

lowers her abdomen and the male straightens his. The pair may remain

stationary or make short flights to other perches. During this time,

the female will not remain unperched. If the male settles on a site

where the female cannot touch the substrate, she continues beating her

wings until the male flies again and lands on a suitable perch. Often

the female will perch after a short flight, leaving the male upright

and unperched, supported only by his abdomen. The male's wings are now

held together, the female's slightly apart. Four to eight minutes

after copulation, the pair flies in tandem to vegetation near the water

surface, after which oviposition takes place. The majority of eggs are

deposited in stems of pondweed and bladderwort.

Nonreceptive Females

Six instances of precopulatory tandem behavior differing markedly

from that described above were observed. These events involved con-

specific individuals. After grasping the female and achieving the

tandem position, the male was obviously encountering difficulty in

flying with her to a perch. The female appeared to be resisting the

male's grasp by turning her head and body. The spasmodic, almost zig-

zag flight probably resulted from the female beating her wings in

opposition to the male. This behavior was termed "refusal motion."

These pairs became disengaged usually within five seconds, the female

then flying away from the water, often followed by the importunate

male. One of the instances involved a female which had stopped ovi-

positing and was apparently leaving the reproductive site for that

day. This behavior is similar to that described by Bick and Bick

(1963) for E. civil. I believe that these females are not responsive

to mating attempts because their fertile egg supply is exhausted, and

it would not be advantageous for them to mate again so soon.

Perched females were often seen curving the abdomen downward and

spreading their wings upon approach of another damselfly (Figure 14).

Similar behavior was observed by Grieve (1937) in Ischnura verticalis

(Say) and interpreted as a posture for attracting males. Bick (1966)

showed that this posture, which he termed "ventral abdominal curving,"

is neither a signal attracting conspecific males nor a cleaning move-

ment, but that it is a threat display enabling the female to maintain

a perch at water. The phenomenon is reviewed by Pajunen (1963).

Females of E. signatum and E. pollutum not in mating readiness and

those recently grasped by a heterospecific male and still being pursued

by males often curve the abdomen ventrally, indicating that this pos-

ture is a signal that she is not receptive. To test this hypothesis,

models were mounted with abdomens decurved and with abdomens straight

and presented simultaneously to E. pollutum males. The results in

Figure 14. Perched, unreceptive female of E. pollutum with
abdomen decurved and wings spread in response to
an approaching male.

Table 2 show that 44 percent of the males grasped models with decurved

abdomens and 56 percent grasped models with straight abdomens. These

responses are not significantly different, as the probability of the

observations is greater than 0.25. This test suggests that the actual

posture does not ward off males; rather the actual movement of decurv-

ing the abdomen appears to be the "nonreceptive signal" to males. To

test this hypothesis, live females were tethered by gluing their legs

to a thin stem and presented to males. Upon male approach, these

females decurved the abdomen and spread their wings; males departed

without landing on such females, evidence that the act of refusal is

important and not the attained posture.

Mating Frequency

Males appear at water nearly every day they are sexually mature,

usually returning to the same general vicinity of the creek, although

not necessarily to the same microarea. Females do not appear at water

every day, and their whereabouts on days away from water are unknown.

Males do not mate every day they are at water, although females

apparently do. No individuals were observed mating more than once

daily. Only one male (E. pollutum) was seen mating on two successive

days. Most individuals probably mate at least twice and probably as

many as five times during their lifespan.

Table 2. Number of Enallagma pollutum males attempting tandem with
simultaneously presented females with straight or decurved
abdominal postures.

Abdominal posture
of female

Number of males
attempting tandem

decurved 18 (44%)
straight 23 (56%)


In search of the mechanisms isolating Enallagma pollutum and

E. signatum, several possibilities were immediately eliminated. First,

there obviously are no habitat or seasonal differences where my studies

were conducted. Secondly, daily mating times are essentially similar,

even though E. pollutum males are usually present at the reproductive

site earlier than males of E. signatum. There appears tobeno difference

in the times at which females come to these sites. Therefore the po-

tential for interspecific matings exists, and proof of isolation was


Examination of more than 250 copulating pairs of each species

resulted in not one instance of interspecific mating, indicating that

E. pollutum and E. signatum are very effectively isolated. Absence of

interspecific copulation eliminates the need for postmating isolation,

and although such a mechanism may exist between these species, it could

not be tested.

In order to determine the moment at which isolation occurs, the

mating sequence was retraced and attention focused on interspecific

mating attempts. A schematic account of this behavior is presented

in Figure 15. The major reason for females being unreceptive is that

they terminated oviposition for that day and were leaving the water

when they were taken by conspecific males. These females perform


d flies to 9, grasps
her and achieves tandem

9 is not sexually 9 is sexually
receptive receptive

tandem position

c is

9 receives stimuli
for discrimination

cf is

successful tandem flight
to a perch

d transfers sperm


Figure 15. Schematic account of mating sequence in Enallagma
pollutum and E. signatum.

refusal motions as previously described (page 44). Observations indi-

cate that females cannot escape males, even though they appear to make

an evasive flight. Sexually responsive females were observed in tandem;

13 of the 50 precopulatory tandem pairs collected (26 percent) involved

heterospecific individuals: seven E. pollutum males took E. signatum

females, two E. pollutum males took E. dubium females, and four E.

signatum males took E. pollutum females. After departure of the hetero-

specific males, eight of these females were successfully mated by con-

specific males; the other five escaped my sight. The heterospecific

males released their tandem grasp before or shortly after reaching a

perch site because of vigorous refusal motions of the females. There-

fore, females do not attempt copulation with heterospecific males, and

reproductive isolation occurs before copulation, either during the

tandem attempt or the tandem flight. The classical mechanical isola-

tion theory, which proposes that the genitalia are incompatible, is

therefore not applicable. In addition, no structural differences in

the genitalia of these two species were found.

Elimination of the above possible mechanisms points to the pres-

ence of ethological isolation. The interspecific tandem attempts

indicate that although females are differently colored, males are un-

able to visually distinguish them. Species-recognition experiments

to test this hypothesis involved presentation of freshly-killed

females to males in apparent mating readiness. Although females are

taken into tandem more commonly when flying, flight is not necessary

for release of male mating behavior. Males of both species take

perched females, facilitating the use of mounted females for testing.

Use of dead females prevented the refusal behavior of live tethered

females that deters males from landing on them. A total of 900

presentations were made. Male responses were recorded as follows:

1 no response; 2 flight toward model but no contact; 3 landing

on model; 4 attempting to clasp the model's thorax with appendages;

5 successful clasping and tandem position. This system is similar

to that used by Paulson (1974). Responses 1 and 2 were considered

negative, i.e., the proper visual stimulus(i) to elicit a mating

response was assumed absent. Responses 3, 4 and 5 were considered

positive, i.e., the proper stimulus was present. Table 3 summarizes

the male responses.

Males of E. pollutum and E. signatum responded to females of both

species in nearly equal frequencies. The probabilities of the chi-

square values under the null hypothesis are approximately 0.10 for

E. pollutum and 0.25 for E. signatum. Therefore the responses are

not significantly different, indicating that males do not visually

distinguish females of the two species and therefore do not control

isolation. The percent negative response by males to conspecific

females was assumed to be a result of males not in mating readiness

or the dead females somehow lacked the proper stimulus necessary to

elicit a mating response from these males.

To further test the males' apparent inability to distinguish

females, freshly-killed females were presented simultaneously to

males as a choice test. Females were mounted one inch apart on

separate pins and were given identical postures. After each positive

male response, their positions were alternated. The positive responses

Table 3. Response of Enallagma pollutum and E. signatum males to 900
presentations of conspecific and heterospecific females.
No movement toward and no contact with the model were scored
as negative responses (-); landing on the model was scored
as a positive response (+).

Females Presented

pollutum signatum X2

+ 203(81%) 188(75%)
pollutum 2.64
47 62

+ 134(67%) 142(71%)
signatum 0.74
66 58

are summarized in Table 4. The probabilities of the chi-square values

are 0.75 for E. pollutum and 0.50 for E. signatum, and are not signifi-

cantly different, further proof that males are unable to visually rec-

ognize females. It seems there would be selection for males to be

able to distinguish females. However, their inability may be explained

by the presence of females intermediate in color. I was able to dis-

tinguish most females using thoracic color, but three E. pollutum had

greenish-yellow thoracic stripes and a greenish-grey abdomen laterally,

and two E. signatum had orangish-green thoracic stripes and a yellowish-

grey abdomen laterally. Therefore, males not discriminating against

color may acquire conspecific mates more often.

Paulson (1974) defined visual isolation as the proportion of

responses that are negative. Expressed thusly, the data in Table 3

indicate that E. pollutum males are 25 percent visually isolated from

E. signatum females, and that E. signatum males are 33 percent visually

isolated from E. pollutum females. However, the percent negative re-

sponse of males to conspecific females must be considered, so that

the corrected visual isolation percentages are 6 percent and 4 per-

cent. These results indicate absence of visual species recognition,

and agree with those of Paulson on four western species of Enallagma.

Paulson further reasoned that low visual isolation percentages

reflect a high degree of mechanical isolation, which he expressed as

the number of attempted tandems that were unsuccessful compared to

the number successful. He apparently assumed that the male receives

information from the fit of his appendages with the female thorax,

and that if the stimuli are improper, he releases and tandem is not

Table 4. Number of positive responses by males to females of both
species when presented simultaneously.






pollutum 18 21 0.13

signatum 13 18 0.80


achieved. This view places the male in the discriminating role and

hence responsible in effecting isolation. My observations indicate

the assumption that males discriminate is invalid, at least in E.

pollutum and E. signatum, for three reasons. First, the female

thoracic parts contacted by the male appendages are very similarly

shaped, and are also variable; it is difficult to imagine how a male

could receive different stimuli from these structures. Second, males

take dead, dried heterospecific females into tandem, fly to a perch

and repeatedly attempt to get them to copulate; after one to five min-

utes, she is released. This behavior indicates that the shape of the

female thorax is not discriminated by males. Third, in the naturally-

occurring interspecific tandem pairs seen, the female was obviously

trying to escape from the male's grasp by turning her body and

apparently beating her wings to oppose the male's direction of flight.

The flight path of such pairs is spasmodic compared to the smooth,

direct flight of conspecific pairs. If a male is successful in bring-

ing a heterospecific female to a perch, she continues resisting by

turning her head, moving her legs and other abnormal movements. The

male releases his appendages and usually flies away, the female often

remaining perched. However, several E. pollutum males remained near

the E. signatum females they had been clasping and after 20 to 40 sec-

onds of vertical abdominal bobbing, flew to the female and again

adopted the tandem position. The female repeated the resistance move-

ments, the male again departing. In one instance, this sequence was

repeated, after which the female flew away. If males were discrimi-

nating, they would not persistently hold heterospecific females in

tandem as described.

For the above reasons, I contend that females discriminate be-

tween heterospecific and conspecific males. The generality of a lower

propensity in females to mate as compared with males supports this

contention. Paulson used live females as models, and the failure of

certain males to achieve tandem with certain heterospecific females

(e.g., E. boreale Selys males with E. carunculatum Morse females) is

probably the result of female refusal motions. Paulson stated that

his category 2a, in which brief contact was made with the female,

appeared to involve "the female attempting to dislodge a responsive

male by vigorous motion." In only one of his mixed-species pairs

was tandem often achieved, that involving E. carunculatum males and

E. boreale females. The success of this pairing seems to indicate a

lack of his mechanical isolation, but the problem was not discussed.

Paulson used the term mechanical isolation in a new sense,

referring to the fit of the male appendages on the female thorax as

a possible nongenitalic lock-and-key type mechanism, and assumed that

lack of visual species recognition by males meant mechanical isolation.

This assumption precludes the possibilities of visual species recogni-

tion by females, and of chemcial and auditory stimuli, and other

behavioral differences.

I could not show conclusively that females are unable to visually

distinguish conspecific and heterospecific males. Female behavior

prior to and during the tandem attempt was observed 27 times, eight

instances involving heterospecific individuals. Females do not change

their style of flight in response to the first male to approach them.

If this first male is conspecific, tandem is achieved, followed by a

smooth flight to a perch. If the first male is heterospecific, tandem

is also achieved, but the female then begins refusal motions and any

ensuing tandem flight is spasmodic. After dislodging the heterospecific

male, her flight near all males is evasive, although she is usually

caught by another male.

I attempted to test the hypothesis that females do not visually

recognize conspecific and heterospecific males by tethering live

females and observing their response to approaching males. However,

tethered females perform refusal motions upon approach of all males;

males come near them but usually do not land, and only one tandem

attempt was observed. The only evidence that females do not recognize

males is their lack of evasive flight from heterospecific males.

If tandem attempts were completely random, a 1:1 ratio of intra-

specific to interspecific tandem attempts would be expected. The

observed ratio is approximately 3:1, the majority (37 of 50) being

intraspecific. The difference is highly significant (P <0.005).

Therefore, mating is not random. Species differences in microhabitat

of both sexes at the water accounts for nonrandom mating. Females of

E. polutum usually make short flights only in vegetation near the

water's edge. Because the majority of E. signatum males hover over

open water, they are less likely to encounter an E. pollutum female

than are E. pollutum males perched near the edge on vegetation. Fe-

males of E. signatum, although flying in vegetation near the edge,

fly also out over the open water. Males of E. pollutum may detect

and catch them when they are near vegetation. The majority of inter-

specific tandem pairs (7 of 11) involved an E. pollutum male and E.

signatum female. This difference does not necessarily mean that females

visually recognize conspecific and heterospecific males; their visual

cue may simply be the vegetation versus the open water.

The above microhabitat difference provides partial isolation of

the species by lessening the probability of interspecific meetings.

This mechanism is not reliable, as many E. signatum males perch near

the edge and E. pollutum males often fly over open water, and inter-

specific tandem attempts occur. An additional mechanism is required

once heterospecific individuals engage in tandem. The three remaining

ethological types of isolating mechanisms by which females could re-

ceive discriminatory information involve chemical, auditory and tac-

tile stimuli.

Pheromone and auditory communication have not been demonstrated

in Odonata. Evidence that females do not attract males by olfactory

or auditory stimuli is that males take dead, dried females into

tandem. As a further test, live females were placed in small cages

set near active males. These females could not be seen by males, and

were not approached, further evidence of a lack of chemcial and audi-

tory stimuli for attracting males. The possibility that the male re-

leases a specific chemical upon contact with the female was not tested.

Contact chemoreception is probably common in insects, although the

precise role in reproductive isolation has received little attention.

Williamson (1906) reported that in Argia the male apical abdominal

tubercles usually leave a white spot on the female thorax, and proposed

the function to be a strengthening of the tandem bond. If these sub-

stances were species-specific, they could be used in communication.

Histological sections of male superior appendages of E. pollutum and

E. signatum show specialized cells, probably secretary in function,

beneath the thick, unsclerotized cuticle of the mesal margin (Figure

16). These cells are columnar, have large nuclei and intercellular

spaces, and appear to be multilayered in places. The function of

these cells may be to secrete continuously the thick cuticle of the

mesal margin during the adult male maturation period. The inferior

appendages, which lack a thick cuticle, do not contain these cells.

Another possible function is secretion of a pheromone during tandem.

Such a chemical may serve to release female mating behavior, or if

specifically different, may be detected by females and isolate the

species. Coeloconic sensillae (pit pegs) are present on damselfly

antennae (Slifer and Sekhon, 1972); they have been assumed to serve

as chemoreceptors for distinguishing differences in pond water

(Steiner, 1948). The thorax bears thin-walled sensillae and small

pegs. Release of a specific chemical during tandem was not tested

because of the uncooperativeness of captive females. Chemical com-

munication is unreported in Odonata, but the possibilities should be

investigated. No behavior which could be called stridulation was

observed during tandem attempts, and no distinct sounds were audible

to me.

Observation of interspecific tandem attempts revealed that the

female does not begin refusal motions until tandem is achieved. This

fact suggests that tactile stimuli are releasers of female mating

behavior and specific differences may isolate the species. An at-

tempt was made to demonstrate that the shape of the male superior

Figure 16. Oblique section through male superior appendage of
E. ollutum; arrow indicates specialized secretary
cells (pul ed away during fixation) beneath the
thickened, unsclerotized cuticle (500X).

appendages provides the stimulus for species recognition by the female.

Newly-emerged males were collected and the soft, yet unsclerotized

superior appendages were altered with a pair of forceps. The tips of

E. signatum appendages were pushed inward to shorten the appendage

whereas those of E. pollutum were compressed dorsoventrally to narrow

them. These altered males were characteristically marked with white

paint on the thorax and a group of unaltered controls were marked dif-

ferently, and both groups released. Many of these general damselflies

were damaged, and the return within two weeks was very low. The

appendages retained their altered shape after sclerotization. Those

males returning were not seen attempting tandem or copulating. In a

second experiment, mature males were collected and the superior append-

ages altered by cutting off the tips in E. signatum and the ventral

lobe in E. pollutum; these individuals were then marked and released.

Over half of these males flew away from the site immediately; those

remaining behaved apparently normally. Of these, only one E. signatum

was seen attempting tandem with a conspecific female; the female per-

formed refusal motions and tandem was disengaged. No altered males

were seen in copulation whereas several control males mated success-


These observations,although not as controlled as desired, indi-

cate that superior appendage shape provides the critical stimulus

for discrimination by females. However, injury to the male during

alteration could have affected the ability to properly grasp the

female. For example, shortening the appendage in E. signatum could

have weakened the bond between appendages and thorax and caused

separation of the pair.

The surface of the female thorax bears numerous setae and other

sensillae which may function as tactile receptors (Figures 17 and 18).

In tandem, the ventral lobe of E. pollutum superior appendages contact

the posterior surface of the female's hind pronotal lobe. These hind

lobes differ significantly in the two species. In E. pollutum, the

posterior surface bears numerous setae and small, sharp ridges near

the dorsal rim, and has a striated ventral rim (Figure 19). In E.

signatum, the surface bears setae on the dorsal rim only, and smaller,

low ridges all over the posterior surface; the ventral rim is quite

smooth (Figure 20). It is highly probable that these sensillae re-

ceive information from the ventral margin of the male superior append-

age. This margin in E. pollutum definitely touches the hind lobe for

at least half the appendage length; in E. signatum, the only point of

contact appears to be the small ventral projection located just dis-

tal to the mid-length of the appendage.

Another difference in the male's superior appendages which could

provide information to females is length: the tip of the E. pollutum

appendage does not extend as far down the side of the female's thorax

as that of E. signatum (compare Figures 12 and 13). Tactile receptors

could easily detect this difference. The mesostigmal plates, on which

the mesal margins of the male superior appendages lie, bear some tac-

tile hairs, but appear to be heavily sclerotized and sculptured to

function as organs for being clasped (Figures 21 and 22). Jurzitza

(1974) found "tactile bristles" on the prothorax and mesostigmal plates

of Ischnura fluviatilis Selys which contact the male appendages during

tandem. I believe that the mesostigmal plates are modified to receive

Figure 17. Scanning electron micrograph of thorax of an E. pollutum
female showing sensory receptors (10OX); pp = prothoracic
pit, hl = hind lobe, mp = mesostigmal plate.

Figure 18. Scanning electron micrograph of thorax of an E. signatum
female showing sensory receptors (100X).

Figure 19. Scanning electron micrograph of hind pronotal lobe
(left side, posterior view) of an E. pollutum female


Figure 20. Scanning electron micrograph of hind pronotal lobe
(left side, posterior view) of an E. signatum female

Figure 21. Scanning electron micrograph of E. pollutum
female mesostigmal plate (170X).

Figure 22. Scanning electron micrograph of E. signatum
female mesostigmal plate (170X).

the specific shape of the male superior appendage in order to afford

a secure tandem bond. The mesa margin of the superior appendage is

unsclerotized and flexible, with many small tubercless," further

allowing for a secure grip on the mesostigmal plates.

There are other ways by which the female could receive tactile

information. During the tandem attempt, for example, there may be

specific differences in the way the male grasps the female with his

legs, or in the way he "bites" the top of her head. However, no

consistent differences in male tandem behavior were detected. The

"biting" appears to function merely to stimulate the female sexually

and perhaps to prevent delay in mating.

Mayr (1970) did not include tactile stimuli as an isolating

mechanism. Barth (1968) found evidence that tactile stimuli, in

addition to chemical stimuli, are releasers of mating behavior in

cockroaches. The tactile isolation mechanism proposed above is best

classified as ethological, and is not a mechanical or lock-and-key

type mechanism. Males can clasp and securely hold heterospecific fe-

males with their appendages, hence there appears to be no mechanical

barrier. Several males of E. weewa, which have very differently

shaped, bilobed superior appendages, effectively clasped and held

presented females of E. pollutum and E. signatum.

Paulson (1974) worked under the hypothesis that those species

differing primarily in structure of male appendages will be found to

be mechanically isolated, whereas those differing primarily in the

coloration of the females will be found to be visually isolated. My

observations on other species of Enallagma at the study sites indicate

that other isolating mechanisms are operative. Enallagma dubium is

reproductively active earlier in the diel cycle than E. pollutum and

E. signatum, although the periods overlap slightly. For the most part,

E. dubium is temporally isolated; however, E. dubium females may still

be at the water when E. pollutum males arrive, and two were seen taken

into tandem by E. pollutum males. They performed refusal motions and

the males departed. The superior appendages of E. dubium males are

similar to those of E. signatum, and females may discriminate against

E. pollutum by appendage shape. However, males of E. dubium are

smaller and very differently colored; the antehumeral stripe is dark

metallic red and very narrow, and the abdomen is entirely black.

Therefore, females may also be able to recognize males visually.

The activity period of E. weewa coincides for the most part with

those of E. pollutum and E. signatum, and males hover over open water

as do E. signatum. No naturally occurring, interspecific tandem at-

tempts were observed, indicating effective visual isolation. The

thorax of E. weewa females is distinctively striped, and mature males

are black except the ninth abdominal segment is blue. They can be

recognized by a human observer from ten feet away.

Enallagma daeckii may be isolated from the other species by

microhabitat, size and behavioral differences. Males and females

usually perch or fly very slowly in shaded areas along the bank,

usually away from the water. Males are very elongate, bright blue

and appear to "search" for the light brown females in the shaded

areas. No interspecific tandem attempts involving E. daeckii were

seen. Enallagma pallidum is similar in appearance to E. daeckii,

but was not found at the Hwy. S-225 site where E. daeckii is common.

These two species may be isolated by habitat differences.

Males of E. divagans are bright blue and black, and are mostly

active earlier in the diel cycle than E. pollutum and E. signatum,

and their flight season is shorter, from March to May. Enallagma

cardenium is a very dark species, mostly black with dark blue thoracic

stripes and second, third, eighth and ninth abdominal tergites.

Enallagma concisum males are bright red, and E. geminatum males are

bright blue; these species were uncommonly seen at the study sites.

These four species are probably visually isolated from E. pollutum

and E. signatum, but isolation among them was not studied.

Of the ten Enallagma species at the study sites, only three,

E. signatum, E. dubium and E. concisum, have similarly shaped

superior appendages. The females are quite similar in appearance

whereas males are distinctively colored. The other seven species

have distinctively shaped male superior appendages, and most are

also distinctively colored or patterned. Evidence of ethological

and ecological isolation exists, and I am uncertain whether the

species differ "primarily" in coloration or appendage shape. I

believe that the exceptions to Paulson's generalizations (which in

essence are the same as those of Williamson, 1906) cast doubt upon

their usefulness, at least in the genus Enallagma.

Lack of courtship display may be the ancestral zygopteran

condition. Based on this assumption, reproductive behavior of

Enallagma and probably most Coenagrionidae can be considered ances-

tral. In his phylogenetic tree of the Odonata, Fraser (1957) placed


Coenagrionidae at the end of a line descending directly from the

Protozygoptera based on morphology. Behavioral evidence substan-

tiates this arrangement. However, courtship displays, considered

derived characteristics, have been found in Platycnemididae, a

family which Fraser apparently considered more ancestral morphologi-

cally than the Coenagrionidae. Knowledge of behavior is still too

scanty for clarification of the phylogenetic relationships of these



1. Reproductive behavior of Enallagma pollutum and E. signatum is

similar in most aspects. The species are semicrepuscular most

of the year, and although activity periods overlap, E. pollutum

usually begins activity earlier in the diel cycle. Enallagma

signatum males usually hover over open water whereas E. pollutum

males perch mostly near the bank.

2. Males, and probably females, are capable of sex recognition. The

males' orange ninth abdominal segment is an important sex-

recognition character.

3. Vigorous intraspecific and interspecific interactions enable males

to maintain a position at the water, an advantage in securing a


4. Pairing occurs at the water without courtship displays. Sperm are

transferred in tandem, in response to female "abdominal probing"

and before copulation. Copulation takes place on vegetation

along the stream bank and lasts an average of 20 minutes. Indi-

viduals may mate once in a day, but rarely mate on successive


5. Females usually oviposit in tandem shortly after copulation.

Eggs are deposited in plant tissue at and below the water surface.

6. The species are very effectively isolated, and isolation is

effected prior to the copulatory act. The mechanical isolation

theory of incompatible genitalia is not applicable.

7. Males are incapable of visually distinguishing conspecific and

heterospecific females, and attempt tandem with either species.

Partial isolation is achieved by differences in microhabitat:

females of E. pollutum usually remain in vegetation near the

bank whereas females of E. signatum eventually fly toward open

water. This difference lessens the probability of interspecific


8. Isolation also occurs during the tandem attempt, at which time

females receive stimuli enabling them to discriminate between

heterospecific and conspecific males. Females grasped by

heterospecific males perform "refusal motions," upon which the

males then release; the females, therefore, control isolation.

9. Cells of a secretary nature are present in the male superior

appendages; possible functions include secretion of the thick

unsclerotized cuticle on the mesal margin, or secretion of a

pheromone. The possibility of chemical isolation by way of

contact chemoreception, as yet unreported in Odonata, requires

further investigation.

10. Tactile stimuli, provided by the shape of the male superior

appendages, appear to be releasers of mating behavior in females,

and specific differences may constitute the mechanism isolating

the species. The female thorax bears numerous tactile sensillae.

This proposed mechanism is ethological, not lock-and-key.


11. Exceptions to the generalizations that species with differently

shaped male appendages are mechanically isolated and species

differently colored or patterned are visually isolated may be

found in the genus Enallagma.


Barth, R. H. 1968. The comparative physiology of reproductive
processes in cockroaches. Part I. Mating behaviour and its
endocrine control. Advances in Reprod. Physiol. 3:167-207.

Beck, W. M. 1965. The streams of Florida. Bull. Fla. State Mus.

Bick, G. H. 1966. Threat display in unaccompanied females of the
damselfly, Ischnura verticalis (Say) (Odonata: Coenagrionidae).
Proc. Entomol. Soc. Wash. 68:271.

Bick, G. H., and J. C. Bick. 1961. An adult population of Lestes
disjunctus australis Walker (Odonata: Lestidae). The Southwest.
Nat. 6:111-137.

Bick, G. H., and J. C. Bick. 1963. Behavior and population struc-
ture of the damselfly, Enallagma civil (Hagen) (Odonata:
Coenagriidae). The Southwest. Nat. 8:58-84.

Bick, G. H., and J. C. Bick. 1965. Sperm transfer in damselflies
(Odonata: Zygoptera). Ann. Entomol. Soc. Amer. 58:592.

Bick, G. H., and J. C. Bick. 1972. Substrate utilization during
reproduction by Argia plana Calvert and Argia moesta (Hagen)
(Odonata: Coenagrionidae Odonatologica 1:3-9.

Bick, G. H., and L. E. Hornuff. 1965. Behavior of the damselfly,
Lestes unguiculatus Hagen (Odonata: Lestidae). Proc. Ind. Acad.
Sci. 75:110-115.

Bick, G. H., and L. E. Hornuff. 1966. Reproductive behavior in the
damselflies Enallagma aspersum (Hagen) and Enallagma exsulans
(Hagen) (Odonata: Coenagriidae). Proc. Entomol. Soc. Wash.

Bick, G. H., and D. Sulzbach. 1966. Reproductive behaviour of the
damselfly, Hetaerina americana (Fabricius) (Odonata: Caloptery-
gidae). Anim. Behav. 14:156-158.

Borror, D. J. 1934. Ecological studies of Argia moesta Hagen (Odonata:
Coenagrionidae) by means of marking. Ohio J. Sci. 34:97-108.

Buchholtz, C. 1951. Untersuchungen an der Libellen-Gattung Calop-
teryx-Leach unter besonderer Berucksichtigung ethologischer
Fragen. Z. Tierpsychol. 8:273-293.

Buchholtz, C. 1955. Eine vergleichende Ethologie der orientalischen
Calopterygiden (Odonata) als Beitrag zu ihrer systematischen
Deutung. Z. Tierpsychol. 12:364-386.

Buchholtz, C. 1956. Eine Analyse des Paarungsverhaltens und der
dabei wirkenden Ausl~ser bei den Libellen Platycnemis pennipes
Pall. und P1. dealbata Klug. Z. Tierpsychol. 13:14-25.

Cantrall, I. J., and T. J. Cohn. 1970. Melanoploid genitalia and
mechanical isolation. Proc. Int. Study Conf. Current and Future
Problems of Acridology, London. Session:Taxonomy. Pp. 35-44.

Consiglio, C. 1974. Some observations on the sexual behaviour of
Platycypha caligata caligata (Selys) (Zygoptera: Chlorocyphidae).
Odonatologica 3:257-259.

Corbet, P. S. 1952. An adult population study of Pyrrhosoma nymphula
(Sulzer):(Odonata: Coenagrionidae). J. Anim. Ecol. 21:206-222.

Corbet, P. S. 1962. Age-determination of adult dragonflies (Odonata).
XI Int. Congr. Entomol., Vienna 1960:287-289.

Corbet, P. S. 1963. A biology of dragonflies. Quadrangle Books,
Chicago. 247 pp.

Currie, N. 1963. Mating behavior and local dispersal in Erythemis
simplicicollis. Proc. 18th Ann. Meeting N. Central Branch,
Entomol. Soc. Amer. 18:112-113.

Dobzhansky, T. 1970. Genetics of the evolutionary process. Columbia
Univ. Press, New York.

Fraser, F. C. 1949. A revision of the Chlorocyphidae with notes on
the differentiation of the Selysian species rubida, glauca,
cyanifrons and curta. Bull. Inst. Sci. Nat. Belg. 25:1-50.

Fraser, F. C. 1957. A reclassification of the order Odonata. Royal
Zool. Soc. New South Wales. 133 pp.

Grieve, E. G. 1937. Studies on the biology of the damselfly,Ischnura
verticalis Say, with notes on certain parasites. Entomol. Amer.

Heymer, A. 1966. Etudes comparees du comportement inn6 de Platycnemis
acutipennis Selys et de Platycnemis latipes Rambur. Ann. Soc.
Entomol. Fr. 2:39-73.

Heymer, A. 1973. Etude du comportement reproducteur et enalyse des
mecanismes declencheurs innes (MDI) optiques chez les caloptery-
gidae (Odon, Zygoptera). Ann. Soc. Emtomol. Fr. 9:219-255.

Howe, R. H. 1917. Manual of the Odonata of New England. Part II.
Mem. Thoreau Mus. Nat. Hist. II: 9-24.

Ito, Y. 1960. Territorialism and residentiality in a dragonfly,
Orthetrum albistylum speciosum Uhler (Odonata: Anisoptera).
Ann. Entomol. Soc. Amer. 53:851-853.

Jacobs, M. E. 1955. Studies on territorialism and sexual selection
in dragonflies. Ecology 36:566-586.

Johnson, C. 1961. Breeding behavior and oviposition in Hetaerina
americana Fabricius and H. titia (Drury) (Odonata: Agriidae).
Can. Entomol. 93:260-266.

Johnson, C. 1962a. Reproductive isolation in damselflies and dragon-
flies (order Odonata). Texas J. Sci. 14:297-304.

Johnson, C. 1962b. A study of territoriality and breeding behavior
in Pachydiplax longipennis (Burmeister) (Odonata: Libellulidae).
The Southwest. Nat. 7:191-197.

Johnson, C. 1962c. Breeding behavior and oviposition in Calopteryx
maculatum (Beauvois) (Odonata: Calopterygidae). Amer. Mid]. Nat.

Johnson, C. 1964. Mating expectancies and sex ratio in the damselfly,
Enallagma praevarum (Odonata: Coenagrionidae). The Southwest.
Nat. 9:297-304.

Johnson, C. 1972. Tandem linkage, sperm translocation, and copulation
in the dragonfly, Hagenius brevistylus (Odonata: Gomphidae). Amer.
Midl. Nat. 88:131-149.

Jordan, K. 1905. Der Gegensatz zwischen geographischer und nichtgeo-
graphischer Variation. Z. wiss. Zool. 83:151-210.

Jurzitza, G. 1974. Rasterelektronenmikroskopische Untersuchungen des
Zangengriffes und der Laminae mesostigmales einiger Coenagrionidae
(Odonata, Zygoptera). Forma et Functio 7:377-392.

Kennedy, C. H. 1922. The ecological relationships of the dragonflies
of the Bass Islands of Lake Erie. Ecology 3:325-336.

Kennedy, C. H. 1928. Evolutionary level in relation to geographic,
seasonal and diurnal distribution of insects. Ecology 9:367-379.

Koen, J. 1937. The dragonflies of the Reelfoot Lake area. J. Tenn.
Acad. Sci. 12:129-153.

Kormondy, E. J. 1959. The systematics of Tetragoneuria based on
ecological, life history, and morphological evidence (Odonata:
Corduliidae). Univ. Mich. Mus. Zool. Misc. Pub. 107:7-79.

Kormondy, E. J. 1961. Territoriality and dispersal in dragonflies
(Odonata). J. N. Y. Entomol. Soc. 69:42-52.

Krieger, F., and E. Krieger-Loibl. 1958. Beitrage zum Verhalten von
Ischnura elegans und Ischnura pumilio (Odonata). Z. Tierpsychol.

Loibl, E. 1958. Zur Ethologie und Biologie der deutschen Lestiden
(Odonata). Z. Tierpsychol. 15:54-81.

Lorkovic, Z. 1958. Some peculiarities of spatially and sexually
restricted gene exchange in the Erebia tyndarus group. Cold
Springs Harbor Symp. Quant. Biol. 23:319-325.

Lutz, P. E., and A. R. Pittman. 1970. Some ecological factors influ-
encing a community of adult Odonata. Ecology 51:279-284.

Mayr, E. 1970. Populations, species, and evolution. Belknap Press
of Harvard Univ. Press, Cambridge, Mass. 453 pp.

Moore, N. W. 1952. On the so-called "territories" of dragonflies
(Odonata-Anisoptera). Behaviour 4:85-99.

Moore, N. W. 1957. Territory in dragonflies and birds. Bird Study

Needham, J. G., and H. B. Heywood. 1929. A handbook of the dragon-
flies of North America. Charles C. Thomas, Springfield, Ill.
378 pp.

Neville, A. C. 1959a. Inter-generic pairing in Odonata. Entomol.
Mon. Mag. 95:179.

Neville, A. C. 1959b. Mating swarm in Enallagma cyathigerum (Charp.)
Odon., Coenagriidae). Entomol. Mon. Mag. 95:181.

Noble, G. K. 1939. The role of dominance in the life of birds.
Auk 56:263-273.

Pajunen, V. I. 1963. On the threat display of resting dragonflies
(Odonata). Ann. Entomol. Fenn. 29:236-239.

Pajunen, V. I. 1964. Mechanism of sex recognition in Leucorrhinia
dubia v. d. Lind., with notes on the reproductive isolation
between L. dubia and L. rubicunda L. (Odon., Libellulidae).
Ann. Zool. Fenn. 1:55-71.

Patrick, 0. R., and P. E. Lutz. 1969. A life-history study of the
damselfly Ischnura posita (Hagen). J. Elisha Mitchell Sci.
Soc. 85:131.

Paulson, D. R. 1973. Temporal isolation in two species of dragon-
flies, Epitheca sepia (Gloyd, 1933) and E. stella (Williamson,
1911) (Anisoptera: Corduliidae). Odonatologica 2:115-119.

Paulson, D. R. 1974. Reproductive isolation in damselflies. Syst.
Zool. 23:40-49.

Rentz, D. C. 1972. The lock and key as an isolating mechanism in
katydids. Amer. Sci. 60:750-755.

Schick, R. X. 1965. The crab spiders of California (Araneida,
Thomisidae). Bull. Amer. Mus. Nat. Hist. 129:1-180.

St. Quentin, D. 1934. Beobachtungen und Versuche an Libellen in
ihren Jagdrevieren. Konowia 13:275-282.

St. Quentin, D. 1964. Territorialitat bei Libellen (Odonata).
Ergebnisse und Ausblicke. Mit. Munch. Entomol. Gesel. Bd. 54:

Slifer, E. H., and S. S. Sekhon. 1972. Sense organs on the antennal
flagella of damselflies and dragonflies (Odonata). Internat. J.
Insect Morph. Embryol. 1:289-300.

Steiner, H. 1948. Die Bindung der Hochmoorlibelle Leucorrhinia dubia
Vand. an ihren Biotop. Zool. Jb. Jena (Syst.) 78:65-96.

Tillyard, R. J. 1913. On some new and rare Australian Agrionidae
(Odonata). Proc. Linn. Soc. New South Wales 37:404-479.

Waage, J. K. 1973. Reproductive behavior and its relation to terri-
toriality in Calopteryx maculata (Beauvois) (Odonata: Caloptery-
gidae). Behaviour 47:240-256.

Walker, E. M. 1912. The North American Dragonflies of the genus
Aeshna. Univ. Toronto Studies Biol. Series 11:1-213.

Walker, E. M. 1915. Notes on Staurophlebia reticulata Burm. Can.
Entomol. 47:387-393.


Watson, J. A. L. 1966. Genital structure as an isolating mechanism
in Odonata. Proc. Roy. Entomol. Soc. London Ser. A 41:171-174.

Webb, G. R. 1947. The mating-anatomy technique as applied to
polygyrid landsnails. Amer. Nat. 81:134-147.

Williamson, E. B. 1906. Copulation of Odonata. Entomol. News 17:


Kenneth Joseph Tennessen was born 10 June 1946 in Ladysmith,

Wisconsin. During his youth, an interest in animals and other natural

things was kindled and continued into high school; attention became

focused on the diversity and abundance of insects. After completing

secondary education at Phillips High School, Phillips, Wisconsin, he

enrolled as a biology major at Wisconsin State University-Platteville.

After two years he transferred to the University of Wisconsin at

Madison, earning a B. S. degree in entomology in 1968. His interest in

Odonata led to entering graduate school at the University of Florida,

assisting Dr. Minter J. Westfall in research for his forthcoming manual

on the Zygoptera of North America. He was soon drafted into the United

States Armed Services and served 19 months active duty including 12

months in South Vietnam. He returned to Florida in 1971 and received

the M. S. degree in March, 1973, working on a systematically difficult

genus of dragonflies. Problems encountered during this study led to

an interest in reproductive isolation and a research problem for the

pursuit of the Ph.D. degree.

Mr. Tennessen is a member of the Entomological Society of America,

Phi Kappa Phi, and Sigma Xi. He is married and has two sons.

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Dale H. Habeck, Chairman
Professor of Entomology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Minter J. 1ostfall r., Cd ariman
Professor cf Zoology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

F. Cliffd Joh'-on, II
Professor of Zoology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Philip S Callahan
Professor of Entomology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Lewis Berner
Professor of Zoology

This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council, and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.

August, 1975

Dean, College of Agr iture

Dean, Graduate School


3 1262 08553 0516

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs