Sexual behavior in North American cicadas of the genera Magicicada and
Okanagana
by
John Richard Cooley
A dissertation submitted in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
(Biology)
The University of Michigan
1999
Doctoral Committee:
Professor Richard D. Alexander, Chairman
Professor George Estabrook
Professor Brian Hazlett
Associate Professor Warren Holmes
©
John R. Cooley
All Rights Reserved
1999
To my wife and family
ii
ACKNOWLEDGMENTS
This work was supported in part by the Horace H. Rackham School of Graduate
Studies, the Department of Biology, the Museum of Zoology (UMMZ), the Frank W.
Ammerman Fund of the UMMZ Insect Division, and by Japan Television Workshop Co.,
LTD.
I thank the owners and managers of our field sites for their patience and
understanding:
Alum Springs Young Life Camp, Rockbridge County, VA.
Horsepen Lake State Wildlife Management Area, Buckingham County, VA.
Siloam Springs State Park, Brown and Adams Counties, IL.
Harold E. Alexander Wildlife Management Area, Sharp County, AR.
M. Downs, Jr., Sharp County, AR.
Richard D. Alexander, Washtenaw County, MI.
The University of Michigan Biological Station, Cheboygan and Emmet Counties, MI.
John Zyla of the Battle Creek Cypress Swamp Nature Center, Calvert County,
MD provided tape recordings and detailed distributions of Magicicada in Maryland. Don
Herren, Gene Kruse, and Bill McClain of the Illinois Department of Natural Resources,
and Barry McArdle of the Arkansas State Game and Fish Commission helped obtain
permission to work on state lands. Keiko Mori and Mitsuhito Saito provided funding and
high quality photography for portions of the 1998 field season. Laura Krueger performed
the measurements in Chapter 7. Artwork in Chapters 7 and 8 is by Melissa Anderson,
Dan Otte, and John Megahan.
iii
I thank Charles L. Remington for introducing me to the complexities of
Magicicada, Oren Hasson, Tom Moore, Mark O’Brien, and Dan Otte for helpful
discussion, and Chris Simon, for her pioneering genetic surveys of 13- year cicadas and
our conversations about cicadas. Andrew Richards has been instrumental in the project
to map the distributions of periodical cicada broods in central Illinois. Although little of
that information is included in this thesis, Andrew’s meticulous work and careful
thinking have been an important component of all our various and ongoing cicada
projects.
My development and maturation as a graduate student were strongly influenced
by the members of my doctoral committee, Richard D. Alexander, George Estabrook,
Brian Hazlett, and Warren Holmes. None of this work could have been possible without
the nearly 40 years of effort of Richard D. Alexander. I thank Dick for reigniting my
early interest in periodical cicadas. Most of the techniques, approaches, and questions in
this work can be traced back to Dick, and the discovery of the new Magicicada species
and the female wing flick signal are the legacies of his pioneering bioacoustical studies
and his admonition to watch and measure everything. It was at Dick’s suggestion that we
followed the behaviors of individually marked females kept separate from males except
while under direct observation, allowing us to observe the complete male-female
rapprochement sequence. What makes Dick uniquely unique is that he has an uncanny
ability, and a proven track record, for being right about how things work. Dick has been
tireless in his support, profound in his suggestions, and unwavering in his commitment to
seeing that this work meets high quality standards.
My friends David C. Marshall and Deborah Ciszek have also provided
irreplaceable support, input, and collaboration. The fieldwork in this thesis is the joint
collaborative effort of David and myself. The pronoun “I” in this thesis is but a
formality; all “I’s” are truly “We’s.” As a collaborative team, we have accomplished far
more than we could have separately, and it has been a joy to work with as insightful,
iv
careful, and thorough a partner as David. This collaboration has been an important part
of my own intellectual development.
My thanks would not be complete without considering my family. I thank my
parents, Rickie and Bill Cooley, for valuing education above all else, my brothers Bill
and Jamie for their interest and technical expertise, my grandmothers, Frances S. Cooley
for her determination, and Jane H. Yager for her creativity and patience, and my
grandfathers, William H. Cooley for teaching me practical ways to construct things, and
Richard S. Yager, for instilling in me a love of the strange, the natural, and the absurd.
My wife, Louanne Reich Cooley, is a constant source of support. She designed an
inexpensive and portable flight cage, contributed her knowledge of northern Michigan to
the studies of Okanagana, and has been incredibly patient and understanding throughout
this process.
Last, I thank my study organisms. They lead bizarre lives—but then, so do we!
v
PREFACE
Sexual behavior in North American cicadas of the genera Magicicada and
Okanagana
by
John Richard Cooley
Chairman: Richard D. Alexander.
The dense, raucous, mass emergences of North American periodical cicadas
(Magicicada spp.) have attracted scientific study since the 17th century. Periodical
cicadas are unique in their combination of long life cycles (13- or 17- years) and
synchronous development, leading to periodical emergences of adults. Although their
long life cycles make it difficult to conduct a longitudinal study of any one periodical
cicada population, because populations in different regions of eastern North America are
divided into asynchronous “broods,” or year classes, in almost any year, it is possible to
study Magicicada. Over four years, I collected natural history information and studied
the mating behaviors of 13- and 17- year Magicicada and two species in a non-periodical
genus, Okanagana.
vi
Chapter 1 is a review chapter, and it owes a major intellectual debt to my
experiences as a co-author of a book chapter1 about the evolution of insect mating
systems. As I became involved in that project, and as I started to construct my thesis, I
realized that I would have difficulty discussing mate choice unless I clarified for myself,
at least, the nature of mate choice. In this chapter, I review the reasons, starting with
asymmetries in parental investment, why mates should be choosy. I continue with
discussions of how sexual conflicts of interest shape mating systems, and the kinds of
choice mechanisms and criteria invertebrates are most likely to employ. I conclude with
some predictions about possible mechanisms and functions of mate choice in periodical
cicadas.
Chapter 2 is largely a methodological chapter. I present data demonstrating that
the methods I used to mark and confine cicadas were not likely to cause mortality or
behavioral changes that would bias my studies.
Chapter 3 is a brief examination of “seminal plugs” in Magicicada. These
structures are usually left behind in the female genital opening after mating. I
demonstrate that these structures contain DNA and are likely composed of dried
ejaculate, not other specially constructed or secreted materials. Seminal plugs do not
prevent future mating attempts, but I suggest that they may be the products of
antagonistic coevolution between females who are time-limited and males who can
reduce threats to their paternity by imposing time-costs on females’ future mating
attempts.
1
Alexander, R. D., D. C. Marshall, and J. R. Cooley. 1997. Evolutionary Perspectives on
Insect Mating. Ch. 1 in B. Crespi, J. Choe, eds. The evolution of mating systems in
insects and arachnids. Cambridge University Press.
vii
Chapter 4 tests two hypotheses for female multiple mating in Magicicada:
Females remate either to replenish depleted supplies or to effect postcopulatory mate
choice. I demonstrate that interrupting a female’s mating tends to promote remating,
while matings with an inappropriate, conspecific mate do not. Further, mated females do
not actively seek additional mating opportunities, nor are they sexually attractive to
males. Interrupted females, however, solicit second matings. These results are most
consistent with the hypothesis that females remate for replenishment, and they disprove
the hypothesis that females use postcopulatory choice mechanisms to make species-level
mating decisions.
Chapter 5 documents a previously unknown female wing flick signal, used by
females to signal sexual receptivity. No such signal has previously been reported in
North American cicadas, although similar signals are known in some Australian and New
Zealand species. This chapter documents the signal in different Magicicada species and
also describes a bizarre effect (behavioral bisexuality) of a fungus disease in males, as
well as a specialized, competitive signal males use to “jam” the acoustical signals of
interloping rivals. The newly described behaviors allow formulation and testing of a
hypothesis for the two-part calling songs of Magicicada species such as M. septendecim .
The two-part structure of these songs facilitate female discrimination of an individual
male’s call against a background chorus.
Chapter 6 describes a new 13- year species, Magicicada neotredecim. M.
neotredecim male calling songs and female mate acceptance criteria are distinct from
those of the similar, synchronic and partially sympatric species, M. tredecim. Perhaps the
most compelling piece of evidence that M. neotredecim and M. tredecim are separate
species is that M. neotredecim has undergone reproductive character displacement as a
result of its interactions with M. tredecim. The existence of this new species suggests
two general hypotheses for speciation in Magicicada, the “nurse brood” hypothesis, and
viii
the “canalization” hypothesis. Each could explain the origin of M. neotredecim, and each
leads to specific predictions that may someday clarify how M. neotredecim came to be.
This chapter ends with a prediction, based on the “nurse brood” hypothesis and patterns
of reproductive character displacement, suggesting what other undiscovered Magicicada
species might be like and where they might be found.
Chapter 7 explores the possibility that females discriminate among conspecifics
on the basis of phenotypic symmetry. Symmetry has been promoted as a universal and
reliable mate choice criterion. I begin by reviewing asymmetry theory in order to clarify
the nature of the relationship between symmetry and mating success. I conclude that
asymmetry has no special status as a mate choice criterion, but rather, it is like any other
condition-dependent trait. I present the results of a study of symmetry and mating
success in periodical cicadas demonstrating that males with more symmetrical forelegs
are most likely to mate. However, the differential success of some males is more likely
due to indirect mate discrimination than to evolved female mate choice mechanisms.
Chapter 8 compares the behaviors of Magicicada to members of another genus,
Okanagana. I present the results of a 3- year biogeographic survey of the distributions of
Okanagana canadensis and O. rimosa in northern Michigan. I then describe the mating
behaviors of both species, which involve females approaching stationary calling males,
and contrast them to those of Magicicada, which involve males searching for females.
From this contrast, I suggest hypotheses for how the unique mating system of
Magicicada may have evolved from mating systems such as those of Okanagana.
ix
TABLE OF CONTENTS
DEDICATION................................................................................................................... ii
ACKNOWLEDGMENTS ............................................................................................... iii
PREFACE......................................................................................................................... vi
LIST OF TABLES .......................................................................................................... xii
LIST OF FIGURES .........................................................................................................xv
LIST OF APPENDICES ............................................................................................. xviii
CHAPTER
I.
MATE CHOICE IN INVERTEBRATES, WITH SPECIAL
REFERENCE TO PERIODICAL CICADAS (MAGICICADA
SPP.) ..................................................................................................................1
II.
EXPERIMENTAL METHODS IN STUDIES OF PERIODICAL
CICADAS: EXAMINATIONS OF POSSIBLE
METHODOLOGICAL BIASES IN THE STUDY OF
PERIODICAL CICADAS (MAGICICADA SPP.) ......................................30
III.
MATE GUARDING AND THE NATURE AND POSSIBLE
FUNCTIONS OF “SEMINAL PLUGS” IN PERIODICAL
CICADAS (MAGICICADA SPP.).................................................................53
IV.
FEMALE MULTIPLE MATING AND MATE CHOICE IN A
PERIODICAL CICADA, MAGICICADA SEPTENDECIM .....................70
V.
SEXUAL SIGNALING IN PERIODICAL CICADAS,
MAGICICADA SPP........................................................................................97
x
VI.
REPRODUCTIVE CHARACTER DISPLACEMENT AND
ALLOCHRONIC SPECIATION IN PERIODICAL CICADAS
WITH A NEW SPECIES, MAGICICADA NEOTREDECIM..................152
VII.
ASYMMETRY AND MATING SUCCESS IN A PERIODICAL
CICADA, MAGICICADA SEPTENDECIM (HOMOPTERA:
CICADIDAE) ...............................................................................................202
VIII. THE GEOGRAPHIC DISTRIBUTION AND MATING
BEHAVIORS OF TWO CICADA SPECIES (OKANAGANA SPP.)
IN NORTHEAST MICHIGAN ..................................................................235
xi
LIST OF TABLES
Table
1.1. Mate discrimination and its alternatives .......................................................27
2.1. Reported effects of methods for marking individual insects. .......................46
2.2. Results of “seminal plug” analysis ...............................................................48
2.3. Cage density resampling results....................................................................49
2.4. Results of pretarsal clipping experiment.......................................................50
2.5. Summary of resampling results comparing remating frequencies in
small bag cages to remating frequency in large flight cage..........................51
3.1. Study sites 1995- 1997..................................................................................65
3.2. Mating durations in Magicicada septendecim, Okanagana rimosa, and
O. canadensis................................................................................................66
3.3. Results, female mating interruption experiment...........................................67
3.4. Results, resampling algorithm to evaluate whether females with
seminal plugs were unlikely to remate .........................................................68
4.1. Mating durations in Magicicada septendecim. .............................................90
4.2. Mating interruption experiment: Likelihood of remating first or
ovipositing first, and presence/absence of sperm plug .................................91
4.3. Counts of eggnests in mating interruption experiments ...............................92
4.4. Results of playback/ mating status experiment.............................................93
xii
4.5. Mating interruption and female wing flick response to playbacks of
male calls ......................................................................................................94
4.6. Census data from female attractiveness experiment.....................................95
5.1. Traits distinguishing Magicicada species...................................................129
5.2. The sexual sequence in Magicicada species...............................................130
5.3. Study Sites, 1995- 1998 ..............................................................................131
5.4. Length of teneral period, in days ................................................................132
5.5. Mated and unmated female M. septendecim responses to 2- minute
playbacks of male calls ...............................................................................133
5.6. Wing flick responses of females to a series of alternating con- and
hetero- specific calls ...................................................................................134
5.7. Male M. septendecim responses to simulated wing-flick signals ...............135
5.8. Male M. –decim movement direction following artificial wing flick
signal...........................................................................................................136
5.9. Experiments on male chorusing behavior...................................................137
5.10. M. septendecim wing-flick signals responses to frequency-modified
playbacks of male song...............................................................................138
5.11. Contrasts of female wing flick responses to playbacks of complete
artificial calling songs and partial artificial calling songs ..........................139
5.12. Effects of background chorus intensity on female wing flick responses
to playbacks of complete artificial calling songs and partial artificial
calling songs ...............................................................................................140
5.13. Effects of M. -decim “interference buzz.” ..................................................141
6.1. Traits distinguishing Magicicada neotredecim and other Magicicada
species.........................................................................................................184
6.2. Relationship of abdomen color to male call pitch and female pitch
preference....................................................................................................185
xiii
6.3. Dominant frequencies of Magicicada septendecim chorus recordings
in the UMMZ sound library........................................................................186
7.1. Tests for normality of signed, scaled asymmetries.....................................224
7.2. Tests for measurement error and handedness .............................................225
7.3. Numbers of cicadas used in analysisof symmetry and mating success ......226
7.4. Pearson pairwise correlation matrix for character asymmetry....................227
7.5. Pearson pairwise correlation matrix for character size ...............................228
7.6. Pearson pairwise correlation matrix for character sizes and asymmetry....229
7.7. Comparison of asymmetry of mated and unmated males for each
character......................................................................................................230
7.8. Results of tymbal rib asymmetry analysis ..................................................231
7.9. Results of resampling algorithm to evaluate whether males with wing
asymmetries were as likely to mate as normal males. ................................232
8.1. Lengths of Okanagana canadensis calling and silent bouts......................256
8.2. Mating durations in Okanagana and Magicicada septendecim..................257
8.3. Responses of male cicadas to presentation of model cicada.......................258
xiv
LIST OF FIGURES
Figure
1.1. Distribution of male traits in relation to female threshold acceptance
value..............................................................................................................28
1.2. Comparison of threshold acceptance criteria and best-of-n choice ..............29
2.1. Survivorship of marked and unmarked cicadas ............................................52
3.1. Photograph of gel showing RAPD-PCR amplification. ...............................69
4.1. Numbers of eggnests laid by female M. septendecim...................................96
5.1. Stylized sonogram of male call/female wing flick coutship duet for M.
-decim .........................................................................................................142
5.2. Sonogram of male M. -decim call and female wing flick response............143
5.3. Sonogram of male M. -cassini call and female wing flick response ..........144
5.4. Sonogram of male M. -decula call and female wing flick responses .........145
5.5
Comparison of male responses to a model female that moves only ...........146
5.6. Male responses to actions of model female that moves and makes
sounds .........................................................................................................147
5.7. Sonogram of response of M. -decim male with Stage I Massospora
cicadina infection to an M. -decim call ......................................................148
5.8. Sonogram of response of M. -cassini male with Stage I Massospora
cicadina infection to an artificial M. -cassini call.......................................149
xv
5.9. Responses of female M. septendecim to playbacks of artificial, pure
tone calls and portions of calls against an artificial background chorus ....150
5.10. Sonogram of male M. -decim call and interference buzz of nearby
male.............................................................................................................151
6.1. Geographic distribution of the seven periodical cicada species .................187
6.2. Comparison of typical 17- year Magicicada chorus with 13- year
chorus containing M. tredecim and M. neotredecim...................................188
6.3. Spectrogram showing a two-banded, mixed-species chorus ......................189
6.4. Air temperature effects on M. -decim chorus frequency ............................190
6.5. Sonogram of male M. neotredecim calls illustrating consistency of
individual male calling song pitch ..............................................................191
6.6. Power spectrum of mixed M. neotredecim and M. tredecim chorus
with accompanying frequency histograms of male song pitches and
female pitch preferences .............................................................................192
6.7. Relative proportions of M. neotredecim and M. tredecim estimated
from chorus recordings of the 1998 emergence of Magicicada 13-year
Brood XIX. .................................................................................................193
6.8. Geographic variation in dominant chorus pitch of M. neotredecim,
showing reproductive character displacement in and near the region of
overlap with the low-pitched M. tredecim. .................................................194
6.9. Schematic representation of calling song displacement in Brood XIX
M. tredecim and M. neotredecim ................................................................195
6.10. Geographic variation in dominant chorus pitch of M. tredecim,
suggesting weak reproductive character displacement...............................196
6.11. Box plots of male calling song pitch, right wing length, thorax width,
and first sternite width of M. tredecim (Sharp Co., AR) and M.
neotredecim (Sharp Co., AR. and Piatt Co., IL).........................................197
6.12. Formation of an incipient Magicicada species by “nurse brood
facilitation” of life cycle mutants................................................................198
xvi
6.13. A model of Magicicada speciation using nurse brood facilitation and
reinforcement of premating isolation..........................................................199
6.14. Two hypotheses for the origins of Midwestern 13- year cicadas ...............200
6.15. A model of Magicicada life cycle evolution via canalization of a
climate-induced life cycle shift...................................................................201
7.1. Characters used in analysis of phenotypic symmetry .................................233
7.2. Comparisons of cumulative asymmetry scores for mated and unmated
males ...........................................................................................................234
8.1. Locations of study sites, Okanagana canadensis, O. rimosa 19961998 ............................................................................................................259
8.2. 1996 distribution map of O. rimosa and O. canadensis ............................260
8.3. 1997 distribution map of O. rimosa and O. canadensis ............................261
8.4. 1998 distribution map of O. rimosa and O. canadensis ............................262
8.5. Location of O. rimosa study site in 1998....................................................263
8.6. Sonogram of male Okanagana canadensis call..........................................264
8.7. Sonogram of male Okanagana rimosa call ................................................265
8.8. Sketches of lateral view of typical Okanagana male genitalia...................266
8.9. Sketches of lateral view of typical Magicicada male genitalia ..................267
xvii
LIST OF APPENDICES
Appendix
A.
C++ code for mating system simulation .....................................................269
B.
Pascal program code for comparing experimental cage densities to
densities of free cicadas observed on cage-sized branches.........................287
C.
1995 Cage mating experiments...................................................................290
D.
C++ code for resampling program to evaluate remating frequencies in
large and small cages ..................................................................................291
E.
C++ code for seminal plug resampling algorithm ......................................302
F.
Male frequency collection, 1998 Brood XIX .............................................313
G.
Playback experiments, 1998 Brood XIX ....................................................316
H.
Locality, chorus pitch data, 1998................................................................317
I.
Holding jig for symmetry measurements....................................................319
J.
Symmetry and size measurements for mated and unmated cicadas ...........320
K.
Code for resampling algorithm to evaluate whether males with
asymmetrical wing venation were over- or under-represented among
males that did not mate in mating experiments .........................................322
L.
Miscellaneous field notes on Okanagana ...................................................326
xviii
CHAPTER I
MATE CHOICE IN INVERTEBRATES, WITH SPECIAL
REFERENCE TO PERIODICAL CICADAS (MAGICICADA SPP.)
Abstract
Most invertebrates have limited adult experiences and mating opportunities.
Thus, studies of invertebrate mate choice require a different approach from many
vertebrate-based mate choice models, which implicitly allow animals to make use of past
experiences and multiple mating opportunities. Careful consideration of the benefits
realizable from mate discrimination as well as the ways in which discrimination may take
place can reveal whether observed mating patterns are manifestations of evolved mate
choice mechanisms or whether they are incidental effects of other behaviors. For
example, for those insects lacking male parental investment and multiple mating
opportunities, female choice criteria would not include material benefits, and mate choice
mechanisms would be unlikely to involve comparisons. Alternatively, if existing male
variation has minimal fitness consequences for females, then the costs of discrimination
may outweigh the benefits, and any observed mating biases may be incidental rather than
1
the results of mate choice. With this in mind, I describe a series of studies designed to
evaluate whether female periodical cicadas (Magicicada spp.) choose mates and if so,
what choice criteria they are likely to use.
Introduction
Every aspect of behavior, life history, and ecology provides unique raw materials
for the operation of sexual selection. Many vertebrates have extended, juvenile periods
in social contact with conspecifics, multiple opportunities for observing all aspects of
adult behavior before becoming sexually mature, multiple breeding opportunities, and the
possibility of gaining proficiency at acquiring mates. Insects, in contrast, often have
single breeding opportunities (i.e., they are semelparous), no experience with
conspecifics prior to mating, and few opportunities for rectifying errors. Life history
patterns common among invertebrates often have no vertebrate parallels; thus, vertebratebased models of sexual selection and mate choice (Bradbury 1981, 1985, Höglund and
Alatalo 1995) may not accurately characterize invertebrate mating systems.
Rather than revise or modify vertebrate mate choice models, this review provides
a general discussion of how the opportunities for mate choice and the benefits from
choosing create the selective pressures shaping a given species’ mating system. This
general discussion of mate choice leads to specific predictions for how invertebrates
should be expected to choose mates or mating opportunities and concludes with a
discussion of mate choice in periodical cicadas (Magicicada spp.), whose lek-like mating
aggregations suggest a history of strong sexual selection.
A caveat
Mate choice itself may be difficult to observe directly, so behaviorists usually
measure correlates of choice, such as polygyny. The hazards of assuming that observed
mating patterns such as extreme polygyny necessarily result from choice are illustrated
2
by the following example: Suppose that in a population of 50 males and 50 females,
males are always sexually receptive, females become receptive sequentially (thus the
operational sex ratio is always 50 males:1 female), and females mate with the first male
encountered. This scenario, written as a C++ computer program (Appendix A) and run
for 10,000 iterations, leads to the following results: First, counting the maximum number
of matings achieved by any male in each iteration, the 95th percentile is 5 matings. That
is, in 9,500 of the iterations, at least one male achieved 5 matings and was notably more
polygynous than others. In rare cases, males received up to 9 matings, and invariably,
some males received no matings. At first glance, the population appears to be
characterized by a 50:50 sex ratio, because a random sample of individuals drawn from
the population would contain equal numbers of males and females. Thus, it would be
tempting to conclude that the mating skew results from some form of mate choice.
However, the operational sex ratio is male-biased, because at any given time only one
female is receptive. The skewed operational sex ratio causes some males, by chance, to
be much more successful than others even though females mate indiscriminately (see
Boggs 1995 for a similar scenario). Some males were more successful than others simply
because they were, by accident, in the right place at the right time.
This example, in which a strongly male-biased operational sex ratio leads to
polygyny, highlights what should be the first question in any study of mate choice: Are
the animals involved behaving such that they receive fitness benefits from participating in
mate choice, or are some patterns better explained as spurious or incidental effects? The
answer to this question lies in understanding the benefits possible from choice and the
ways in which animals might obtain them.
Mate choice
The distinction between mate choice and its alternatives corresponds to Darwin’s
original distinction between sexual and natural selection (Table 1.1; Darwin 1859, 1871,
3
Arnold 1983). As with Darwin’s forms of selection, mate choice and its alternatives are
not easily separable. Mate choice results from individual actions that reduce the
probability of mating with some members of the opposite sex (Parker 1983a). Mate
choice includes both direct and indirect choice (see below), or all circumstances in which
the identity of a mating partner affects fitness, either through genetic materials provided,
or through potential mates’ differential ability to deliver resources or appear at times or
locations most appropriate for mating (see Halliday 1983). In choosy species, members
of the chosen sex can take actions that increase their own chances of mating success. The
alternatives to mate choice are context choice and haphazard mating. Context choice
occurs when fitness benefits result from restricting mating to particular contexts, but
potential mates have no differential abilities to appear in the appropriate context and
cannot take actions to increase their mating success; thus mating is independent of
partner identity. Under haphazard mating, differences in mating context or partner
identity have no influence on fitness.
Why choose mates at all?
Each sex strives to defend its post-zygotic investment against preemption or
usurpation by members of the other sex (see Wiley and Poston 1996). Post-zygotic
investment is the “parental investment” of Trivers (1972; also Alexander and Borgia
1979) and refers to any investment that is irretrievably committed to a particular zygote
or collection of zygotes. In contrast, pre-zygotic investment is mating effort or resources
committed to gametes (or collections of gametes) or to the production and/or delivery of
gametes before fertilization. As with any artificial dichotomy, it is not always easy to
distinguish pre- and post-zygotic investment. Even the term “post-zygotic investment” is
misleading; the terms “investment in zygotes,” or “zygotic investment” are more
accurate, because they refer to investment that has its most profound effects after
fertilization, although it may be dispensed at any time before or after zygote formation
4
(Alexander and Borgia 1979). Variations in zygotic investment directly affect offspring
quality. Thus, parental care by either sex is investment in zygotes, because it is delivered
to particular offspring. Such care may actually be dispensed before the zygote is formed;
for instance, egg yolk is investment by a female in her zygotes, because variation in the
amount or quality of the yolk affects a particular zygote’s likelihood of success. In
contrast, male courtship is mating effort, or investment in gametes, not zygotes, because
different levels of investment affect only the likelihood that a male will, by delivering
gametes, produce a zygote, not the success of any particular zygote. Like any resource,
the zygotic investment of one sex is available for use and exploitation by members of the
other.
Control of fertilization
Each sex strives to control the act of fertilization in order to defend its zygotic
investment against preemption or usurpation. The costs of both finding additional mates
and of zygotic investment influence sexual choosiness, but these factors are not
independent. As members of one sex increase their zygotic investment, members of the
other sex seek them out, so high zygotic investment is correlated with low mate search
costs (Parker 1983a). Thus, the sex with the greater zygotic investment has the greater
incentive to protect its investment and avoid mating indiscriminately, because its
reproductive output is limited more by cost of the investment than by the cost of finding
mating partners, while the sex with the lesser zygotic investment is limited by its ability
to find willing mates and cannot afford to forego mating opportunities.
In many sexual species, females initially have the greater zygotic investment; they
produce large, well-provisioned gametes specialized for post-zygotic success (Parker et
al. 1972, Alexander and Borgia 1979). Delivery of resources to gametes makes the
resources available to zygotes immediately after fertilization (Alexander and Borgia
1979). Associated with gamete provisioning is physical control and protection of the
5
gametes, probably for the following reasons: Females that evolved to dedicate greater and
greater amounts of resources to zygotes also evolved to protect their investment by
increasing physical control over them, such as by retaining them within the body after
fertilization. As females increased their control, they gained confidence of the maternity
of the zygotes they invested in and benefited by investing even earlier, in gametes. Thus,
for many females, the distinction between prezygotic and zygotic investment is blurred,
because investment in gametes translates directly into investment in zygotes. As females
gained control over gametes, such as by retaining them inside their bodies for as long as
possible, they also increased their control over the fertilization process.
Conflicts of interest between the sexes
The sexes’ relative levels of post-zygotic investment (which, as noted above, are
related to confidence of parenthood) affect the degree of conflict over mating decisions.
If the strategies of the sexes are identical, e.g., if both sexes have substantial zygotic
investment, or if neither sex invests substantially and therefore neither discriminates, then
there is little conflict over investment (Trivers 1972, Alexander and Borgia 1979, Parker
1979). When one sex invests more in zygotes than the other invests, the sexes will be in
conflict over mating decisions (Parker 1979). Sexual conflict will, in each sex, favor the
evolution of adaptations for controlling fertilization and thwarting the ability of the other
sex to do so (see Rice 1996). Because the optimal mating strategy for each sex is
contingent upon the other’s strategy, the sexes evolve in response to each other, with both
sexes using a variety of strategies to gain control over fertilization (Parker 1979, 1983b,
Alexander et al. 1997). Females with internal fertilization have an immediate control
advantage; male responses to this advantage include genitalia that place sperm in close
proximity to a female’s eggs (Parker 1970, Alexander et al. 1997), contributed substances
that alter female physiology (see Lum 1961; Riemann and Thorson 1969; Burnet et al.
1973; Leopold 1976; Scott 1986; Chen et al. 1988; Kingan et al. 1993; Gromko,
6
Newport, and Kortier 1984; Gromko, Gilbert, and Richmond 1984, Wolfner 1996;
Eberhard 1997 pp. 37- 40), or any number of other physiological or behavioral
adaptations. Each sexes’ adaptations manipulate the other’s mating costs and benefits to
effect control over fertilization. For instance, a male might restrict access to some
valuable resource (such as a nesting site), trading resource access for matings. If the cost
to a female of losing control of fertilization is less than the cost to her of losing access to
the resource, then she will accept the trade. Each sexes’ adaptations to gain control may
promote counter adaptations, leading to evolutionary races in which a wide range of
strategies and possible counter strategies contribute to the complexity of the mating
system (Parker 1983b). By inference, any species characterized by complex sexual
interactions during sexual rapprochement likely exhibits some form of mate choice,
because elaborate sexual interactions suggest that members of the chosen sex can take
actions to improve their odds of mating. It is the mutual contingency of the sexes’
mating strategies, a consequence of mate choice, that leads to complex sexual adaptations
(Alexander et al. 1997).
Selection and mate choice
Mate choice is an adaptation for producing genetically superior offspring,
acquiring direct phenotypic benefits convertible into reproductive output, or both.
Genetic benefits from choice fall into two non-exclusive categories: (1) Arbitrary benefits
or (2) Offspring quality benefits. Arbitrary benefits consist of the genes coding for
ornaments or other traits enhancing sexual attractiveness. Offspring quality benefits
include the genetic materials necessary for producing offspring well equipped to survive
and reproduce.
For example, in a given habitat, male Colias butterflies with the most appropriate
enzyme variants not only gain from efficiencies in flight and foraging, but they are also
more able to produce long and successful courtship displays (Watt et al. 1986). Females
7
mating with the most actively displaying males produce offspring that realize flight and
foraging benefits and sons that produce more attractive sexual displays (Weatherhead and
Robertson 1979; but see Kirkpatrick 1985). In this case, choosy females gain genetic
benefits with aspects of both arbitrary and offspring quality benefits.
Females may also discriminate among males in order to receive phenotypic
benefits. Such benefits include time and effort conserved by mating with easily located
males, acquisition of male-provided nuptial resources, protection from injury or other
harm, and reduced costs, or at least minimization of losses, associated with acquiescence
to persistent male harassment. To the extent that male ability to provide phenotypic
benefits is heritable, female choice for such benefits will impose selection on males, and
thus phenotypic and genetic benefits are not sharply distinct.
Mechanisms of mate choice: Direct vs. indirect
Mate choice involves either direct assessment of potential mates or indirect biases
(Wiley and Poston 1996, Sullivan 1989). In species with female choice, direct choice
requires individual recognition and memory of male characteristics, and it requires
females to perceive their different mating options. Indirect choice requires that the
highest quality males have stereotypical characteristics, and that females act in ways that
discriminate against males not exhibiting such qualities even though they may not have
perceived the different mating options available to them.
Females may favor a particular male indirectly, rather than directly, on the basis
of several possible cues:
1. Females may be attracted to the loudest or most easily located signals apart from
any comparisons among those signals (The “passive attraction” hypothesis;
Parker 1983a).
2. Females may accept only those males that have dominated in a contest with others.
Male-male contests, in effect, provide résumés of the competitors’ lifetime
achievements (Borgia 1979). A female may be absent for the contest itself, but
the behavior of the contestants after the contest restricts her mating options. For
8
example, Burk (1983) discusses how losers in male-male contests in the field
cricket Teleogryllus oceanicus reduce their calling; thus, females are more likely
to mate with the victors. Moore and Moore (1988) present similar results for the
cockroach Nauphoeta cinerea.
3. Females may favor males on the basis of some trait or resource. A female may
feed on a copulatory gift or secretion, breaking off the mating when the gift is
gone and thus disfavoring males with inadequate or poor-quality gifts. Thornhill
(Thornhill and Alcock 1983) discusses how male scorpionflies providing inferior
nuptial gifts are less successful as mates, because females stop mating once they
have consumed the gift.
4. Females may engage in behaviors that discriminate among males according to
their relative performance with respect to some environmental cue. Females may
accept only males who, because of particular sensory, locomotory, or
thermoregulatory abilities, are able to be in the right place at the right time.
Honeybee mating swarms may be an example of such mate discrimination (see
Winston 1987).
Others have classified these possibilities differently; for instance, Lloyd’s “passive
filtering” would include aspects of 1 and 4 above; his “passive selection” is most similar
to 2; “active filtering includes 1 and 4; while “active selection” includes aspects of 2, 3,
and 4 (Lloyd 1979).
At the crudest level, all females exercise some degree of “indirect” mate choice
by successfully mating only with living, sexually mature males (Parker 1979). It is trivial
to point out that selection thus favors females who make this distinction, but this
distinction is hardly the result of any specially evolved or elaborated adaptation. To the
extent that more subtle male variation is heritable and has fitness consequences for
females, selection can reinforce both sexes’ phenotypic cues and predictable behaviors
involved in pairing, leading to evolved mate choice mechanisms.
Mate choice criteria
Mate choice criteria fall on a continuum of two extremes: (1) Females may,
actively or passively, make relative comparisons among available mating partners (“bestof-n” choice, Janetos 1980), or (2) Females may accept as mates males who meet or
exceed internally generated criteria (“threshold” or “minimal criterion” choice; Janetos
9
1980). Each of these models leads to distinct predictions about the kinds of behaviors
males and females should exhibit (Janetos 1980, Alexander et al. 1997). Note that mate
choice criteria do not constrain the evolution of mate choice mechanisms (either direct or
indirect), and all four permutations of mate choice mechanisms and choice criteria are
theoretically possible.
Direct relative comparisons: Best-of-n choice. Females exercising
best-of-n choice seek to identify and mate with the best available male in a group of nmales. Best-of-n choice criteria are relative, open-ended, and require a comparison of
potential mates. As long as such relative comparisons do not exact prohibitive costs, and
as long as female fitness covaries with incremental changes in the resources or genetic
materials provided by mates, selection will favor females using best-of-n choice. Bestof-n choice is functionally equivalent to threshold choice (see below) in which the
acceptance criteria are changeable and influenced by the kinds of males a female
encounters. The mechanisms involved in best-of-n choice need not be complex: Females
could choose the best mate of those available by setting their acceptance threshold to the
value of the first male encountered, raising it after encountering a higher quality male,
and mating with the highest quality male when the threshold becomes fixed or remains
stable after repeated encounters. Male-male contests could also restrict sexual access of
all but the best males present. Females may require potential partners to simultaneously
perform tests of skill to reveal relative quality. High costs, in terms of time and effort,
and limited opportunities for making comparisons make short-lived, semelparous animals
such as insects unlikely to exercise best-of-n choice (Alexander et al. 1997).
“Indirect” or “non-simultaneous” relative comparisons. Males,
through active, intrasexual competition, can sort themselves in ways that limit sexual or
resource access by lesser quality individuals (Cox and LeBoeuf 1977) and thus females’
mating options will be restricted as if they were exercising best-of-n choice. Females
10
making indirect, relative choices do not require simultaneous tests of skill; instead, they
create situations in which males compete with each other before mating, and they will
mate readily even if only one male is present.
Some “lekking” insects may provide examples of indirect best-of-n choice: If
male insects form aggregations through which receptive females fly, and if males within
an aggregation chase, catch, and mate with the females, these behaviors place premiums
on male flight and perceptual capabilities and cause male mating success to covary with
male quality. Lloyd (1979) points out how, by imposing sexual selection on such traits,
females will cause males to compete to demonstrate their prowess (p. 337). If more than
one male is present in such a “lek,” it provides an arena in which male competitive
interactions sort males on the basis of quality. Insects that engage in multi-male
courtship chases, such as honeybees (see Winston 1987 p. 207), midges (McLachlan and
Cant 1995), some Lepidoptera (Sullivan 1981, Pliske 1975) and syrphid flies (see
Chapman 1954, Cooley “hotrod” model, unpublished) may all have mating systems
characterized by this kind of choice.
Threshold or minimal criterion choice. Females exercising threshold
choice discriminate against a class of potential mates that fail to meet minimal acceptance
criteria (Janetos 1980). Males meeting the criteria are all equally acceptable, regardless
of differences among them. Females acquire the choice criteria without regard to the
characteristics of the male population, and once acquired, the criteria are fixed and
unchangeable. The feature most distinguishing best-of-n from threshold choice is that
under threshold choice, all acceptable males are equally acceptable independent of other
qualities, and differences among acceptable males have no appreciable fitness
consequences for females, while under best-of-n choice, in a given group of males, there
is one best mate, because each male’s acceptability as a mating partner depends both on
his own qualities and on the characteristics of the population at large. Some mate choice
11
mechanisms, such as “one step” or movable threshold mechanisms may appear to involve
threshold choice, but from the standpoint of sexual selection they are more similar to
best-of-n than to threshold choice because they involve relative comparisons and
“thresholds” generated only after females sample the male population (Janetos 1980).
If threshold choice criteria exclude a large proportion of the males in the
population, then females with lower acceptance thresholds will locate mates more readily
than females using arbitrarily high criteria (Fig. 1.1). This increased efficiency may
provide phenotypic benefits compensating any loss in mate quality, promoting an
increase in the proportion of females with lower criteria in future generations. Thus,
arbitrarily high criteria are unlikely to persist, and threshold choice is unlikely to lead to
the extremes of sexual selection possible under best-of-n choice.
If male variation with respect to mate choice criteria is heritable, and if female
choice criteria are unchanging and uniform across all females, then over the course of
several generations, unless mutation rates are extremely high, threshold-based mate
choice should diminish the proportion of unacceptable males, because all offspring will
have fathers that met or exceeded the choice criteria. The exception, in which a long
history of threshold choice would not lead to the elimination of unacceptable variants, is
the case of species discrimination by threshold choice, because all the unacceptable mates
would belong to a different species. Within a species, males subjected to threshold
choice would be selected to exceed the acceptance threshold to the smallest extent that
ensures that they will mate and that their offspring also exceed the threshold. After
several generations of threshold choice, females would face a pool of potential mates
most of whom slightly exceed the acceptance threshold, but among whom there would be
few consequential differences. Thus, differences in male quality could be trivial in the
face of the direct advantages females might gain from choosing mating times and
locations that maximize their reproductive output. In species for which this is true, there
12
would be little evidence of mate choice; instead, females would mate so as to maximize
direct benefits (such as associated with time or place of mating) to themselves while the
identity of their mate would be largely irrelevant.
Many studies of insect behavior seem to suggest threshold choice. Scorpionflies
(Hylobittacus apicalis ) obtaining nuptial gifts above a certain size threshold are accepted
by females, while those offering smaller gifts are rejected or allowed only brief
copulations with little or no chance of paternity (Thornhill 1980, Thornhill and Alcock
1983 p. 367). Females do not differentiate among males providing acceptable gifts, but
appear to reject mates on the basis of a fixed gift size criterion or threshold.
Moore and Moore (1988) demonstrated that female roaches (Nauphoeta cinerea )
use absolute, rather than relative, preferences when mating. Male roaches formed stable
dominance hierarchies, and females appeared to make best-of-n choices, because they
decreased courtship time and increased copulation time as male rank increased.
However, female behavior was independent of the number of males present, and females
deprived of any previous experience with males mated with the first male presented.
Thus, females could not have made relative comparisons among dominant and
subordinate males, and their mating behavior appears to involve an acceptance threshold,
which all males in the experiment met or exceeded.
The action of an absolute mate acceptance threshold in Nauphoeta cinerea does
not explain fully why female behavior was different for males of different dominance
rank. Some aspects of male behavior provide clues: For example, subordinate males
generally delay longer before initiating courtship (Moore and Breed 1986), courtship and
mating duration are influenced by quantity and quality of male pheromone (Breed et al.
1980, Moore 1988), and these effects are independent of male-male competition.
Furthermore, when females are presented with mating partners sequentially instead of
simultaneously, they do not appear to discriminate (Breed 1983). The behavior of
13
females presented with solitary males is evidence for threshold choice; however,
superimposed on this choice is another factor: Inherent qualities of dominant males
apparently cause them to be more easily accepted or perceived by females, independent
of the action of female acceptance criteria.
Passive attraction. “Passive attraction” (Parker 1983a) blurs the distinction
between relative and threshold choice mechanisms. Under the passive attraction
hypothesis, females may have an acceptance threshold, and males act in ways that ensure
they will meet or exceed that threshold. If males employ mate attraction signals or
complex courtship, the more readily females perceive such signals, the more likely they
are to mate (Parker 1983a). Thus, females may tend to approach signaling males who are
more obvious without directly making a relative comparison. For example, if two males
were to sing at equal volume, one near a female and one far, then we would predict that
the female should approach the nearest one. Similarly, if two males were at equal
distance from a female, but one produced a louder signal than the other, the female
should approach the louder male. This “passive attraction” can appear to be best-of-n
choice if two males court simultaneously, because the female will respond only to the
more apparent male. In reality, the female perceives only one of the males, and is not
actually choosing. Animals assorting by passive attraction can appear as though they are
selecting mates, but any choice is indirect at best. It is possible that the mate sorting
qualities of female perception are incidental and provide no selective benefits over and
above the benefits of being able to detect mates at all. Males, however, will be subject to
selection, to the extent that male ability to make use of appropriate advertisement times,
locations, channels, and intensities is heritable (Alexander et al. 1997). Although passive
attraction in some ways sorts males relatively, it does not necessarily lead to male-female
antagonistic coevolution because the sexes do not evolve strategies in response to each
other. Instead, one sex evolves strategies to cope with the unchanging problem of being
distinguishable from its surroundings or with the problem of prevailing in competition
14
with members of its own sex. However, if increased male display intensity thwarts
female interests, intersexual competition becomes a more important selective force, and,
rather than passive attraction, the mating system may be better characterized by a sexual
conflict-based model such as chase-away selection (Holland and Rice 1998).
Jang and Greenfield (1996) demonstrated that female wax moths (Achroia
grisella) were most likely to orient towards louder male signals, male signals with more
rapid pulses, and male signals with longer pulses. Thus, the more evident a male’s
signal, the greater the chance of a female response. Sometimes both a mate acceptance
threshold and “passive attraction” seem to be involved in mate choice, as suggested by
the example of Nauphoeta cinerea, (see above), as well as by male wing amputation
experiments in fruit flies (Drosophila melanogaster ; Robertson 1982), in which the
greater the area of wing removed, the longer the courtship duration before copulation.
Male Drosophila without wings rarely mated, but as long as males had some wing area
remaining and were able to produce some song, they eventually mated. A threshold, in
which females require males to produce some song before mating, could explain why
wingless males failed to mate, and female difficulty in perceiving quiet males could
explain why partially amputated males courted longer before copulating.
Conflicts of interest and timing of mate choice
Mate choice could occur at any stage of the sexual sequence, including after
copulation. It is in the interests of the choosy sex (usually the female) to effect mate
choice as early in the mating sequence as possible. Early in the sequence, she retains
substantial control over the mating decision; however, she also has little information on
which to base her actions (Alexander et al. 1997). Thus, females should seek to cause
males to reveal information about themselves before males have an opportunity to obtain
a mating by force. Because the effectiveness of forcing increases as male and female
come into closer proximity and more intimate contact, it is in the male’s interests either
15
to: (1) convince the female to commit to him early in the sequence, or (2) delay the
female’s final mating decision until he approaches and has a high probability of
compelling a mating in the event the female rejects him. Males should be selected, in so
far as is possible, to refrain from including in calling songs or other sexual
advertisements any information that could jeopardize their ability to approach or be
approached closely by females. Thus, there is a conflict of interest between male and
female over the nature and timing of information to be transmitted, and there is also a
conflict over when in the sequence the female should make her final mating decision.
Multiply-mating females could take actions that affect paternity after they have
engaged in one or more matings. However, if a female must copulate in order to gather
information to make her mating decision, she places copulating males in a strong position
to force fertilization upon her (Sivinski 1984). Because of the threat of forcing,
postcopulatory mate choice should be restricted to rare cases in which females can, with
certainty, separate copulation success and fertilization success (Eberhard 1985). Females
exercising such choice should posses special adaptations for storing, manipulating, or
discarding sperm (see Hellreigel and Ward 1998), or they should exhibit a strict pattern
of last-male sperm priority so that they can retain control over fertilization by restricting
sexual access.
General predictions about mate choice mechanisms
Females choosing the best-of-n available mates can favor rare, novel mutants for
no reason other than their novelty or extremeness. The sexual selection advantages of
such traits can cause them to spread even if they are costly in terms of natural selection.
In contrast, sexual selection under threshold choice sets a lower limit for mate
acceptability, and there is little or no mating advantage to exceeding that limit; for
equally acceptable males facing threshold choice, novelty is at best irrelevant, and, since
any costs are unopposed by sexual selection, novelty may be a liability (Fig. 1.2). One
16
exception is that threshold choice can favor novel traits that enhance a male’s ability to
be detected or noticed (passive attraction); females favoring such traits are usually not
favoring novelty per se, but augmentation of existing mate location and recognition
mechanisms instead. For similar reasons, in species where the only benefits from mate
choice are genetic, threshold choice is unlikely to evolve because only best-of-n choice
can grant rare, novel mutant males a selective advantage allowing the spread of the
mutation. Thus, best-of-n choice is a likely prerequisite of “runaway” evolution and the
evolution of sexual ornaments.
Invertebrates’ general lack of male resource investment would seem to place a
premium on females’ abilities to discriminate among potential mates (Emlen and Oring
1977, Alexander and Borgia 1979, Andersson 1994, Höglund and Alatalo 1995). Any
benefits from discrimination are likely to be genetic, and obtainable only through some
form of relative comparison; further, invertebrate life history characteristics make it
likely that such comparisons are direct. Even if males provide no nuptial resources, and
even if females obtain no genetic advantages for their offspring by selecting mates,
females could still materially benefit by mating at a time or location optimal for their own
interests, or choosing a mating context.
Mate choice in periodical cicadas
13- and 17- year periodical cicadas (Magicicada spp.1) provide an unparalleled
opportunity to examine mechanisms of mate choice. Male Magicicada join mixedspecies singing aggregations attractive to sexually receptive females. Males in these
aggregations engage in chorusing (or “sing-fly”) behavior (Alexander 1968, 1975), in
which they produce a small number of calling phrases (1- 3 in the M. –cassini and M.
1
For convenience the Magicicada cognate species groups are referred to using the following shorthand:
M-decim: M. septendecim (17), M. tredecim (13), and M. neotredecim (13); M -cassini: M. cassini (17) and
M. tredecassini (13); M. -decula: M. septendecula (17) and M. tredecula (13).
17
–decim species or a short calling bout in the M. –decula species) and then fly to another
calling perch. Upon hearing a male’s calling phrase, a receptive female flicks her wings,
producing a visual and acoustical signal similar to that reported in other cicada species
(Chapter 5). Females signal only when they are mature and ready to mate, most signaling
females copulate within several minutes, and most male mating attempts in the absence
of a signal fail. A male perceiving a wing-flick signal ceases sing-fly behavior and
begins courtship, engaging in an acoustical duet with the signaling female while moving
towards her. Most females mate only once (Chapter 4).
Alexander (1975) considered periodical cicada male singing aggregations to be
non-resource based leks, in which females have opportunities to discriminate among
mates. Because unmated female Magicicada eventually oviposit normal-appearing,
presumably unfertilized eggs (Graham and Cochran 1954, pers. obs.), it is unlikely that
females visiting male aggregations receive male donated resources (or gain access to
resources other than sperm) necessary for reproduction. Magicicada females could
choose among mates at several stages of the sexual sequence:
(1). They could choose during early stages of courtship when males call and
females answer.
(2). They could choose during late stages of courtship when males and females
approach and achieve physical contact.
(3). They could exercise postcopulatory choice by remating with higher quality
males should they become available.
Any benefits females receive as a direct consequence of discriminating among potential
mating partners must be genetic (because males provide no material resources), or must
include phenotypic benefits other than male-donated gifts, such as reductions in the
immediate costs of copulation by reducing time and effort wasted in locating appropriate
mates or reducing risks of disease, damage, or predation. Alternatively, instead of
choosing among potential partners, female cicadas may gain more from mating to
18
maximize direct phenotypic benefits to themselves, such as by selecting an appropriate
time and safe location for mating independent of male identity.
Female discrimination against heterospecific males. The Magicicada
species M. -decim, M. -cassini, M. –decula are sympatric over most of their ranges
(Alexander and Moore 1962). Because females respond sexually only to conspecific
male calls (Chapters 5, 6), hybrid matings are unlikely. Females avoiding hybrid matings
conserve time and resources that would otherwise be wasted in fruitless matings. If
female Magicicada mate choice behaviors evolved in the context of hybrid avoidance,
and if the differences in the Magicicada species’ sexual behaviors result in part from past
episodes of character displacement (Chapter 6), threshold choice mechanisms, based
upon species-typical behaviors such as courtship, would be the simplest and least costly
means for females to discriminate against inappropriate mates. Because there is little or
no sexual conflict of interest over species identity if individuals of neither sex realize
appreciable fitness benefits from cross-species matings, if females discriminate only
against heterospecifics, males should be selected to cooperate instead of deceiving
females.
Female discrimination against diseased males. Magicicada are
vulnerable to an entomophagous fungal infection (Massospora cicadina Peck; Soper
1974, Soper et al. 1976). It is possible that the tendency or predisposition to become
infected is inheritable. Infected cicadas are contagious and, in advanced stages of the
infection, sterile. One simple way that females could reject infected mates is to fail to
respond to their calls, or, if the fungus does not become contagious until males cease
normal calling, to refuse to mate with silent males.
Even if there are no genetic differences among males concerning ability to resist
or survive infection, the direct phenotypic benefits to females of avoiding infected males
are substantial. Females should avoid contagious, infected males at all costs and should
19
reject infected males as early in the mating sequence as possible. Even if males exhibit
no heritable genetic variation for susceptibility, females could benefit from
discriminating against infected males because the fungus is invariably fatal, it reduces
infected individuals’ fitness, it is avoidable, and females thus receive fitness payoffs from
avoiding it. If this is an important influence on the Magicicada mating system, then
females can avoid a high phenotypic cost by forcing males to engage in behaviors, such
as joining singing aggregations and participating in complex courtship, likely to reveal
male infection status and help females identify risky matings.
Unless female choice criteria are such that they exclude only sterile, fungusinfected males (who, on their own behalf, have no reproductive interest and thus no
conflict of interest with the females) then male and female interests will conflict.
Specifically, until infected males become sterile, they should strive to disguise their
infection, while females should seek to cause males to court in ways that expose
infections (Alexander et al. ms). For example, if females use male song as an
“uncheatable honest indicator” of male quality, then they should be reluctant to mate
unless their suitors sing, they should place a premium on song power output and the
endurance of singing males, and they should discriminate against males with unusual or
atypical songs characteristic of infected males. If infected and contagious males remain
acceptable to females, then, perhaps because of the low frequency of infection, it is likely
that females gain no net selective advantages from expending any effort to discriminate
against rare infected males.
Female choice of mating time and location. Females may receive direct
phenotypic benefits by restricting mating to optimal times and places. For instance,
females may minimize exposure to predation by avoiding small populations and visiting
only the densest aggregations (see Hamilton 1971; Alexander and Moore 1962, Williams
et al. 1993). Females may also minimize risks by mating early in the day and terminating
20
copulation before cooler evening temperatures force them to remain relatively immobile
until morning, or females may avoid mating until they have found a secure perch from
which they are unlikely to fall; copulating pairs on the ground are relatively immobile
and likely exposed to greater risk of predation from nocturnal rodents (pers. obs.). As
long as males and females face different costs for finding mature, healthy mates, their
interests over timing and location of mating will differ. If male periodical cicadas
typically encounter only one or a few potential mates during their lifetimes, then if they
encounter one at a suboptimal time or location, they will persist in courtship. Females, in
turn, will be selected to resist forced matings. The net result will be that male and female
daily activity patterns will differ, with females tending to become unreceptive before
males stop chorusing, and with decreases in activity related to perceived costs of locating
future mates. If female choosiness is related primarily to mating context, partner identity
will be relevant only if males have heritable differences in their abilities to appear in the
appropriate context.
21
Literature Cited
Alexander, R. D. 1968. Arthropods. Pp. 167- 216 in: T. A. Sebeok (ed). Animal
Communication.Techniques of Study and Results of Research. (Indiana
University Press: Bloomington, IN).
Alexander, R. D., 1975. Natural selection and specialized chorusing behavior in
chorusing insects. Pp. 35-77 in: D. Pimentel, ed. Insects, Science, and Society.
(Academic Press, New York).
Alexander, R. D., and T. E. Moore. 1962. The evolutionary relationships of 17-year and
13-year cicadas, and three new species. (Homoptera: Cicadidae, Magicicada).
Misc. Publ. no. 121. Univ. of Michigan Museum of Zoology, Ann Arbor, MI.
Alexander, R. D., and G. Borgia. 1979. On the origin and basis of the male-female
phenomenon. Pp. 417- 440 in: M. S. Blum and N. A. Blum, eds. Sexual selection
and reproductive competition in insects. (Academic Press: New York).
Alexander, R. D., D. C. Marshall, and J. R. Cooley. 1997. Evolutionary Perspectives on
Insect Mating. Pp. 1-31 in B. Crespi, J. Choe, eds. The evolution of mating
systems in insects and arachnids. (Cambridge University Press).
Alexander, R. D., J. R. Cooley, and D. C. Marshall. (ms). A specialized fungal parasite
(Massospora cicadina) modifies the behavior of periodical cicadas (Homoptera:
Cicadidae: Magicicada).
Andersson, M. 1994. Sexual selection. (Princeton University Press: Princeton NJ).
Arnold, S. J. 1983. Sexual selection: The interface of theory and empiricism. Pp. 67- 107
in: P Bateson, ed. Mate Choice. (Cambridge University Press).
Boggs, C. L. 1995. Male nuptial gifts: Phenotypic consequences and evolutionary
implications. Ch. 10 in: S. R. Leather and J. Hardie, eds. Insect Reproduction.
New York: CRC Press.
Borgia, G. 1979. Sexual selection and the evolution of mating systems. Pp. 19- 80 in M.
S. Blum and N. A. Blum, eds. Sexual selection and reproductive competition in
insects. (Academic Press: New York).
Bradbury, J. W. 1981. The evolution of leks. Pp. 138- 169 in: R. D. Alexander and D. W.
Tinkle, eds. Natural selection and social behavior. (Chiron Press: New York).
Bradbury, J. W. 1985. Contrasts between insects and vertebrates in the evolution of male display,
female choice, and lek mating. Fortschr. Zool. 31: 273-289.
22
Breed, M. D. 1983. Cockroach mating systems. Ch. 12 in: D. T. Gwynne and G. K.
Morris, eds. Orthopteran mating systems: Sexual competition in a diverse group
of insects. (Westview Press, Boulder Colorado).
Breed, M. D., S. K .Smith, and B. G. Gall. 1980. Systems of mate selection in a
cockroach species with male dominance hierarchies. Anim. Behav. 28: 130-134.
Burk, T. 1983. Male aggression and female choice in a field cricket (Teleogryllus
oceanicus ): The importance of courtship song. Ch. 5 in: D. T. Gwynne and G. K.
Morris, eds. Orthopteran mating systems: Sexual competition in a diverse group
of insects. (Westview Press, Boulder Colorado).
Burnet, B., K. Connolly, M. Kearney, and R. Cook. 1973. Effects of male paragonial
gland secretion on sexual receptivity and courtship behaviour of female
Drosophila melanogaster. J. Ins. Physiol. 19: 2421- 2431.
Chapman, J. A., 1954. Studies on summit-frequenting insects in western Montana.
Ecology 35: 41-49.
Chen, P. S., E. Stumm-Zollinger, T. Aigaki, J. Balmer, M. Bienz, and P. Böhlen. 1988.
A male accessory gland peptide that regulates reproductive behavior of female D.
melanogaster. Cell 54: 291- 298.
Cox, C. R., and B. J. LeBoeuf. 1977. Female incitation of male competition: A
mechanism in sexual selection. Am. Nat. 111: 317-335.
Darwin, C. 1859. On the origin of species. A facsimile of the first edition, with an
introduction by Ernst Mayr (Cambridge: Harvard University Press, 1964).
Darwin, C. 1871. The descent of man, and selection in relation to sex. With an
introduction by J. T. Bonner and R. M. May (Princeton: Princeton University
Press, 1981)
Eberhard, W. G. 1985. Sexual Selection and Animal Genitalia. (Harvard University
Press: Cambridge, MA).
Eberhard, W. G. 1997. Cryptic female choice. Pp. 32- 57 in B. Crespi, J. Choe, eds. The
evolution of mating systems in insects and arachnids. (Cambridge University
Press).
Emlen, S. T., and L. W. Oring. 1977. Ecology, sexual selection, and the evolution of
mating systems. Science (Wash. D. C. ) 197: 215- 223.
Graham, C., and A. B. Cochran. 1954. The periodical cicada in Maryland in 1953. J.
Econ. Ent. 47: 242- 244.
Gromko, M. H., M. E. A. Newport, and M. G. Kortier. 1984. Sperm dependence, of
female receptivity to remating in Drosophila melanogaster . Evolution 38: 12731282.
Gromko, M. H., D. G. Gilbert, and R. C. Richmond. 1984. Sperm transfer and use in the
multiple mating system of Drosophila. In: R. L. Smith, ed. Sperm competition
and the evolution of animal mating systems. (New York: Academic Press).
23
Halliday, T. R., 1983. The study of mate choice. Pp. 3- 32 in: P Bateson, ed. Mate
Choice. (Cambridge University Press).
Hamilton, W. D. 1971. Geometry for the selfish herd. J. Theor. Biol. 31: 295-311.
Hellriegel B., and P. I. Ward. 1998. Complex female reproductive tract morphology: Its possible
use in postcopulatory female choice. J. Theo. Biol. 190: 179-186.
Höglund, J. and R. V. Alatalo. 1995. Leks. (Princeton University Press: Princeton NJ).
Holland, B., and W. R. Rice. 1998. Perspective: Chase-away sexual selection:
Antagonistic seduction versus resistance. Evolution 52: 1-7.
Janetos, A. C. 1980. Strategies of female mate choice: A theoretical analysis. Behav.
Ecol. Sociobiol 7: 107-112.
Jang, Y., and M. D. Greenfield. 1996. Ultrasonic communication and sexual selection in
wax moths: Female choice based on energy and asynchrony of male signals.
Anim. Behav. 51: 1095- 1106.
Kingan, T. G., P. A. Thomas-Laemont, and A. K. Raina. 1993. Male accessory gland
factors elicit change from “Virgin” to “Mated” behaviour in the female corn
earworm moth Helicoverpa zea. J. Exp. Biol. 183: 61- 76.
Kirkpatrick, M. 1985. Evolution of female choice and male parental investment in
polygynous species: The demise of the “sexy son.” Am. Nat. 125: 788-810.
Leopold, R. A. 1976. The role of male accessory glands in insect reproduction. Ann.
Rev. of Emtomol. 21: 199-221.
Lloyd, J. E. 1979. Sexual selection in luminescent beetles. Pp. 293- 342 in: M. S. Blum
and N. A. Blum, eds. Sexual selection and reproductive competition in insects .
(Academic Press: New York).
Lum, P. T. M. 1961. The reproductive system of some Florida mosquitoes. II. The male
accessory glands and their role. Ann. of the Ent. Soc. of Amer. 54: 430- 433.
McLachlan, A., and M. Cant. 1995. Small males are more symmetrical: Mating success
in the midge Chironomous plumosus L. (Diptera: Chironomidae). Anim. Behav.
50: 841- 846.
Moore, A. J. 1988. Female preferences, male social status, and sexual selection in
Nauphoeta cinerea. Anim. Behav. 36: 303- 305.
Moore, A. J., and M. D. Breed. 1986. Mate assessment in a cockroach, Nauphoeta
cinerea Anim. Behav. 34: 1160- 1165.
Moore, A. J., and P. J. Moore. 1988. Female strategy during mate choice: Threshold
assessment. Evolution 42: 387- 391.
Parker, G. A., 1970. Sperm competition and its evolutionary consequences in the insects.
Biological Reviews 45: 525- 567.
24
Parker, G. A., 1979. Sexual selection and sexual conflict. Pp. 123- 167 in M. S. Blum
and N. A. Blum, eds. Sexual selection and reproductive competition in insects .
(Academic Press: New York).
Parker, G. A., 1983a. Mate quality and mating decisions. Pp. 141-166 in: P Bateson, ed.
Mate Choice, (Cambridge University Press).
Parker, G. A. 1983b. Arms races in evolution—and ESS to the opponent-independent
costs game. J. Theor. Biol. 101: 619-648.
Parker, G. A., R. R, Baker, and V. G. F. Smith. 1972. The origin and evolution of gamete
dimorphism and the male-female phenomenon. J. Theor. Biol. 36: 529-553.
Pliske, T. E. 1975. Courtship behavior of the Monarch Butterfly, Danaus plexippus L.
Ann. Ent. Soc. of Amer. 68: 143- 151.
Rice, W. R. 1996. Sexually antagonistic male adaptation triggered by experiental arrest
of female evolution. Nature (Lond.) 381: 232-234.
Riemann, J. G., and B. J. Thorson. 1969. Effect of male accessory material on
oviposition and mating by female house flies. Ann. Ent. Soc. Am. 62: 828- 834.
Robertson, H. M. 1982. Female courtship summation in Drosophila melanogaster.
Anim. Behav. 30: 1105- 1117.
Scott, D. 1986. Inhibition of female Drosophila melanogaster remating by a seminal
fluid protein (esterase 6). Evolution 40: 1084- 1091.
Sivinski, J. 1984. Sperm in competition. Ch. 3 in: Smith, R. L., ed. 1984. Sperm
competition and the evolution of animal mating systems. (Academic Press: New
York).
Soper, R. 1974. The genus Massospora entomopathogenic for cicadas. Part I. Taxonomy
of the genus. Mycotaxon 1:13-40.
Soper, R., A. J. Delyzer, and L. F. R. Smith. 1976. The genus Massospora
entomopathogenic for cicadas. Part II. Biology of Massospora levispora and its
host Okanagana rimosa, with notes on Massospora cicadina on the periodical
cicadas. Ann. Ent. Soc. Amer. 69: 89- 95.
Sullivan, B. K. 1989. Passive and active female choice: A comment. Anim. Behav. 37:
692- 694.
Sullivan, R. T. 1981. Insect swarming and mating. Florida Entomologist 64: 44- 65.
Thornhill, R. 1980. Mate choice in Hylobittacus apicalis (Insecta: Mecoptera) and its
relations to some models of female choice. Evolution 34: 519- 538.
Thornhill, R.A., and J. Alcock. 1983. The evolution of insect mating systems.
(Cambridge: Harvard University Press).
Trivers, R. L. 1972. Parental investment and sexual selection. In: B. Campbell, ed.
Sexual selection and the descent of man, 1871- 1971. (Aldine: Chicago).
25
Watt, W. B., P. A. Carter, and K. Donohue. 1986. Females’ choice of “good genotypes”
as mates is promoted by an insect mating system. Science (Wash., D. C.) 233:
1187- 1190.
Weatherhead, P. J., and R. J. Robertson. 1979. Offspring quality and the polygyny
threshold: “The sexy son hypothesis.” Am. Nat. 113: 201-208.
Wiley, R. H., and J. Poston. 1996. Indirect mate choice, competition for mates, and
coevolution of the sexes. Evolution 50: 1371- 1381.
Williams, K. S., K. G. Smith, and F. M. Stephen. 1993. Emergence of 13-yr periodical
cicadas (Cicadidae: Magicicada): phenology, mortality, and predator satiation.
Ecology 74: 1143- 1152.
Winston, M. L. 1987. The biology of the honey bee. (Harvard University Press:
Cambridge, MA).
Wolfner, M. F. 1996. Tokens of Love: Functions and regulation of Drosophila male
accessory gland products. Insect Biochem. Molec. Biol. 27: 179- 192.
26
Table 1.1. Mate discrimination and its alternatives.
Kind of choice
mate choice
context choice
none
Identity of
mate matters
yes
no
no
Benefits
genetic, material
material
none
27
Selection
sexual
natural
none
28
t, threshold value
Male quality, q
Figure 1.1. Distribution of male traits in relation to female threshold acceptance value, t. Females do not accept males
of quality q < t. As t decreases, number of males excluded from mating (n) also decreases.
n, number of
males excluded
Number of males
29
A
B
C
Figure 1.2. Comparisons of threshold mate acceptance criteria (A, B), and best-of-n choice criteria (C). Increasing width of line
(A, B, or C) indicates increasing probability that a male of quality q will be accepted as a mate. (A) Under threshold choice,
females do not accept males with quality q < t1 , excluding some males from mating. (B) At a lower value of t (t2), number of
males excluded also decreases. Under threshold choice, all males of quality q > t have an equal probability of mating.
(C) Under best-of-n choice, as male quality q increases, so does probability of being accepted as a mate. Best-of-n choice can
lead to runaway evolution, because a mutant male with quality q > l (limit of male variation), is potentially more attractive than
any other male in the population.
number of males
t2, female acceptance threshold
t1, female acceptance threshold
l, limit of male variation
q, male quality
CHAPTER II
EXPERIMENTAL METHODS IN STUDIES OF PERIODICAL CICADAS:
EXAMINATIONS OF POSSIBLE METHODOLOGICAL BIASES IN THE
STUDY OF PERIODICAL CICADAS (MAGICICADA SPP.)
Abstract
Periodical cicadas (Magicicada spp.) are ideal subjects for studies of mating
behavior because adults can be kept and mated in captivity. However, for the results of
captive studies to be applicable to natural populations, the methods used must not bias or
alter mating behavior. For instance, marking allows focal-individual studies, but chemical
effects of marking could cause premature death or reduced likelihood of mating. Further,
cages could bias sexual behaviors, since confined females are limited in their abilities to
escape persistent, harassing males. In this paper we demonstrate that the methods we use
to mark and cage periodical cicadas are harmless. Our marking methods have no
measurable effects, the densities in our cages are not unrealistically higher than densities of
free cicadas on branches, and we uncover no evidence that matings observed in cages result
30
from forcing, or that smaller, denser cages increase males’ abilities to force matings upon
sexually unreceptive females.
Introduction
Our long-term studies of periodical cicada (Magicicada spp.) mating behavior
include experiments and observations in which we confine individually marked periodical
cicadas in closed mesh cages (Chs. 3, 4, 5, 6, 7, 8). Increased mortality or morbidity are
possible undesirable effects of such manipulation. Other effects could be more subtle; in
mate choice studies, cage and marking effects could create false positive results by causing
cicadas to be unusually active. Conversely, if manipulation disturbs cicadas or causes them
to expend significant effort attempting to escape cages or remove marks, mate choice
experiments could yield false negative results. Perhaps the most insidious problem of cage
studies is that they may prevent or at least alter normal sexual rapprochement behaviors by
placing members of both sexes in close, unsolicited proximity. To assess the seriousness
of these problems in our studies of periodical cicadas, we present data addressing changes
in mortality and sexual behavior due to marking or cage confinement artifacts and discuss
ways of minimizing biases.
Methods and Results
Individual marking
The ease of use of different marking methods and their appropriateness for
particular study designs are well reviewed (Southwood 1978, Walker and Winewriter
1981, Kearns and Inouye 1993). Above and beyond logistical concerns, the possible
effects on the study subjects further constrain the choice of marking method. Commonly,
studies making use of marking techniques assume no ill effects. Table 2.1 summarizes
literature reporting effects of individual marking (not mass-marking) methods on study
organisms. We located sources using published reviews and electronic databases
31
(Zoological Record, Wilson Indexes to Journal Articles, and Biological Abstracts) and
included only papers containing data specifically addressing possible mortality or morbidity
due to marking. Of several hundred papers examined, we found only 20 fitting these
criteria. Table 2.1 is neither exhaustive nor prescriptive, but rather supplements earlier
reviews (Southwood 1978, Walker and Winewriter 1981, Kearns and Inouye 1993). Few
general trends are apparent, except that marks applied topically to sclerotized parts seemed
to have few deleterious effects, while dusts, powders, or, in the case of soft-bodied
insects, some solvents, must be used cautiously. The lack of general trends means that
controlled testing of a marking method is the only way to ensure its appropriateness for any
given study.
Individual marking has proven effective in our research on periodical cicadas
(Magicicada septendecim (L.)), allowing us to track the behaviors of individual insects or
classes of insects as they emerge, mature, mate, and die (Chs. 3, 4, 5, 6, 7, 8). One
marking technique we have used is to apply colored enamel paint dots to the dorsal thoracic
surface. Because published reports contain no clear information on the toxicity of the
solvents in these paints (xylene, xylol and n- propoxy propanol) we conducted experiments
to determine whether this marking method has adverse effects on our study animals.
Methods. We compared the survivorship and mating activity of marked and
unmarked adult Magicicada septendecim in a recently logged clearing along Route 639 in
the Horsepen Lake State Wildlife Management Area, Buckingham County, VA, in midMay 1996. We placed two 1 x 2 m black fiberglass mesh bags over living vegetation and
stocked each cage with 10 unmarked and 10 marked mature, unmated female cicadas. The
cicadas were marked or handled before being put in the cages, and were assigned to the
cages in no particular order.1 We marked cicadas by using a flat toothpick to apply one
1
In this design, the 40 female cicadas, not the cages, are replicates; the females were divided into two cages
for convenience and to prevent severe overcrowding. The experiments do not test for a cage effect, but
rather test for effects of two different treatment regimes on the females. The two treatment regimes are
entirely independent of the cages and cage assignments.
32
yellow and two blue dots of Testors’ lead-free gloss enamel (Testors Corp., Rockford IL,
61104) to the dorsal surface of the thorax. The primary solvents in Testors’ enamel paints
are xylol and n- propoxy propanol, which are toxic to humans in high doses. We used no
anesthetic during marking, and we were careful to prevent paint from contacting the wing
articulations or the articulation between the head and thorax. The total surface area covered
by paint was between 6 and 10 mm2. To allow matings, we placed ten sexually mature
adult males collected from an active chorus into each cage. Cicada densities in these cages
were thus high (see below), but intentionally so, to expose any differential survival ability
of marked and unmarked cicadas. From May 23- June 4, we censused the cages 8 times
by counting and removing dead cicadas and recording whether or not each was marked or
showed evidence of a mating sign (or “seminal plug;” White 1973). We scored females
exhibiting seminal plugs as having mated; those lacking the mark as not having done so.
We preserved all cicadas in 70% ethanol, and at the final census, we removed all remaining
cicadas.
Results. We compared the survivorship of the marked and unmarked cicadas
using Cox regression analysis, implemented with the PHREG function in SAS (SAS
Institute Inc., 1996). This technique allowed us to include individuals who survived
through to the end of our observations. The small difference in survivorship curves for
painted and unpainted individuals was not statistically significant (Wald Chi-Square-2.32,
1 d.f., P ≤ 0.13). If anything, marked cicadas seemed to survive longer than unmarked
cicadas, although this pattern is not statistically significant. The two cages do appear to
have different mortality rates (Wald Chi-Square=14.02, 1 d.f., P ≤ 0.0002), attributable to
the death of substantial amounts of vegetation in one cage, although the increased mortality
in this cage was distributed uniformly among marked and unmarked cicadas.
33
A chi-square goodness-of-fit analysis of seminal plug presence/absence data reveals
no significant pattern. Marked and unmarked females were equally likely to exhibit this
evidence of mating. Data are summarized in Table 2.2 and Figure 2.1.
Cage artifacts
In our studies of periodical cicada behavior, we make extensive use of mesh bag
cages. Cages allow recovery of marked individuals, facilitate experiments with
manipulated individuals, and permit isolation from free cicadas so that observations may be
restricted to cicadas whose entire adult histories are known. Our cages consist of fiberglass
hardware cloth folded over living vegetation suitable for feeding or oviposition, giving an
approximate volume of 200 liters. We are careful to construct cages so that the highest
points, in which cicadas tend to congregate, have few corners or narrow spaces that might
trap or cause injury. In experiments where we watch and record cicada behaviors, we
typically stock cages with 10- 20 cicadas; when we are less concerned about the details of
behavior (such as our paint marking experiments), or for storing cicadas, we typically place
20- 40 cicadas per cage.
Each sex appears subject to sex-specific cage mortality. Females, whether mated or
not, oviposit in cage branches, sometimes to a degree that kills or weakens the vegetation.
This problem can be alleviated by moving ovipositing females to fresh branches on a
regular basis. Males, especially those caged near unmated females, tend to damage
themselves by struggling with each other or by attempting to escape. The severity of this
effect can be reduced by housing males at low densities somewhat away from females.
Since both of these problems could severely compromise long-duration experiments, we
have kept cage experiments short and maintained and stocked cages in a manner that
minimizes these effects.
34
Of greater concern are the ways in which cages may alter cicada sexual behavior by
unnaturally confining males and females in close proximity, perhaps promoting male sexual
coercion. For example, females confined with males in cages are denied the opportunity to
participate in normal rapprochement behaviors. If critical components of the mating
sequence occur during rapprochement, such as if females normally accept or reject a
courting male before he comes within one meter of her, females in small cages would not
be able to engage in normal mate rejection behaviors. Observations do not support the
view that male forcing behavior is generally successful in periodical cicadas; of
approximately 100 courtships of caged cicadas observed in 1996, three appeared to involve
male forcing, and only one of these three resulted in copulation. Nevertheless, it is still
possible that cages may facilitate male forcing behaviors, and thus confinement in cages
may compromise mating experiments.
We used three approaches to evaluate the possibility of a cage effect. First, to
determine whether our experimental cage densities are unrealistically high, we compared
the densities in our cages to densities of natural cicada aggregations. Because of the
difficulties of sexing uncaged cicadas, our analysis concerns only cicada densities and not
sex ratios within cages. Second, we caged both normal and physically handicapped males
with receptive females and examined the relative mating success of each class of male.
Comparing the success of normal and handicapped males provides information about
forcing behaviors and whether such behaviors might be exacerbated by cage studies. Last,
we compared female remating frequencies in a large 4m x 4m x 4m screen “flight cage” to
frequencies in small bag cages. Mated females are generally unreceptive (Chapter 4) so
extreme polyandry in this experiment would indicate that confinement reduces the
effectiveness of female resistance. If cages affect rapprochement or escape behaviors, then
the severity of a cage effect should depend on the size of the cage. Demonstrating similar
remating frequencies in large and small cages would suggest the absence of a cage effect.
35
Cage density surveys; methods. On May 14 and 17, 1996, an observer
(JRC) walked along Virginia Route 639, an unpaved road with little traffic, and counted the
numbers of cicadas visible on 39 branches of approximately the same size as those inside
our mesh bag cages. We developed a computer resampling algorithm (Appendix B) to
evaluate the extremeness, relative to our branch scan data, of specified cage densities.
Given a hypothetical cage density, the algorithm randomly picks, with replacement, a value
from the 1996 branch scan data and calculates whether the picked value is more or less
extreme than the specified density by comparing the distances to the scan data mean. The
algorithm repeats these calculations 10,000 times and reports the percentage of iterations in
which the picked value was more extreme than the specified value. This percentage is
analogous to a P-value indicating the likelihood of, by chance alone, picking a value from
the data set more extreme than the specified cage density. The algorithm calculates such Pvalues for many hypothetical cage densities, generating 95% confidence intervals for the
1996 scan data.
Results. Our branch survey counts ranged from 3 to 27 (mean 12.61, ± 5.6).
The resampling algorithm suggests that densities of up to 25 cicadas per caged branch are
not extreme in comparison to the 1996 scan data (Table 2.3) and are comparable to natural
cicada densities on similar branches. Greater densities are atypical of 1996 scan data, and
behavioral data from such cages may be suspect.
Pretarsal clipping experiment; methods. Cicadas have a pair of sharp
terminal claws (or pretarsal ungues; Snodgrass 1993) on each leg. Claw damage reduces
cicadas’ ability to grasp vegetation or other cicadas. Damage to the foreleg claws appears
to increase mortality, because cicadas apparently use their front legs almost as antennae to
investigate the characteristics of a surface before venturing onto it; cicadas with damaged
front claws tend to remain motionless, perpetually clawing at the surface in front of them.
36
Cicadas with claws other than the front pair removed appear to have reduced gripping
abilities but are still mobile.
When a male cicada attempts to force a copulation, he climbs on the back of a
female and must resist her attempts to dislodge him. A female so threatened flaps her
wings vigorously, and is usually successful in preventing mating, unless the male is able to
immobilize her wings or insert his genitalia before being dislodged (JRC pers. obs.).
Given the chance, males will attempt to mount and copulate, but successful forced matings
require substantial male effort. In contrast, in a mating that involves normal courtship and
apparent mate acceptance by the female, the male meets little or no resistance when he
attempts to mount.
If males are manipulated by removing some of their claws, they should be less
successful at some of the behaviors required for forcing matings. If handicapped and
normal males have equal access to females, but normal males mate more frequently, forcing
behaviors are likely an important male tactic, perhaps encouraged by cage studies.
Alternatively, no difference in mating frequencies would suggest that male forcing
behaviors are not generally successful.
To evaluate whether bag cages allow males to trap females and force matings upon
them, we placed twelve mature unmated female cicadas into three hardware cloth cages,
four females per cage. Cages enclosed living vegetation, providing access to live branches
for feeding and oviposition. Into each cage we also placed eight mated males obtained by
separating mating pairs. Males can mate more than once; we used mated males to establish
that all males were acceptable as mates prior to the manipulations. Four of these eight
males had their two midleg pretarsal claws removed with a fingernail clipper, while the
other four were handled similarly but left intact. We censused the cages regularly and
recorded the identities of copulating males. Copulations are distinguishable from other
37
postures, because even though body orientations vary, the male genitalia are clearly
engaged.
Claw removal had observable effects on the cicadas’ mobility and agility, but no
evident immediate effects on mortality. If forcing behaviors are prevalent in bag cages,
then unmanipulated and fully agile cicadas should obtain a greater proportion of matings
than should males whose claws have been removed and who are thus at a disadvantage.
Results. The experiment ran for approximately 71 hours, or until all but two
females copulated. The first matings occurred 40 hours after the experiment began, and the
last matings at 68 hours. No rematings were allowed. Data were pooled across replicates,
and pooled results were analyzed using a two- tailed Fisher’s Exact Test. Manipulated
(clipped) and unmanipulated cicadas were statistically equally likely to obtain matings
(Table 2.4). For comparison, a data set of this size showing a significant treatment effect
would need to have two or fewer males mate in one treatment and eight or more in the
other. With the caveat that this experiment involves small sample sizes, it does not support
the view that male forcing behaviors are generally prevalent or successful, because males
whose ability to force matings was impaired were at no disadvantage compared to normal
males. Thus, we have no evidence that male forcing behaviors are a significant
complication in cage experiments.
Cage size effects; methods. Unconfined, singly mated female Magicicada do
not exhibit normal female behaviors indicating sexual receptivity (Chapter 5). If such
females are generally able to avoid encountering males and remating, then a cage effect may
have its most observable impact on female remating frequencies. A cage effect would also
likely involve a relationship between the cage volume, or cage surface-to -volume ratio, and
the severity of the effect, because in the smallest cages, females should be least able to
escape unsolicited courtships and males should be most able to find females.
38
One way to test for the operation of a cage effect, then, is to compare remating
behavior in small cages with behavior in large cages. In 1995, we stocked one large cage
(the “flight cage”) and many smaller cages (“bag cages”) with cicadas. To determine
whether remating frequencies were higher in the bag cages than expected from flight cage
data we used a Monte-Carlo computer resampling algorithm to compare remating
frequencies in the flight cage and bag cage data.
During the 1995 Brood I periodical cicada emergence at Alum Springs, Rockbridge
County, Virginia, we stocked a 4m x 4m x 4m nylon mesh cage (our “flight cage”) with 79
adult female Magicicada septendecim, 86 adult male M. septendecim, 14 adult female M.
cassini, and 10 adult male M. cassini. Magicicada nymphs tend to emerge from the ground
and undergo their final molt in the evening (Maier 1982); the morning after ecdysis, newly
emerged cicadas (“teneral” cicadas) are readily identifiable by their pale color, soft bodies,
and by their position low in the vegetation. We captured teneral cicadas, individually
marked them, placed them in the cage, and recorded the dates, times, and individuals’
identities for all matings. Based on day of tenerality (capture), we assigned each cicada to a
“day-cohort” useful in estimating the onset of sexual maturity. We recorded all observed
mating pairs in the cage. Out of 79 female M. septendecim, only one mated with a male M.
cassini ; in addition, one female M. cassini mated with a M. septendecim male. Because
cross-species interactions appear minimal, M. cassini will not be considered further in
discussions of this experiment. Of the 78 female M. septendecim that mated, most did so
by the eighth day of their adult lives, and 13 mated more than once before ovipositing.
Although the flight cage experiments ran for 20 days, because cicadas in the cage
belonged to several day-cohorts, their behavior is best recorded with respect to day from
tenerality, not day of first cage stocking. Therefore, only information on the behavior of
78 female cicadas for a period from tenerality to 12 days of adulthood was available. Not
all of the cicadas belonging to later cohorts have information available about the later days
39
(days 8-12) of their adult lives, because the experiment was terminated less than 12 days
after the latest cohorts were collected. It is not known whether cicadas in later cohorts
would have remated, because they may only have had enough time to mate once. Thus,
calculations of remating frequency in the flight cage are conservative, because any errors in
the collection or interpretation of the flight cage data can lead only to underestimates of
remating frequencies. This conservative figure for female remating provides the maximum
potential contrast to cicada behavior in the smaller cages.
Of the five bag cage experiments conducted in 1995, only two allowed remating
and are appropriate for this analysis. The design and results of each are summarized in
Appendix C. Although the flight cage data covers a period of 12 days, and these
experiments ran for 11 days, these time periods are comparable, since the number of days
is less important than day quality (in terms of weather, etc.). In any case, all but one
mating in the flight cage occurred before the mating cicada’s tenth day from tenerality, well
before the end of the flight and bag cage experiments.
A cage effect that scales with cage size should manifest itself as higher remating
frequencies in the small bag cages than in the large flight cage. We compared mating
frequencies using a C++ resampling algorithm (Appendix D). The algorithm resamples the
flight cage data on singly-and doubly- mated females to create sets of sizes comparable to
the bag cages (n= 8 or n= 4 as appropriate), and it tallies the number of females in the
resampled data that mated twice. The program iterates this resampling 10,000 times to
generate a frequency distribution, and thus percentiles, of the outcomes of the resampling.
Results. The second replicate in the first experiment I may show a cage effect,
since half of its 8 females remated, which is outside the 95th percentile of the resampling
algorithm, and more that twice the remating frequency in any other cage (Table 2.5).
Interpretation of these results requires caution, because two of the rematings in this cage are
unusual, and unlike any others in these experiments. One of the four rematings in this cage
40
occurred after a mating involving conflict between two males. In this particular instance,
two males struggled with each other, both attempting to insert their genitalia into the
female. It was not possible to determine whether either male was successful in copulating.
When the cage was next observed, one hour later, a period too short to be considered a
normal mating (Chapter 4), the female was alone and not mating. Approximately 20 hours
later, she engaged in a normal mating.
Another of the rematings in this replicate was the only post-oviposition remating in
this study. This study was not designed to evaluate post-oviposition remating, which may
occur for reasons distinct from pre-oviposition rematings (Chapter 4). If pre- and postoviposition remating are treated as different phenomena, and if the post-oviposition
remating is therefore eliminated from consideration, the results in this replicate are not
significantly different from flight cage results. We scored both of these unusual situations
as matings, but neither would have met the stricter criteria typical of our other experiments.
If either one of the unusual matings in replicate 2 is discounted, then this replicate does not
have a remating frequency higher than that in the flight cage.
Conclusions
We have uncovered no evidence that careful marking with enamel paints affects
cicadas’ mortality or likelihood of mating, nor is there any evidence that cicada densities in
our experimental cages are unrealistically high. The pretarsal clipping experiment suggests
that males prevented from forcing are no less successful than normal males, and that male
forcing strategies are not a significant factor in bag cage experiments. The cage size
resampling algorithm demonstrates the absence of a cage effect whose severity scales with
cage size. We have thus uncovered no evidence that our methods significantly alter cicada
behavior in ways that would generate false or misleading results.
41
From observing cicada behavior in cages, however, we can add some notes of
caution. In 1997, we placed female M. septendecim and male M. cassini in cages away
from a male M. septendecim chorus. Under normal circumstances, we observed few
sexual interactions among these cicadas. When we played male M. septendecim song,
however, female M. septendecim responded with a receptivity signal (Chapter 5). This
signal, which is similar in all Magicicada species, apparently incited the male M. cassini
males in the cages to increase their sexual activities, and in some cases, a heterospecific
male successfully engaged in a cross-species mating by mounting and copulating with a
female who had responded to conspecific playbacks.
In another experiment, we constructed several dozen model cicadas from blocks of
wood, painted them black, and attached them to the branches of a tree enclosed in a large
flight cage. We stocked the cage with male M. septendecim and M. cassini. Normally, the
males paid no attention to the models, but if we produced a simulated female wing-flick
signal in time with a male call, males directed increased attention to the blocks, in some
cases mounting and copulating. Male M. cassini in particular seemed to increase their
sexual activity after we produced a simulated female wing-flick signal, and one male’s
excitement seemed to have a synergistic effect on the others.
The experimental results in this paper demonstrate the effectiveness of female
resistance, and that cages do not cause unreceptive females to mate. However, close
confinement of several males and females in cages might compromise mate choice
experiments, because if one female in a cage produces a wing-flick signal, indicating sexual
receptivity, all the males in the cage may increase their sexual activity. In small, densely
stocked cages, males can reach a mating frenzy, in which they interfere with and attempt to
replace other courting males. Our observations indicate that these problems are more
severe in M. cassini than in M. septendecim, so it appears advisable to avoid using the
former species in mating experiments in which several males are confined in one cage.
42
Cage experiments in general require close observation and careful design to ensure that
male-male competition does not unduly influence mating. Criteria for scoring matings must
be designed so that unusual situations, such as post-oviposition mating, male-male
competition, and interloping attempts, do not contribute to false positive results.
43
Literature Cited
Brussard, P. F., 1971. Field techniques for investigations of population structure in a
“ubiquitous” butterfly. Journal of the Lepidopterists’ Society. 25: 22- 29.
Caprio, M. A., D. Miller, and E. Grafius. 1990. Marking adult Colorado potato beetles,
Leptinotarsa decemlineata (Coleoptera: Chrysomelidae), using paper labels. Great
Lakes Entomologist 23: 13-18.
Davey, J. T. 1956. A method of marking isolated adult locusts in large numbers as an aid
to the study of their seasonal migrations. Bull. Ent. Res . 46: 797-802.
Dobson, R. M., J. W. Stephenson, J. R. and Lofty. (1958). A quantitiative study of a
population of wheat bulb fly, Leptohylemia coarctata (Fall.), in the field. Bull. Ent.
Res. 49: 95-111.
Dobson, R. M., and M. G. Morris. 1961. Observations on emergence and life-span of
wheat bulb fly, Leptohylemia coarctata (Fall.), under field cage conditions. Bull.
Ent. Res. 51: 803- 821.
Dreyer, H., and J. Baumgartner. 1997. Adult movement and dynamics of Clavigralla
tomentosicollis (Heteroptera: Coreidae) populations in cowpea fields of Benin,
West Africa. J. Econ. Entomol. 90: 421-426.
Fales, J. H., O. F. Bodenstein, G. D. Mills, Jr., and L. H. Wessel. 1964. Preliminary
studies on face fly dispersion. Ann. Ent. Soc. Amer. 57: 135-137.
Gallepp, G. W. and A. D. Hasler. 1975. Behavior of larval caddisflies (Brachycentrus
spp.) as influenced by marking. Am. Midl. Nat. 93: 247-254.
Gangwere, S. K., W. Chavin, and F. C. Evans. 1964. Methods of marking insects, with
especial reference to Orthoptera (Sens. Lat.). Ann. Ent. Soc. Amer. 57: 662-669.
Garcia, M. A., and L. M. Paleari. 1990. Marking cassinidae (Coleoptera: Chrysomelidae)
larvae in the field for population dynamics studies. Ent. News 101: 216-218.
Greenslade, P. J. M. 1964. The distribution, dispersal, and size of a population of Nebria
brevocillis (F.) with comparative studies on three other Carabidae. J. Anim. Ecol.
33: 311- 333.
Hager, B. J., and F. E. Kurczewski. 1986. Nesting behavior of Ammophila harti
(Fernald)(Hymenoptera: Sphecidae). American Midland Naturalist 116: 7- 24.
Hart, D. D. and V. H. Resh. 1980. Movement patterns and foraging ecology of a stream
caddisfly larva. Can. J. Zool. 58: 1174-1185.
Hunter, P. E., 1960. Plastic paint as a marker for mites. Ann. Ent. Soc. Amer. 53: 698.
44
Kearns, C. A. and D. W. Inouye, 1993. Techniques for pollination biologists. (University
Press of Colorado: Niwot, CO).
Maier, C. T. 1982. Obervations on the seventeen-year periodical cicada, Magicicada
septendecim (Hemiptera: Homptera: Cicadidae). Ann. Ent. Soc. Amer. 75: 14- 23.
Mascanzoni, D., and H. Wallin. 1986. The harmonic radar: a new method of tracing
insects in the field. Ecol. Ent. 11: 387-390.
Mead-Briggs, A. R. 1964. Some experiments concerning the interchange of rabbit fleas,
Spilopsyllus cuniculi (Dale), between living rabbit hosts. J. Anim. Ecol. 33: 13-26.
Paulson, G. S., and R. D. Akre. 1991. Role of predaceous ants in pear Psylla
(Homoptera: Psyllidae) management: estimating colony size and foraging range of
Formica neoclara (Hymenoptera: Formicidae) through a mark-recapture technique.
J. Econ. Entomol. 84: 1436-1440.
Rieske, L. K. and K. F. Raffa, 1990. Dispersal patterns and mark-and-recapture estimates
of two pine root weevil species, Hylobius pales and Pachylobius picivorus
(Coleoptera: Curculionidae), in Christmas tree plantations Environmental
Entomology 19: 1829- 1836.
SAS Institute Inc., 1996. SAS/STAT® Software: Changes and Enhancements through
Release 6.11. Cary, NC: SAS Institute Inc.
Sempala, S. D. K. 1981. The ecology of Aedes (Stegomyia) africanus (Theobald) in a
tropical forest in Uganda: mark-release-recapture studies on a female adult
population. Insect Sci. Application 1: 211-224.
Snodgrass, R. E. 1993. Principles of Insect Morphology (Cornell University Press: Ithaca,
NY).
Southwood, T. R. E., 1978. Ecological Methods, with particular reference to the study of
insect populations. (Chapman and Hall: London).
Stuart, R. J. 1986. Use of polyester fibers to mark small leptothoracine ants
(Hymenoptera: Formicidae). J. Kans. Ent. Soc. 59: 566-568.
Walker, T. J., and S. A. Winewriter. 1981. Marking techniques for recognizing individual
insects. Florida Entomologist 64: 18- 29.
White, J. 1973. Viable hybrid young from crossmated periodical cicadas. Ecology 54: 573580.
45
46
None
None
“Sharpie” marker (xylol solvent)
Acrylic paint
Paper labels affixed with cynoacrylate glue
Acetone wash, followed by
paper labels affixed with cynoacrylate glue
Plastic tags affixed with wire or elastic band
Brussard 1971
Rieske and Raffa 1990
Caprio et al. 1990
Caprio et al. 1990
Hart and Resh 1980
Dicosmoecus gilvipes (Trichoptera)
Leptinotarsa decemlineata (Coleoptera)
Leptinotarsa decemlineata (Coleoptera)
Hylobius pales (Coleoptera)
Pachylobius picivorus (Coleoptera)
Erebia epipsodea (Lepidoptera)
Clavigralla tomentosicollis (Hemiptera)
Fingernail polish (acetone based)
Dreyer and
Baumgartner 1997
None
None
None
None
Toxic
None
None
Behavioral
changes
Toxic
None
None
Musca autumnalis (Diptera)
Charidotis punctatostriata (Coleoptera)
Lepthohylemia coarctata (Diptera)
Schistocerca gregaria (Orthoptera)
Formica neoclara (Hymenoptera)
Lepthohylemia coarctata (Diptera)
None§
None†
Fingernail polish (acetone based)
Fingernail polish (acetone based)
Enamel paint (petroleum-solvent based)
Oil-based paint
Paulson and Akre 1991
Dobson et al. 1958
Ammophila harti (Hymenoptera)
Nebria brevicollis (Coleoptera)
None
None
Fales et al. 1964
Garcia and Paleari 1990
Enamel paint (xylol, propanol based)
Enamel paint (petroleum-solvent based)
Hager and Kurczewski
1986
Greenslade 1964
Laelaspis georgiae (mite)
Leptinotarsa decemlineata (Coleoptera)
None
Effect*
Brachycentrus spp. (Trichoptera)
Enamel paint (xylol, propanol based)
Enamel paint (xylol, propanol based)
Hunter 1960
Caprio et al. 1990
Magicicada septendecim (Homoptera)
Melanoplus sanguinipes (Orthoptera)
Taxon
Lacquer-based paint (nitrocellulose)
Oil-based paint (Rubine Toner or
Monolite Yellow in wood rosin and kerosene)
Gallepp and Hasler 1975 Typewriter correction fluid
Enamel paint (xylol, propanol based)
This study
Dobson et al. 1961
Davey, 1956
Marker
Reference
Table 2.1. Reported effects of methods for marking individual insects, grouped by general method and organized chronologically within each grouping.
47
Colored thread or paper attached with glue
Polyester fiber tied around body
Melted crayon, colored ink, Aniline dyes, or
Titanium oxide spray
Metallic dust
Fluorescent powder, pigment in oil spray,
Oil and Cellulose paints, dry pigment
Wing amputation or clipping
Gangwere et al. 1964
Stuart 1986
Gangwere et al. 1964
Gangwere et al. 1964
Gangwere et al. 1964
Gangwere et al. 1964
Spilopsyllus cuniculi (Siphonaptera)
Pterostichus melanarius, P. niger, Harpalus
rufipes, Carabus granulatus (Coleoptera)
Periplaneta americana (Blattaria)
Acheta domesticus (Orthoptera)
Acheta domesticus (Orthoptera)
Acheta domesticus (Orthoptera)
Acheta domesticus (Orthoptera)
Leptothorax spp., Harpagoxenus spp.
(Hymenoptera)
Aedes africanus (Diptera)
* “None:” no effects reported; “Toxic:” increased mortality reported; “Injury:” unspecified damage or mortality noted.
§ Pers. Comm.
† Dispersal attributable to handling effects noted.
‡ Method proved to be impermanent and of limited use.
Mead-Briggs 1964
Tarsal clipping
Mascanzoni and Wallin 1986
Radar tags
Fluorescent powder
combined with Duco paint
Sempala 1981
Table 2.1 (Continued)
None
None
None for
moderate treatment
None‡
Toxic in
large doses
None‡
None
None‡
Toxic
Table 2.2. Numbers of marked and unmarked females showing a “seminal plug,” or
evidence of mating, and χ 2 comparison of seminal plug frequency in marked and
unmarked cicadas. Equal numbers of marked and unmarked females were assigned to two
cages, with an excess of males. Total females: Total number of females of each category
present. Seminal plug present: total number of females at end of experiment with seminal
plug. Marking does not affect likelihood of mating.
Total
females
“seminal plug”
present
χ 2 (2- tail)
P
Cage A
Marked
Unmarked
Total
10
10
20
6
7
13
0.038
P>>0.1 (n.s.)
Cage B
Marked
Unmarked
Total
10
10
20
5
6
11
0.045
P >>0.1 (n.s.)
Combined cages
Marked
20
Unmarked
20
Total
40
11
13
24
0.083
P >>0.1 (n.s.)
48
Table 2.3. Cage density resampling results, 10,000 iterations. Algorithm
determines whether a specified cage density is greater than densities of
cicadas on actual branches by resampling, with replacement, branch density
data and generating percentiles. Cage densities above 25 are significantly
greater than naturally-occurring densities on branches of similar size.
Hypothetical cage density
12
14
16
25
26
P
0.85
0.82
0.51
0.053
0.028*
* indicates significance
49
Table 2.4. Results of pretarsal clipping experiment. Mating likelihoods of normal males
and males with center tarsal claws removed were compared by confining 4 unmated
females in each of 3 cages, and adding 4 normal and 4 clipped males to each cage. Counts
are of normal and tarsal clipped males mating in each cage. 10 total matings occurred. In
the combined replicates, the number of normal and tarsal clipped males mating do not differ
(Fisher’s Exact 2- Tail P ≤ 0.656).
Cage
1
2
3
Combined
Males mating
Normal
Clipped
3
1
2
1
2
1
6
4
50
51
4
3
2
2
# Females
8
8
8
8
4
4
4
Replicate
Experiment 1:
(A)†
(A)§
B
C
Experiment 2:
D
E
F
10,000
10,000
10,000
10,000
10,000
10,000
10,000
# Resampling
iterations
5150
5150
5150
290
1299
3960
3960
# Iterations with
rematings =
observed rematings
† If both questionable matings in replicate A are included as first matings (see text).
§ If only one questionable mating in replicate A is included as a first mating (see text).
1
1
1
Observed
rematings
≤ 0.515
≤ 0.515
≤ 0.515
≤ 0.029
≤ 0.130
≤ 0.396
≤ 0.396
P
Table 2.5. Summary of resampling results comparing remating frequencies in small bag cages to remating frequency in large flight
cage. Algorithm starts with data on singly- and doubly- mated females from large flight cage. The algorithm draws samples of size n
from this population, where n is the size of the population in small bag cages (8 for experiment 1; 4 for experiment 2). The algorithm
then tallies the number of resampling iterations in which the number of multiply mated females in the sample is greater than or equal
to the number of observed rematings in the small bag cages.
52
Number Living
5/23/96
Date
6/3/96
6/2/96
6/1/96
5/31/96
5/30/96
5/29/96
5/28/96
5/27/96
5/26/96
5/25/96
5/24/96
5/22/96
5/21/96
5/20/96
5/19/96
5/18/96
5/17/96
5/16/96
5/15/96
5/14/96
5/13/96
5/12/96
5/11/96
Figure 2.1. Survivorship of marked and unmarked cicadas in two cages. Triangular and square markers indicate different
cages, filled markers indicate marked treatment, and open markers indicate unmarked treatment. Cicadas in the two cages
differed in survivorship, but within a cage, marking did not affect survivorship.
0
2
4
6
8
10
12
6/4/96
CHAPTER III
MATE GUARDING AND THE NATURE AND POSSIBLE FUNCTIONS
OF “SEMINAL PLUGS” IN PERIODICAL CICADAS (MAGICICADA
SPP.)
Abstract
After mating, a female periodical cicada (Magicicada spp.) usually exhibits a mass,
or “seminal plug,” blocking her genital opening. This mass contains DNA and appears to
be composed of dried seminal fluid rather than being specially secreted or constructed.
Although the “seminal plug” does not prevent additional matings, it may be part of a male
strategy to make remating so costly, in terms of time, that mated females will be sexually
unreceptive. Evidence that time is valuable to female Magicicada includes the limited nature
of oviposition space, the monopolization of oviposition space by the first females to find it,
and the apparent space-conserving nature of Magicicada eggnests. Thus, “seminal plugs”
may be part of a male paternity-assurance strategy even though they do not prevent
matings.
53
Introduction
As long as a given male’s sperm remain in a female’s reproductive tract, it is not in
his interests to facilitate future mating by that female. However, because females may seek
opportunities to remate in order to acquire resources or manipulate offspring paternity
(Chapter 4), the sexes may be in conflict over remating. In species where males’ paternity
is threatened by opportunities for polyandry, selection will favor males that evolve methods
for blocking or preventing female remating, such as by inserting a “mating plug,” or
barrier, into the female genital opening. Even a current tendency for females not to remate
could be the direct result of past male efforts to devalue females’ remating incentives. For
example, by inserting a mating plug, a male imposes costs on his mate and reduces her
potential benefits from remating by forcing future mating efforts to include time spent
removing or circumventing the plug. If females are time-limited, even if mating plugs are
not absolute barriers to remating, females with such plugs may benefit from avoiding costly
second matings by becoming sexually unreceptive after their first mating.
Mated female periodical cicadas (Magicicada spp.) usually exhibit a whitish or
yellowish ovoid mass lodged in their genital openings (White 1973). Although the name
applied to this structure, “seminal plug,” implies that it is an adaptation for mate-guarding,
there are no detailed studies of its function. Magicicada seminal plugs could be the
incidental results of ejaculate desiccation upon exposure to air. However, it is also possible
that males use seminal plugs to protect their paternity, because under certain conditions
female Magicicada remate (Chapter 4).
In other insect species, seminal plugs may function in biasing paternity by causing
ejaculate to be retained within the spermatheca, by excluding additional sperm during the
period that previously acquired sperm may be used (or stored for future use), or by
permanently excluding additional sperm (Parker 1970, Boorman and Parker 1976, Leopold
54
1976, Ehrlich and Ehrlich 1978, Thornhill and Alcock 1983: 339, Alcock 1994, Polak et
al. 1998). Extended copulation may serve similar functions (Alcock 1994, Parker 1970).
Functional seminal plugs are often constructed of materials other than ejaculate. For
example, honeybee (Winston 1987) and harvester ant (Hölldobler 1976) males leave their
genitalia as plugs. In some Diptera, “sperm plugs” appear to be derived from
spermatophore materials (see Neilsen 1959, Lum 1961, Kotrba 1996, Yasui 1997), while
in many Lepidoptera, males apply sealing substances to the female genital opening after
inserting a spermatophore (Labine 1964, Matsumoto 1987, Ehrlich and Ehrlich 1978).
Determining whether “seminal plugs” are male paternity-assurance adaptations or
whether they are instead functionless consequences of matings is not a simple task. Two
general predictions distinguish between functional and nonfunctional explanations for
seminal plugs. (1) If a female insect can be caused to mate and acquire a seminal plug, and
if she is prevented from acquiring a normal sperm supply but does not remate, the plug
itself may be an effective deterrent to matings; alternatively, if such females show no
decrease in likelihood of remating, remating propensity is likely controlled by factors such
as spermathecal fullness. (2) If males use special materials or processes to construct or
insert seminal plugs, the costs of doing so must be offset by some functional benefit, such
as the ability of plugs to block matings; if, on the other hand, seminal plugs are incidental
results of ejaculate drying, their functional significance and adaptive nature would be less
certain.
I used several approaches to assess whether male Magicicada behave in ways that
protect their paternity and whether seminal plugs are part of a paternity-assurance strategy.
First, I evaluated whether mating durations are far in excess of what would be required for
complete insemination; an excess would imply that males use mating duration as a form of
mate-guarding. I also interrupted matings to determine whether seminal plugs are
deposited at the conclusion of the normal mating sequence and whether they are produced
55
when males are removed suddenly and without warning. I determined whether seminal
plugs deter or prevent subsequent matings, and I analyzed the substances contained in them
to evaluate whether they are specially constructed or secreted. I confined my research to
the 17- year species Magicicada septendecim, but since M. cassini and M. septendecula, as
well as the 13- year species M. tredecim, M. neotredecim, M. tredecassini, and M.
tredecula have similar mating systems and exhibit seminal plugs, the results are likely
general to the genus Magicicada.
Methods and Results
Mating duration
In 1995 (study sites listed in Table 3.1), I captured and individually marked
(Cooley et al. 1998, Chapter 2) teneral females and allowed them to mature in ca. 200 liter
single-sex storage cages. Cages were constructed by folding a piece of hardware cloth
over living hardwood vegetation. When the females were mature, I placed 5- 10 in newly
constructed cages with fresh vegetation, and to each cage I added males captured from
those active in nearby choruses. To establish starting and stopping times for each observed
copulation, I monitored cages continuously or scanned them every 10-15 minutes, or ad lib
scanned when I heard male courtship sounds. Mating durations are thus, in some cases,
estimates ± 15 minutes. In 1996- 1998, I conducted similar experiments on 8 individuals
belonging to two species of the genus Okanagana to provide comparative data (4 each O.
canadensis and O. rimosa). I observed these cicadas continuously during courtship and
mating and examined females for the presence of “seminal plugs” after mating.
Results. For the 61 Magicicada mating pairs I observed, mating duration
averaged 838.8 (± 1074) minutes. Since poor weather may extend matings over days
(unpub. data), if 17 matings lasting more than 10 hours are excluded, the remaining 43
56
matings had an average duration of 243.4 (± 121.48) minutes (Table 3.2). Matings in
Okanagana were much shorter, with average durations under 30 minutes (Table 3.2).
Seminal plug deposition
In 1996, I collected 65 unmated, teneral female Magicicada septendecim and
allowed them to mature as described above. When they were mature, I allowed them to
mate with males caught from those present singing and flying in nearby choruses. I
interrupted 38 of the resulting matings after 1 or 2 hours by suddenly, grasping the base of
the male’s genitalia, gently working male and female genitalia apart, and noting the
presence or absence of a “seminal plug.”
Results. Although interrupted matings could result in seminal plugs, indicating
that the plug may be formed if the male transfers at least some ejaculate, interruption
significantly decreases the likelihood that a seminal plug will be present (Table 3.3). All
seminal plugs were clearly distinguishable from the absence of a plug, but plugs resulting
from an interrupted copulation were more likely to have a soft, glossy appearance than
plugs from uninterrupted matings. Because some females involved in terminated
copulations exhibited sperm plugs, the insertion or creation of this plug requires either (1)
no special actions or processes, or (2) that the male engage in processes that he can initiate
and complete with little or advance preparation.
When I separated copulating pairs, in several cases I noted a small, whitish drop of
ejaculate clinging to the males’ genitalia. In one case, the male continued ejaculating after
separation, and the ejaculate quickly dried into a solid, whitish mass reminiscent of seminal
plugs. This dried mass had the same color and consistency a typical seminal plug, except
that it was of a more uniform consistency; the distal end of seminal plugs (the end visible in
the genital opening) tended to be drier and more solid than the soft, moist proximal end.
57
These observations of drying ejaculate demonstrate that it can harden into a solid mass
without additions or manipulations by the male.
Female remating
I allowed the 36 of the mating-interrupted females the opportunity to remate with
males recently captured from the nearby chorus. Of the 36 females, 11 exhibited seminal
plugs. When given the opportunity to remate, 20 of the 36 females in our mating
interruption experiment remated; 4 of these 20 females exhibited seminal plugs. I
compared remating frequencies of females with and those without seminal plugs using both
a Chi-squared goodness of fit test and a computer resampling algorithm (Appendix E). The
resampling algorithm randomly draws 20 “remating” samples from a population composed
of 11 females with mating plugs and 25 without (total 36). The algorithm tallies the
number of females in each sample with seminal plugs and repeats the process 100,000
times to generate percentiles. These percentiles can be used to determine whether the
number of remating females with seminal plugs is smaller than expected by chance, as
would be predicted if the seminal plug prevented copulation.
Females with seminal plugs were not underrepresented among remating females
(Table 3.4). The seminal plug is therefore not a barrier to remating.
Seminal plug composition
In 1997, I captured approximately 25 mated female M. septendecim and, using
forceps, removed seminal plugs from their genital openings. I preserved the plugs in 70%
ethanol. To isolate high molecular weight DNA, I used Qiagen tissue preparation columns
(Qiagen corp., Cat. 29304) to extract DNA from preserved plugs or from 70% ethanol
preserved adults collected in 1996. I used a modified version of the tissue protocol
included with the kit:
58
1. I removed and macerated a leg from each adult cicada, or I combined several
seminal plugs into one microfuge tube.
2. I added 180 ul Qiagen buffer ATL and 20 ul Proteinase K stock solution without
vortexing. I incubated the solution for a minimum of 24 hours at room
temperature.
3. Without vortexing, I added 200 ul Qiagen buffer AL and incubated at 70°C for at
least 10 min.
4. Without vortexing, I added 210 ul 100% ethanol.
5. I washed and eluted the DNA twice with warm distilled water as per instructions.
I amplified extracted DNA using a Randomly Amplified Polymorphic DNA
oligonucleotide primer (RAPD; Haymer 1994, Hadrys et al. 1992) obtained from the
University of British Columbia, with adult cicadas as positive controls and distilled water
as our negative control in all amplifications. After screening approximately 200 other 10base primers, I found that the primer sequence (CCT GCG CTT A) reliably produces a
characteristic double or single band in M. septendecim with a minimum of amplification
artifacts. I visualized amplified DNA on 1% agarose gels stained with ethidium bromide,
photographed the gels, and noted the presence or absence of bands.
Results. The DNA extraction and amplification demonstrates that seminal plugs
contain DNA (Fig. 3.1), a finding consistent with the hypothesis that they are formed of
dried ejaculatory material instead of special-purpose substances. Qualities of the ejaculate
alone can account for the properties of seminal plugs. As I extracted the seminal plugs, I
noted that the proximal ends of most plugs were bifurcated, suggesting that they extended
into paired spermathecae. The texture of the plugs was consistent with the hypothesis that
they were formed by progressive drying of ejaculatory fluid, starting from the more
exposed, distal end.
59
Discussion
The properties of seminal plugs can be accounted for by the properties of ejaculate
alone. Males do not appear to employ special substances or structures other than ejaculate
to construct seminal plugs; thus, they are not unambiguously male paternity assurance
adaptations, although the hardening properties of ejaculate may be the product of selection
for ejaculate to form barriers to remating. The remarkably long mating durations and
“seminal plugs” of Magicicada may be parts of a male paternity-protection strategy. Mating
interruption experiments (Chapter 4) have demonstrated that females allowed to mate for
more than 2 hours tend to be sexually unreceptive and unlikely to remate; the longer
average mating durations (ca. 5 hours) typical of Magicicada may be a form of male mate
guarding (see Thornhill and Alcock 1983:340, Alcock 1994). Furthermore, the hooks and
complex structure of male Magicicada genitalia (Chapter 8) may allow males to extend
mating duration contrary to female interests, allowing the ejaculate to harden and making
future matings costly or difficult. In contrast to Magicicada, Okanagana have short mating
durations, simple genitalia, lack a seminal plug, and, presumably, have a greater
confluence of interest between the sexes over mating duration.
Paradoxically, although Magicicada have long lifespans, time may be the adults’
greatest enemy. For example, since females appear to avoid constructing new eggnests
directly over pre-existing ones, oviposition space in dense emergences may be limited and
available only to the first finder. The ideal eggnest density probably represents a complex
balance between overly dense eggnests, leading to “flagging,” or death of twigs (and the
eggs therein), and overly sparse eggnests, which might be vulnerable to predators or
parasitoids (Marlatt 1923). Of these competing sources of egg mortality, flagging is certain
to occur on densely oviposited branches, while the risks from ovipositing in sparsely used
branches are uncertain; thus females may be better off to err on the side of ovipositing in an
area of sparse eggnests than to risk having their eggnests “flag.” The prevalence of
60
flagging may thus be weak evidence suggesting the scarcity of suitable oviposition space
and the lack of oviposition options for many females.
The structure of Magicicada eggnests also supports the claim that oviposition space
is limited. Magicicada eggnests are bifurcated, unlike those of other cicadas ovipositing in
live wood, such as Okanagana. Bifurcated eggnests take up approximately as much linear
space on a branch as do single eggnests, so Magicicada eggnests contain approximately
twice as many eggs as Okanagana nests. Magicicada’s bifurcated eggnests may be an
adaptation to ensure complete oviposition and monopolization of unused twigs when
limited oviposition space is available on a “first come, first serve” basis.
Female- female competition for finding oviposition space may place a premium on
efficient use of time and ability to locate and take maximum advantage of unused twigs.
Males may use the currency of time to coerce females to acquiesce to male interests. Even
without being an absolute barrier to copulation, seminal plugs may make future matings
costly in terms of time, because future mates would have to remove or circumvent the
seminal plug. By engaging in lengthy copulations, obstructing the female reproductive
tract, and thereby ensuring that any future copulations would also be lengthy, males may
reduce females’ potential benefits from remating.
One hypothesis for how Magicicada male and female interests in mating duration
have evolved is as follows: Because males that monopolized the paternity of their mates’
offspring sired more offspring and were more successful than males whose sperm was
diluted or discarded because of remating, some males increased their control of paternity by
increasing mating duration. Over evolutionary time, as males first began to extend
matings, the sexes may have been in conflict over the time costs of mating. As males
escalated these costs beyond a certain point, and as males began to employ more and more
effective forcible restraining devices, such as complex genitalia and ejaculate that forms
seminal plugs (or any other strategy that functions to increase intermating interval), male
61
and female interests would have become more congruent, in that females could not afford
to permit additional matings. Those females acceding to male interests in monopolizing
paternity would have had an advantage over those engaging in multiple lengthy matings, so
females would have evolved to be unreceptive to multiple matings. Thus, by the process
described above, a seminal plug that did not prevent mating may have been part of a male
strategy to negate any advantages females could realize from remating. This hypothesis
would be disproven by findings that (1) seminal plugs provide no impediment to future
matings and males have no adaptations for evading or removing them; or (2) seminal plugs
are found in similar cicada species (such as Okanagana spp.) in which there is no evidence
of male-female conflict over mating duration.
62
Literature Cited
Alcock, J. 1994. Postinsemination associations between males and females in insects: The
mate-guarding hypothesis. Ann. Rev. Entomol. 39: 1-21.
Boorman, E., and G. A. Parker. 1976. Sperm (ejaculate) competition in Drosophila
melanogaster, and the reproductive value of females to males in relation to female
age and mating status. Ecological Entomology 1: 145- 155.
Cooley, J. R., Hammond, G. S., and D. C. Marshall. 1998. The effects of enamel paint
marks on the behavior and survival of the periodical cicada, Magicicada
septendecim (L.) (Homoptera) and the lesser migratory grasshopper, Melanoplus
sanguinipes (F.) (Orthoptera). Great Lakes Entomologist 31: 161-168.
Ehrlich, A. R., and P. R. Ehrlich. 1978. Reproductive strategies in the butterflies: I.
Mating frequency, plugging, and egg number. J. Kans. Ent. Soc. 51: 666- 697.
Hadrys, H., M. Balick, and B. Schierwater. 1992. Applications of random amplified
polymorphic DNA (RAPD) in molecular ecology. Molecular Ecology. 1: 55-63.
Haymer, D. S. 1994. Random amplified polymorphic DNAs and microsatellites: What are
they, and can they tell us anything we don’t already know. Ann. Ent. Soc. Amer.
87: 717- 722.
Hölldobler, B. 1976. The behavioral ecology of mating in harvester ants (Hymenoptera:
Formicidae: Pogonomyrmex ). Behav. Ecol. Sociobiol. 1: 405- 423.
Kotrba, M. 1996. Sperm transfer by spermatophore in Diptera: new results from the
Diopsidae. Zool. J. Linn. Soc. 117: 305- 323.
Labine, P. A. 1964. Population biology of the butterfly, Euphydryas editha I. Barriers to
multiple inseminations. Evolution 18: 335- 336.
Leopold, R. A. 1976. The role of male accessory glands in insect reproduction. Ann. Rev.
Emtomol. 21: 199-221.
Lum, P. T. M. 1961. The reproductive system of some Florida mosquitoes. II. The male
accessory glands and their role. Ann. Ent. Soc. Amer. 54: 430- 433.
Marlatt, C. L. 1923. The periodical cicada. U.S.D.A. Bureau of Entomology Bulletin 71,
183 pp.
Matsumoto, K. 1987. Mating patterns of a sphragis-bearing butterfly, Luehdorfia japonica
Leech (Lepidoptera: Papilionidae), with descriptions of mating behavior. Res.
Pop. Ecol. 29: 97- 110.
Nielsen. E. T. 1959. Copulation of Glyptotendipes (Phytotendipes) paripes Edwards.
Nature (London) 184: 1252- 1253.
Parker, G. A. 1970. Sperm competition and its evolutionary consequences in the insects.
Biological Reviews 45: 525- 567.
63
Polak, M., W. T. Starmer, and J. S. F. Barker. 1998. A mating plug and male mate
choice in Drosophila hibisci Brock. Anim. Behav. 56: 919- 926.
Thornhill, R. and J. Alcock. 1983. The evolution of insect mating systems. (Harvard
University Press: Cambridge, MA)
White J. 1973. Viable hybrid young from crossmated periodical cicadas. Ecology 54: 573580.
Winston, M. L. 1987. The biology of the honey bee. (Harvard University Press:
Cambridge, MA)
Yasui, Y. 1997. Sperm competition and the significance of female multiple mating in the
predatory mite Parasitus fimetorum. Experimental and Applied Acarology 21: 651664.
64
65
I
II
III
Brood
Okanagana canadensis/ O. rimosa
1996- 1998
Species
Year
Magicicada septendecim
1995
1996
1997
Table 3.1. Study sites 1995- 1997.
UMBS
Siloam Springs
Horsepen Lake SWF
Siloam Springs SP
Location
Cheboygan/ Emmet
Rockbridge
Buckingham
Brown, Adams
County
MI
VA
VA
IL
State
Open field
Logged site
Logged site
Clearing
Characteristics
Table 3.2. Mating durations in Magicicada septendecim
and Okanagana. O. canadensis mating lasted approximately
4 times as long as the average of the others, so two
averages, one with this long mating, and one without, are
calculated for this species.
Species
O. rimosa
Average
O canadensis
Average
Average w/o outlier
Magicicada septendecim
Average (n= 43)
Duration (min: sec)
19:00
17:00
22:09
17:04
18:48 ± 2:25
80:00
19:06
15:09
23:18
34:23 ± 30:35
19:11 ± 4:05
195
195
295
167
235
235
310
60
305
354
124
440
310
220
263
354
360
263
65
148
310
405
243:24 ± 121:48
66
315
330
442
297
235
195
500
124
111
295
331
440
64
444
295
360
250
310
526
91
60
Table 3.3. Results, female mating interruption experiment. For each treatment (disrupt
1 hr., disrupt 2 hr., combined 1 and 2 hr disruption, and no disrupt), numbers of females
without seminal plugs (No SP), with firm seminal plugs (Firm SP), and with soft seminal
plugs (Soft SP) were counted. There were no statistical differences between 1 and 2 hour
mating disruptions, so these treatments were combined. Females whose matings were
interrupted were less likely to exhibit seminal plugs than were females with uninterrupted
matings (Fisher’s Exact 2- tail, P < 0.001). Soft seminal plugs were more prevalent in
interrupted matings than in uninterrupted matings (Fisher’s Exact 2- tail P < 0.001).
Mating
interruption
Total
females
No SP
Firm SP
Soft SP
Firm + Soft
1 hr.
13
8
1
4
5
2 hr.
25
17
6
2
8
≤ 2 hr.
(combined)
38
25
7
6
13
Not
interrupted
27
2
25
0
25
67
Table 3.4. Numbers of females with and without seminal plugs (SP) remating when
given free access to males. The presence of a seminal plug did not affect female likelihood
of remating: If rematings were expected to be evenly distributed between females with and
without plugs, 6 females with plugs would be expected to remate. 4 did : X 2 P > 0.10.
The data were also analyzed with a resampling algorithm, which drew a sample of 11
females from the “without SP” data: In 100,000 iterations of the algorithm, the proportion
of iterations in which 4 or fewer plugged females were included in the sample was 0.2214.
Remating
Not remating
Females with SP
Females without SP
4
16
7
9
68
69
Figure 3.1. Photograph of gel showing RAPD-PCR amplification with primer (CCT GCG CTT A).
Adults and seminal plugs show characteristic amplification products, demonstrating that seminal plugs
contain DNA. If plugs did not contain DNA, they would show no amplification product. Gel is 1% agarose
stained with EtBr, negative control is distilled water, positive controls are adult cicadas.
Seminal Plug
Seminal Plug
Adult
Adult
Adult
H2O
CHAPTER IV
FEMALE MULTIPLE MATING AND MATE CHOICE IN A PERIODICAL
CICADA, MAGICICADA SEPTENDECIM
Abstract
Explanatory hypotheses for polyandry in the periodical cicada Magicicada
septendecim fall into two categories: 1. Females remate because of inadequate insemination;
and 2. Females remate to effect mate choice. I gave mated females the opportunity to
remate after manipulating both mating duration and mate quality. Females whose first
matings were terminated prematurely tended to remate, while females allowed complete,
undisturbed matings with inappropriate, heterospecific males did not. After participating in
a complete, undisturbed mating, females were no longer sexually receptive nor were they
attractive to males. Thus, when placed in situations where they were either inadequately
inseminated or mated with low-quality mates, only females in the former manipulations
mated multiply, suggesting that polyandry in M. septendecim is best understood as the
result of sperm limitation rather than as an evolved mate choice strategy.
70
Introduction
Multiple mating is sometimes considered to be evidence that females exercise
postcopulatory mate choice (Eberhard 1985, 1996, 1997). Selection favors the evolution
of mate choice mechanisms when individuals discriminating among potential mates produce
genetically superior offspring, acquire phenotypic benefits specific to a given mate and
convertible into reproductive output, or both. Postcopulatory mate choice is a possibility
for females that mate multiply and for which insemination and fertilization are separated in
time (e.g.; Eberhard 1985, Birkhead and Møller 1993, LaMunyon and Eisner 1993,
Eberhard 1996, 1997), although some models of postcopulatory choice have been
subjected to strong criticism (Alexander et al. 1997). Alternatively, females may remate to
acquire sufficient sperm (or other resources) to complete their reproduction, in which case,
mate choice need not be invoked. Magicicada septendecim 1 females sometimes remate
(Chapter 2, Alexander 1968), although anecdotal evidence suggests that remating is not
common: During the 1998 emergence of periodical cicadas, mating pairs became
increasingly rare as the emergence progressed, the opposite of the pattern expected if
females typically remate. To evaluate whether female M. septendecim remating is an
evolved mate choice mechanism, I first discuss the general benefits females might obtain by
remating and whether such benefits involve mate choice. Although, in order to identify
selective contexts of remating, the benefits of polyandry are presented as if they belong to
strict categories, this is undoubtedly an oversimplification, because the different kinds of
benefits are likely not exclusive. I conclude by presenting the results of manipulations
designed to evaluate the causes of polyandry in M. septendecim.
Remating as bet-hedging. Multiply mating females may gain from potential
increases in offspring diversity or from bet hedging against genetically incompatible sires
1
This paper specifically concerns Magicicada septendecim. However, its findings may be applicable to
other species in this genus since the sexual behaviors of all Magicicada species share general similarities.
71
(Walker 1980, Smith 1984, Watson 1991, Zeh and Zeh 1996, 1997, Zeh 1997, Brown
1997a, Tregenza and Wedell 1998). If females remate for diversity or bet-hedging, most
or all females in a population should mate multiply, mate quality should not necessarily
increase with each successive mating (see Watson 1991), nor should mating with a lowerquality male necessarily trigger remating. If females remate only to increase offspring
diversity, sperm displacement must be incomplete, or, if complete, females must oviposit
between matings to allow multiple paternity.
Remating for replenishment. Limited female sperm storage capacity, variable
ejaculate volumes, limits to males’ ability to fully inseminate in a single mating, and
disturbances during mating may promote serial polyandry because remating allows females
to replenish or supplement inadequate sperm supplies (Boorman and Parker 1976, Walker
1980, Gromko et al. 1984, Ward et al. 1992). Females remating to replenish depleted
supplies are expected to oviposit or engage in other sperm- or resource- depleting activities
between successive matings. Some female dipterans, including Drosophila (Pyle and
Gromko 1978, Gromko and Markow 1993), Ceratitis (Nakagawa et al. 1971, Whittier and
Shelley 1993) and Rhagoletis (Neilson and McAllan 1965), homopterans (Nilaparvata
lugens; Oh 1979), and heteropterans (Riptortus, Sakurai 1996, 1998a, b) experience
reversible decreases in rates of ovipositing normal, fertilized eggs (considered an index of
fecundity and fertility), several days after mating. In Riptortus and Nilaparvata, coincident
with decreasing oviposition rate is a return to sexual behaviors typical of unmated females.
In each of the above examples, fertility and fecundity returned to high levels subsequent to
remating.
Remating and phenotypic benefits. Just as females may remate to replenish
depleted sperm, if remating increases female resource access, females may remate to obtain
direct phenotypic benefits. Mating males may provide nuptial gifts, spermatophores,
(Morris 1979, Sakaluk and Cade 1982, Thornhill and Alcock 1983, Thornhill 1984,
72
Gwynne 1986, Goulson et al. 1993, Kaitala and Wiklund 1995, Brown 1997a, b, 1999)
increased predator protection, or access to limited resources (as in resource-based leks;
Alexander 1975, Hoglund and Alatalo 1995). Promiscuous females could collect
phenotypic benefits in the same manner as any acquirable resource; here, the price of
acquisition is permitting male sexual access. Female “foraging” for male-provided
resources may be at least partially responsible for polyandry in some Lepidoptera
(Rutowski 1984, Svärd and McNeill 1994, Karlsson 1996,) and perhaps in some
Orthoptera (Sakaluk and Cade 1982, Brown 1997b) and Hymenoptera (Alcock et al.
1977). Phenotypic benefits from polyandry also include reduced costs, or at least
minimization of losses, associated with acquiescence to persistent male harassment
(Trillmich and Trillmich 1984, Clutton-Brock and Parker 1995, Jormelainen and Merilaita
1995, Arnqvist 1997, Crean and Gilburn 1998), although an effective female response to
this male tactic is to leave the harassing male’s vicinity (see Alexander and Moore 1962,
Stone 1995). Females remating only to replenish depleted supplies, and not to effect mate
choice, would be expected to remate when they run low on supplies and not necessarily
whenever a high quality male becomes available.
If male variations affecting ability to produce, acquire, or defend high-quality
resources or nuptial gifts are heritable, then direct phenotypic benefits from choosing males
with superior resources could be accompanied by genotypic benefits through jointlyproduced offspring, and female remating could be a form of postcopulatory mate choice.
Remating and genetic benefits. If females remate to improve the genetic
constitution of their offspring, then they are exercising mate choice. Genetic benefits from
multiple mating include the advantages of cuckolding a genetically inferior parental male
(Westneat et al. 1990, Birkhead and Møller 1993, Kempenaeres et al. 1992), and, under
restricted conditions, benefits from sperm competition within the female, such that male
offspring sired by especially competitive sperm might themselves produce competitive
73
sperm (Harvey and May 1989, Curtsinger 1991). If females remate to improve the genetic
quality of their offspring, no complex mechanisms need be invoked; for instance, females
with “last male” sperm priority using “threshold criteria” (Janetos 1980; Alexander et al.
1997) to choose mates could, by raising their choice threshold after each mating, “revise”
their previous choices by mating with successively higher quality males (see Gabor and
Halliday 1997). If males are readily available and not costly to acquire, then choosy
females first mated with inferior males should be most likely to remate, and females should
grant all their paternity to their best mate, wasting none on inferior ones. If remating
tendency is independent of mate quality, replenishment seems a more likely explanation.
Bias against heterospecifics. Bias against heterospecifics is often an example
of indirect mate choice (Chapter 1) rather than an evolved mate choice mechanism.
Females benefit from such biases by obtaining appropriate genetic materials for their
offspring. One form of such bias is conspecific sperm precedence, in which, regardless of
mating order, few hybrid offspring are produced when a female is mated with a series of
con- and hetero- specifics. Conspecific sperm precedence is reported in flour beetles
(Wade et al. 1994), grasshoppers (Bella et al. 1992), sunflowers (Rieseberg et al. 1995),
ground crickets (Howard and Gregory 1993, Gregory and Howard 1994; Howard,
Gregory, Chu, and Cain 1998; Howard, Reece, Gregory, Chu, and Cain 1998), and
several Drosophila species (Price 1997). If precopulatory behaviors allow a significant
proportion of heterospecific matings, females could recover from such mistakes by
remating (see Jamart et al. 1995). However, this pattern is not necessarily evidence of an
evolved post-copulatory mate choice mechanism. Instead, it may result from the inferior
competitive ability of heterospecific sperm and seminal products in the environment of the
female reproductive tract, especially if the interactions between female and ejaculate are
shaped by a long history of coevolutionary antagonism over control of fertilization (Rice
1996, Price 1997, Alexander et al. 1997, Rice 1998).
74
Cross-mated females capable of postcopulatory mate discrimination would be
expected to remate before ovipositing unless sperm priority was first in/first out, there was
no sperm mixing, and there was no way to remove stored spermathecal sperm except
through fertilization. Otherwise, the sperm dilution or replacement caused by subsequent
matings prior to oviposition would be to the female’s advantage, since it would minimize
the production of hybrid offspring, each of which represents a loss of potential fitness. If
females mated to heterospecific males do not actively seek subsequent matings with
conspecifics, then it is unlikely that they have evolved to exercise postcopulatory mate
choice.
Polyandry in M. septendecim
Precopulatory behaviors may provide female periodical cicadas an efficient first
defense against heterospecific males; precopulatory species discrimination based on song
characteristics has been demonstrated (Chapter 6), and, in addition, 78 of 80 female M.
septendecim in one of our cage experiments avoided heterospecific matings, a figure
comparable to data reported in White (1973). However, if, in spite of precopulatory
discrimination, heterospecific matings are a significant threat to females, then it is possible
that remating behaviors evolved to allow mate choice revision. Females could also remate
to discriminate among conspecific sires. To evaluate whether female remating is likely to
function in mate choice, I conducted mating experiments to determine the contexts in which
females remate, playback experiments to evaluate female sexual receptivity after mating,
and experiments to evaluate female sexual attractiveness after mating.
Female remating experiments
For M. septendecim, at the present time, we are unable to identify a class of inferior
conspecifics or determine whether, within a species, males vary in ways likely to promote
female mate choice. One way to cause females to remate is to interrupt their first matings.
75
Interruption causes remating either because it simulates a mating attempt by an inferior male
or because it results in incomplete insemination and an inadequate sperm supply. The mate
choice through remating hypothesis predicts that females should remate immediately in
response to mating with an “inferior” male and grant paternity only to later mates and that
the length of time allowed before interruption should not influence likelihood of remating.
In contrast, the replenishment hypothesis predicts that interrupted females will oviposit,
and as they deplete their sperm stores, they will remate; thus females allowed the least
amount of time for their first matings should be the most likely to remate immediately.
Furthermore, if females have an evolved mechanism for improving their circumstances
after mating with an inappropriate mate, then females mated with heterospecifics should
have high pre-oviposition remating frequencies.
Methods. I studied Brood II periodical cicadas in a large open field near
Horsepen Lake State Wildlife Management Area, Buckingham County, VA, during an
emergence lasting from late April through early June 1996. Vegetation in the field
consisted primarily of oak, maple, and tulip tree stump sprouts from a logging operation
several years before. The field was surrounded by a mixed forest composed of mature
hardwoods, primarily oaks and maples, and planted pines.
The morning after they molt, newly emerged cicadas (“teneral” cicadas) are readily
identifiable by their pale color, soft bodies, and by their position low in the vegetation. I
collected unmated, teneral female cicadas daily and placed them into fiberglass mesh
storage cages one meter in length and approximately 60 cm in diameter placed over living
vegetation. After females had matured in the storage cages without access to males, I
marked them with either dots of enamel paint or with symbols drawn onto their wings with
laboratory markers. Neither marking method has adverse effects on cicadas (Chapter 2;
Cooley et al. 1998). I caught males from those present singing and flying in the area of the
experiments and marked them in the same manner as the females.
76
I placed 5- 10 mature females and fewer males (to discourage usurpation or
harassment) in nylon tulle cages similar to the storage cages. I stocked these mating cages
between 8 A. M. and 12 noon, and most matings occurred shortly thereafter. To establish
a starting time for each observed copulation, I monitored cages continuously or scanned
them every 10-15 minutes, stopping the scan only when I had seen every cicada in the
cage. Alternatively, I ad lib scanned when I heard male courtship songs. Starting times are
thus, in some cases, estimates ± 15 minutes. Matings whose beginning times could not be
established within this degree of accuracy were used for the “no interruption” treatment (see
below). Any cicadas failing to mate the same day they were placed in mating cages were
discarded.
Mated female periodical cicadas usually exhibit a mating sign or “seminal plug” in
their genital opening (White 1973), although this structure is not always present and does
not prevent subsequent matings (Chapter 3). A seminal plug appears as a whitish or
yellowish ovoid mass when the female’s ovipositor sheath is gently pulled open. The plug
appears to be composed of dried seminal fluid (Chapter 3), and its presence is thus an
indicator that the male has transferred at least some ejaculate. In all experiments, I noted
the presence or absence of a seminal plug after each mating.
I interrupted conspecific matings after 15 min., 1 hour, 2 hours, or not at all by, at
the appropriate time, quickly removing mating pairs from the mating cages, grasping the
base of the male genitalia, and gently working male and female genitalia apart. Once
separated, males were immediately preserved in 70% ethanol. In addition, I mated female
M. septendecim with male M. cassini by confining them together in small cages. I left
mating pairs not to be disrupted undisturbed until they stopped mating, at which time I
preserved the males. I checked females for the presence of a seminal plug immediately
after disruption or cessation of mating.
77
After they had mated once, or after their first mating had been disrupted, females
were placed into cages where they had access to conspecific males and appropriate
oviposition space. I surveyed these cages at least once per hour and immediately placed
females observed ovipositing into separate oviposition cages, one female per cage. In
addition, I subjected some females to a treatment in which their mating was interrupted after
two hours and they were placed directly into oviposition cages. Once placed in oviposition
cages, females were permitted no further mating opportunities.
Oviposition cages were either surplus mating cages or fiberglass screen cages
approximately one meter in length and 30 cm in diameter. I placed all oviposition cages
over oak, maple, or beech vegetation that had been thoroughly inspected for the absence of
eggnests and that had a minimum of one meter of woody branch suitable for oviposition. I
distributed females among oviposition cages haphazardly with respect to tree species, cage
type, and cage location and removed each female from its oviposition cage when it died,
became moribund, or at the end of the experiment. I left all oviposition cages intact and
closed until the branch broke due to excessive oviposition (“flagging”) or until the end of
the experiment, at which time I removed, counted, labeled, and preserved all eggnests in
70% ethanol.
Results. Normal, uninterrupted matings lasted an average of 243.4 minutes
(Table 4.1). Females whose first matings were interrupted after 15 minutes or one hour
were most likely to remate before ovipositing, those interrupted after two hours were
equally likely to remate or not, and those not interrupted usually did not remate before
ovipositing (Table 4.2). Females whose matings were disrupted within 2 hours of starting
also tended to lack a seminal plug. Female M. septendecim first mated with M. cassini
males did not remate before ovipositing. One female whose mating was uninterrupted and
apparently normal did remate before ovipositing.
78
There is a statistical difference in number of pre-remating eggnests constructed by
females interrupted at two hours and females who were not interrupted (Table 4.3, Fig.
4.1). However, there was no difference between the total number of eggnests produced by
heterospecific-mated, mating-interrupted females allowed to remate, and uninterrupted
females.
Female sexual receptivity
Upon hearing a male’s calling phrase, a receptive female flicks her wings with a
quick motion (Chapter 5); this signal causes the male to continue courtship and attempt to
copulate. If females solicit second matings as a form of mate choice, then we would expect
them to employ many of the same courtship behaviors, such as signaling, as in first
matings, because non-signaling females would have difficulty obtaining mates.
Mating status and receptivity. I identified a high quality recording of
Magicicada septendecim song in the University of Michigan Museum of Zoology Sound
Library. I imported this recording into a Macintosh IIcx computer using Mac Recorder
hardware and Sound Edit software. I removed extraneous background noise, isolated a
single male song, and confirmed that females signal in response to playbacks of this song
(Chapter 5).
In 1997, at Siloam Springs State Park, Brown and Adams Counties, Illinois,
(Brood III), I collected 44 unmated teneral female M. septendecim and allowed them to
mature in single-sex storage cages. Half (22) were mated to males captured from the
chorus, half were left unmated. Mated and unmated females into two cages, and in each of
4 subsequent days, I played recorded M. septendecim calling songs to the surviving
females for 2 minutes and recorded the number mated and unmated females responding.
Results. None of the mated females responded to playbacks, while at least half
of the unmated females responded each day with wing flicks (Table 4.4).
79
Mating completeness and receptivity. In June 1997, I conducted
experiments combining the approaches of the female remating experiments and the
playback experiments to determine how mating interruption would affect the likelihood that
a female would solicit a second mating. As before, I allowed the cicadas to mature in
fiberglass cages and then marked them individually prior to use in experiments. After the
cicadas were mature, I allowed them to mate, disrupting the matings after 15 min, 1 hour, 2
hours, or not at all. I discarded the males and placed the females in cages without access to
males. I visited female cages at least twice daily, played artificial male calling songs, and
recorded female responses.
Results. Females whose matings were disrupted were likely to respond
positively to male song, while normally mated females never responded. All disruption
treatments resulted in an equal likelihood of female response (Table 4.5). These results are
compatible with the 1996 experiments, in which all disruption treatments led to an elevated
incidence of remating.
Female sexual attractiveness
Our playback experiments demonstrate that normally mated females are not sexually
receptive to courting males. Males’ willingness to court mated and unmated females is
potentially a separate issue (for a possible example see Gromko and Markow 1993). If
there is even a slight chance that a mated female will remate, such as if she were to remate
to effect mate choice, then males should find mated females sexually attractive and court
them. If, on the other hand, female remating is an unusual event, one that happens only
under special circumstances, males should seek unmated females instead of wasting time
and effort courting mated and unreceptive females, and they should not behave as though
mated females are sexually attractive.
80
Methods. Unmated female M. septendecim confined in cages signal to courting
males outside. Cages containing signaling females thus are attractive to males, and males
seeking mates congregate on the outsides of such cages. I evaluated the relative sexual
attractiveness of mated and unmated females by confining them in cages and censusing the
males observed on the outsides of the cages.
I erected three large, cylindrical, white tulle cages 1.5 m in diameter and 2 m tall
over stump sprouts, all at the same distance from the forest edge in our Horsepen Lake, VA
study site. I collected unmated and mated females in the same manner as for the playback
experiments and placed 13 mated females in one cage, 13 unmated females of the same age
in another, and nothing in the third cage. The cages were scanned at intervals no greater
than one hour over the course of a two-day period and recorded the number of males on the
exterior surface of each cage. At each census, all males were colllected from the exterior of
the cages and removed to a location at least 15 m distant.
Results. Because the males at each census time were unlikely to be the same, each
census is treated as an independent data point. Significantly greater numbers of males were
observed on the cage containing unmated females than on the empty cage, demonstrating
that it was the females in the cage, not the cage itself, that attracted courting males.
Significantly fewer males visited the cage containing mated females than the cage
containing unmated females, indicating that males are not strongly sexually attracted to
mated females. (Table 4.6).
Discussion
Mating aggregations suggest the action of sexual selection (Emlen and Oring 1977,
Alexander and Borgia 1979, Andersson 1994, Höglund and Alatalo 1995). Alexander
(1975) considered periodical cicada male singing aggregations to be non-resource based
leks. In such leks, males neither control nor have exclusive access to resources (other than
81
gametes) required by females, and thus females visit the lek only for the purpose of mating.
Although female periodical cicadas occasionally remate, and although the situations faced
by females in these experiments would seem to promote postcopulatory mate choice, the
data provide scant evidence of a postcopulatory mate choice mechanism2. If females were
able to accomplish mate choice by remating, they might be expected to detect the
inappropriateness of a heterospecific mate or the inferiority of an abbreviated ejaculate and
remate as quickly as possible. However, in our experiments, only females whose first
matings were extremely brief were attractive to males, actively solicited matings, and
reliably remated before ovipositing.
The readiness of mating interrupted or cross-mated females to oviposit makes
evolved postcopulatory mate choice doubtful. The failure of cross-mated females to remate
before ovipositing suggests that females are unable to evaluate male quality after the mating
act has commenced or that they realize no benefits from doing so. Thus, any speciesidentification mechanisms of periodical cicada mating behavior must be precopulatory.
Magicicada’s complex, species-specific courtship makes it unlikely that females are
routinely in close proximity to heterospecifics or that they normally mate with them. The
failure of female postcopulatory behavior to discriminate against mates as unsuitable as
heterospecifics and the demonstrated post-mating change in both female receptivity and
sexual attractiveness indirectly support the importance of precopulatory discrimination in
Magicicada.
For periodical cicada females, mating is not known to provide any phenotypic
benefits, either through ejaculatory nutrients or nuptial gifts, nor do males restrict female
access to resources (other than sperm). Phenotypic benefits through reduced costs from
2
One female whose mating was uninterrupted and apparently normal did remate before ovipositing; this
observation has no real significance, but it is consistent with female remating behavior being an adaptation
to discriminate against poor quality mates, because at least some females who were randomly paired with
males would be expected to remate because at least some would have been presented with a better alternative
to their first mating.
82
acquiescence to male harassment may be applicable, but as noted elsewhere, females can
simply fly away from persistent courting males. Furthermore, because the only male
behavior that effectively blocks the access of other males is copulation (pers. obs), mating
with a persistent male would only immunize female cicadas from courtship during the
actual mating act, would give the female no additional feeding time (feeding may be carried
out alone or simultaneously with many other behaviors, including copulation; pers. obs.
and similar to ladybird beetles, see Obata 1988) and could only cost oviposition time.
Another direct cost may concern male health; in Magicicada, contagious and fatal infections
of Massospora cicadina Peck, an entomophagous fungus, produce a class of potential
mates best avoided and with whom any interactions carry a risk of infection (Soper 1974,
Soper et al. 1976, Alexander et al. ms.). Yet female polyandry cannot be explained as an
infection-avoidance tactic or as a tactic to avoid genetically susceptible males because
females should be selected to identify and avoid infected males early in the courtship
sequence and avoid excessive intimate contact. Lastly, the fact that few females mate
multiply suggests that there are not likely to be significant direct phenotypic benefits
promoting remating.
Females must evidently require some threshold quantity of sperm before
ovipositing; once they use up a substantial quantity of stored sperm, they may remate.
Evidence in support of this statement includes the results that females interrupted at two
hours produced, on average, fewer eggnests before remating than the total number
produced by females who were not interrupted (Table 4.3). However, the experiments
also indicate that females interrupted at two hours and deprived of further mating
opportunities are capable of producing as many eggnests as uninterrupted females.
Possibly, when a female’s sperm stores fall below some threshold volume, she will remate
if given the opportunity; yet regardless of whether she has access to mating opportunities or
not, she will continue to oviposit for as long as possible. This pattern would be consistent
with the existence of an evolved mechanism that caused females to replenish sperm
83
supplies well before exhausting their spermathecal contents, perhaps a form of bet-hedging
against difficulties of efficiently finding a mate. Alternatively, if it is never difficult for
females to find mates, females may have no evolved mechanisms to stop oviposition.
Females may maladaptively oviposit unfertilized eggs (Graham and Cochran 1954) after
exhausting their sperm stores because the ready availability of receptive mating partners
means that females are never in the position of lacking sperm; thus we would expect no
history of selection on females for suspension of oviposition pending sperm replenishment.
84
Literature Cited
Alcock, J., G. C. Eickwort, and K. R. Eickwort. 1977. The reproductive behavior of
Anthidium maculosum (Hymenoptera: Megachilidae) and the evolutionary
significance of multiple copulations by females. Behav. Ecol. Sociobiol. 2: 385396.
Alexander, R. D., and G. Borgia. 1979. On the origin and basis of the male-female
phenomenon. In: M. S. Blum and N. A. Blum, eds. Sexual selection and
reproductive competition in insects. ( Academic Press: New York).
Alexander, R. D. 1968. Arthropods. Pp. 167-216 in: T. A. Sebeok, ed. Animal
Communication. Techniques of Study and Results of Research. (Indiana Univ.
Press: Bloomington, IN).
Alexander, R. D., 1975. Natural selection and specialized chorusing behavior in chorusing
insects. Pp. 35-77 in: D. Pimentel, ed. Insects, Science, and Society. (Academic
Press, New York).
Alexander, R. D., D. C. Marshall, and J. R. Cooley, 1997. Evolutionary Perspectives on
Insect Mating. Pp. 1- 31 in B. Crespi, J. Choe, eds. The evolution of mating
systems in insects and arachnids. (Cambridge University Press).
Alexander, R. D., J. R. Cooley, and D. C. Marshall. (ms). A specialized fungal parasite
(Massospora cicadina) modifies the behavior of periodical cicadas (Homoptera:
Cicadidae: Magicicada).
Alexander, R. D., and T. E. Moore. 1962. The evolutionary relationships of 17-year and
13-year cicadas, and three new species. (Homoptera: Cicadidae, Magicicada).
Misc. Publ. no. 121. Univ. of Michigan Museum of Zoology, Ann Arbor, MI.
Andersson, M. 1994. Sexual selection. (Princeton University Press: Princeton NJ).
Arnqvist, G. 1997. The evolution of water strider mating sytems: Causes and
consequences of sexual conflicts. Pp. 146- 163 in B. Crespi, J. Choe, eds. The
evolution of mating systems in insects and arachnids. (Cambridge University
Press).
Bella, J. L., R. K. Butlin, C. Ferris, and G. M. Hewitt. 1992. Asymmetrical homogamy
and unequal sex ratio from reciprocal mating-order crosses between Chorthippus
parallelus subspecies. Heredity 68: 345- 352.
Birkhead, T. and A. Møller. 1993. Female control of paternity. TREE 8: 100- 104.
Boorman, E., and G. A. Parker. 1976. Sperm (ejaculate) competition in Drosophila
melanogaster, and the reproductive value of females to males in relation to female
age and mating status. Ecological Entomology 1: 145- 155.
Brown, W. D. 1997a. Female remating and the intensity of female choice in black-horned
tree crickets, Oecanthus nigricornis. Behavioral Ecology 8: 66-74.
85
Brown, W. D. 1997b. Courtship feeding in tree crickets increases insemination and female
reproductive life span. Anim. Behav. 54: 1369-1382.
Brown, W. D. 1999. Mate choice in tree crickets and their kin. Ann. Rev. Entomol. 44:
371- 396.
Clutton-Brock, T. H., and G. A. Parker. 1995. Sexual coercion in animal societies.
Anim. Behav. 49: 1345- 1365.
Cooley, J. R., G. S. Hammond, and D. C. Marshall. 1998. The effects of enamel paint
marks on the behavior and survival of the periodical cicada, Magicicada
septendecim (L.) (Homoptera) and the lesser migratory grasshopper, Melanoplus
sanguinipes (F.) (Orthoptera). Great Lakes Entomologist 31: 161-168.
Crean, C. S., and A. S. Gilburn. 1998. Sexual selection as a side-effect of sexual conflict
in the seaweed fly, Coelopa ursina (Diptera: Coelopidae). Anim. Behav. 56: 14051410.
Curtsinger, J. W. 1991. Sperm competition and the evolution of multiple mating. Am. Nat.
138: 93-102.
Eberhard, W. G. 1985. Sexual Selection and Animal Genitalia. (Harvard University Press:
Cambridge, MA).
Eberhard, W. G., 1997. Cryptic female choice. Pp. 32- 57 in B. Crespi, J. Choe, eds.
The evolution of mating systems in insects and arachnids. (Cambridge University
Press).
Eberhard. W. G. 1996. Female control: Sexual selection by cryptic female choice.
(Princeton University Press: Princeton, NJ).
Emlen, S. T., and L. W. Oring. 1977. Ecology, sexual selection, and the evolution of
mating systems. Science (Wash. D.C. ) 197: 215- 223.
Gabor, C. R., and T. R. Halliday. 1997. Sequential mate choice by multiply mating
smooth newts: Females become more choosy. Behav. Ecol. 8: 162- 166.
Goulson, D., M. C. Birch, and T. D. Wyatt. 1993. Paternal investment in relation to size
in the deathwatch beetle, Xestobium rufovillosum (Coleoptera: anobiidae) and
evidence for female selection for large mates. J. Ins. Behavior 6: 539- 547.
Graham, C., and A. B. Cochran. 1954. The periodical cicada in Maryland in 1953. J.
Econ. Ent. 47:242-244.
Gregory, P. G., and D. J. Howard. 1994. A post-insemination barrier to fertilization
isolated two closely related ground crickets. Evolution 48: 705-710.
Gromko, M. H., and T. A. Markow. 1993. Courtship and remating in field populations of
Drosophila. Anim. Behav. 45: 253- 262.
Gromko, M. H., M. E. A. Newport, and M. G. Kortier. 1984. Sperm dependence of
female receptivity to remating in Drosophila melanogaster. Evolution 38: 12731282.
86
Gwynne, D. T. 1986. Courtship feeding in katydids (Orthoptera: Tettigoniidae):
investment in offspring or in obtaining fertilizations? Am. Nat. 128: 342- 352.
Harvey, P. H., and R. M. May. 1989. Out for the sperm count. Nature (London) 337:
508- 509.
Höglund, J. and R. V. Alatalo. 1995. Leks. (Princeton University Press: Princeton NJ).
Howard, D. J., and P. G. Gregory. 1993. Post-insemination signaling systems and
reinforcement. Phil. Trans. Roy. Soc. Lond. B. Biol. Sci. 340: 231-236.
Howard, D. J., M. Reece, P. G. Gregory, J. Chu, and M. L. Cain. 1998. The evolution
of barriers to fertilization between closely related organisms. Pp. 279-288 in D. J.
Howard and S. H. Berlocher, eds. Endless Forms: Species and Speciation.
(Oxford University Press: New York).
Howard, D. J., P. G. Gregory, J. Chu, and M. L. Cain. 1998. Conspecific sperm
precedence is an effective barrier to hybridization between closely related species.
Evolution 52: 511-516.
Jamart, J. A., P. Caseres, M. C. Carracedo, and R. Piñero. 1995. Consequences of
homo- and heterospecific rapid remating on the fitness of Drosophila melanogaster
females. J. Ins. Physiol. 41: 1019-1026.
Janetos, A. C., 1980. Strategies of female mate choice: A theoretical analysis. Behav.
Ecol. Sociobiol. 7: 107-112.
Jormelainen, V. and S. Merilaita. 1995. Female resistance and duration of mate-gurading
in three aquatic peracarids (Crustacea). Behav. Ecol. Sociobiol. 36: 43- 48.
Kaitala, A. and C. Wiklund. 1995. Female mate choice and mating costs in the
polyandrous butterfly Pieris napi (Lepidoptera: Pieridae) J. Ins. Behavior 8: 355363.
Karlsson, B. 1996. Male reproductive reserves in relation to mating system in butterflies:
A comparative study. Proc. Roy. Soc. Lond. B. 263: 187-192.
Kempenaers, B., G. R. Verhayen, M. Van der Broeck, T. Burke, C. Van Broeckhoven,
and A. A. Dhondt. 1992. Extra-pair paternity results from female preference for
high-quality males in the blue tit. Nature (London) 357: 494-496.
LaMunyon, C. W. and T. Eisner. 1993. Postcopulatory sexual selection in an arctiid moth
(Utethesia ornatrix) Proc. Nat. Acad. Sci. USA 90: 4689- 4692.
Morris, G. K. 1979. Mating systems, paternal investment, and aggressive behavior of
acoustic orthoptera. Florida Entomologist 62: 9- 17.
Nakagawa, S., G. J. Farias, D. Suda, R. T. Cunningham, and D. L. Chambers. 1971.
Reproduction of the Mediterranean fruit fly: Frequency of mating in the laboratory.
Ann. Ent. Soc. Amer. 64: 949-950.
Neilson, W. T. A., and J. W. McAllen. 1965. Effects of mating on fecundity of the Apple
Maggot, Rhagoletis pomonella (Walsh). Canadian Entomologist 97: 276- 279.
87
Obata, S. 1988. Mating refusal and its significance in females of the ladybird beetle,
Harmonia axydris . Physiol. Entomol. 13: 193- 199.
Oh, R. J. 1979. Repeated copulation in the brown planthopper, Nilaparvata lugens Stål
(Homoptera: Delphacidae). Ecological Entomology 4: 345- 353.
Price, C. S. C. 1997. Conspecific sperm precedence in Drosophila. Nature (London) 388:
663-666.
Pyle, D. W., and M. H. Gromko. 1978. Repeated mating by female Drosophila
melanogaster: The adaptive importance. Experientia 15: 449- 450.
Rice, W. R. 1996. Sexually antagonistic male adaptation triggered by experimental arrest
of female evolution. Nature (London) 381- 232- 234.
Rice, W. R. 1998. Intergenomic conflict, interlocus antagonistic coevolution, and the
evolution of reproductive isolation. Pp. 261-270 in D. J. Howard and S. H.
Berlocher, eds. Endless Forms: Species and Speciation. (Oxford University Press:
New York).
Rieseberg, L. H., A. M. Desrochers, and S. J. Youn. 1995. Interspecific pollen
competition as a reproductive barrier between sympatric species of Helianthus
(Asteraceae). Am. J. Bot. 82: 515-519.
Rutowski, R. L. 1984. Production and use of secretions passed by males at copulation in
Pieris protodice (Lepidoptera: Pieridae). Psyche 91: 141- 152.
Sakaluk, S. K., and W. H. Cade. 1982. Orthopteran Mating Systems: sexual competition
in a diverse group of insects. (Westview Press: Boulder, CO).
Sakurai, T. 1996. Multiple mating and its effect on female reproductive output in the bean
bug Reptortus clavatus (Heteroptera: Alydidae). Ann. Ent. Soc. Amer. 89: 481485.
Sakurai, T. 1998a. Variation in time to sperm depletion and oviposition patterns in females
of Riptortus clavatus (Heteroptera : Alydidae). Ann. Ent. Soc. Amer. 91: 737-740.
Sakurai, T. 1998b. Receptivity of female remating and sperm number in the sperm storage
organ in the bean bug, Riptortus clavatus (Heteroptera : Alydidae). Res. Popul.
Ecol. 40: 167- 172.
Smith, R. L. 1984. Human sperm competition. Ch. 19 in: Smith, R. L., ed. Sperm
competition and the evolution of animal mating systems. (Academic Press: New
York).
Soper, R. 1974. The genus Massospora entomopathogenic for cicadas. Part I. Taxonomy
of the genus. Mycotaxon 1:13-40.
Soper, R., A. J. Delyzer, and L. F. R. Smith. 1976. The genus Massospora
entomopathogenic for cicadas. Part II. Biology of Massospora levispora and its
host Okanagana rimosa, with notes on Massospora cicadina on the periodical
cicadas. Ann. Ent. Soc. Amer. 69:89-95.
88
Stone, G. N. 1995. Female foraging responses to sexual harassment in the solitary bee
Anthophora plumipes. Anim. Behav. 50: 405- 412.
Svärd, L., and J. N. McNeil. 1994. “Female benefit, male risk” polyandry in the true
armyworm, Pseudaletia unipunctata. Behav. Ecol. Sociobiol. 35: 319- 326.
Thornhill, R. 1984. Alternative female choice tactics in the scorpionfly Hylobittacus
apicalis (Mecoptera) and their implications. Amer. Zool. 24: 367- 383.
Thornhill, R. and J. Alcock. 1983. The evolution of insect mating systems. (Harvard
Unversity Press: Cambridge, MA).
Tregenza, T., and N. Wedell. 1998. Benefits of multiple mates in the cricket Gryllus
bimaculatus. Evolution 52: 1726-1730.
Trillmich, F. and K. G. K. Trillmich. 1984. The mating systems of pinnipeds and marine
iguanas: convergent evolution of polygyny. Biol. J. Linn . Soc. 21: 209- 216.
Wade, M. J., H. Patterson, N. W. Chang, and N. A. Johnson. 1994. Postcopulatory,
prezygotic isolation in flour beetles. Heredity 72: 163- 167.
Walker, W. F. 1980. Sperm utilization strategies in nonsocial insects. American
Naturalist 115: 780- 799.
Ward, P. I., J. Hemmi, and T. Roosli. 1992. Sexual conflict in the dung fly Sepsis
cynipsea. Functional Ecology 6: 649-653.
Watson, P. J. 1991. Multiple paternity as genetic bet-hedging in female sierra dome
spiders, Linyphia litigiosa (Linyphiidae). Anim. Behav. 41: 343- 360.
Westneat, D. F, P. W. Sherman, and M. L. Morton. 1990. The ecology and evolution of
extra-pair copulations in birds. Current Ornithology 7: 331- 369.
White J., 1973. Viable hybrid young from crossmated periodical cicadas. Ecology 54: 573580.
Whittier, T. S., and T. E. Shelley. 1993. Productivity of singly vs. multiply mated female
Meditteranean fruit flies, Ceratitus capitata. J. Kans. Ent. Soc. 66: 200- 209.
Zeh, J. 1997. Polyandry and enhanced reproductive success in the harlequin-beetle-riding
pseudoscorpion. Behav. Ecol. Sociobiol. 40: 111- 118.
Zeh, J. A., and D. W. Zeh. 1996. The evolution of polyandry I: Intragenomic conflict and
genetic incompatibility. Proc. R. Soc. Lond. B. 263: 1711- 1717.
Zeh, J. A., and D. W. Zeh. 1997. The evolution of polyandry II: Post-copulatory defences
against genetic incompatibility. Proc. R. Soc. Lond. B. 264: 69-75.
89
Table 4.1. Mating durations (minutes) in Magicicada
septendecim.
Average (n= 43)
195
195
295
167
235
235
310
60
305
354
124
440
310
220
263
354
360
263
65
148
310
405
243:24 ± 121:48
90
315
330
442
297
235
195
500
124
111
295
331
440
64
444
295
360
250
310
526
91
60
Table 4.2. Mating interruption experiment: Likelihood of remating first or ovipositing
first, and presence/absence of seminal plug. All P-values Bonferroni corrected for 18
comparisons. ≤ 1hr category includes combined data from 15 minute/ 1 hr. disruption; ≤ 2
hr category includes combined data from all disruption treatments.
Treatment
Remate first
Oviposit first
Disrupt at 15 min
14
0
Disrupt at 1 hr.
11
1
Disrupt at 2 hr.
14
11
Not disrupted
Conspecific (M. septendecim) mate
1
31
Heterospecific (M. cassini) mate
0
5
Comparison
Remating Frequency
15 min. vs. not disrupt
1 hr. vs. not disrupt
2 hr. vs. not disrupt
≤1 hr. vs. not disrupt
≤2 hr. vs. not disrupt
15 min. vs. 1 hr.
15 min. vs. 2 hr.
1 hr. vs. 2 hr.
M. cassini vs. not disrupt
SP
No SP
0
1
6
14
9
19
26
2
5
0
P (Fisher’s Exact 2- Tail)
< 0.001*
< 0.001*
< 0.001*
< 0.001*
< 0.001*
< 1.000
< 0.040*
< 0.220
< 1.000
Presence of seminal plug
15 min. vs. not disrupt
1 hr. vs. not disrupt
2 hr. vs. not disrupt
≤1 hr. vs. not disrupt
≤2 hr. vs. not disrupt
15 min. vs. 1 hr.
15 min. vs. 2 hr.
1 hr. vs. 2 hr.
M. cassini vs. not disrupt
* indicates statistically significant difference.
91
< 0.001*
< 0.001*
< 0.001*
< 0.001*
< 0.001*
< 1.000
< 1.000
< 1.000
< 1.000
Table 4.3. Counts of female M. septendecim eggnests in mating interruption
experiments. Conspecific mates are M. septendecim; heterospecific mates are M. cassini.
Mate
Uninterrupted mating:
Conspecific
Conspecific
Conspecific
Conspecific
Conspecific
Conspecific
Conspecific
Conspecific
Conspecific
Average
Heterospecific
Heterospecific
Heterospecific
Heterospecific
Average
Eggnests
after first
mating
Remated
Additional
eggnests after
second mating
29
31
14
26
34
22
23
34
1
23.8 ± 10.7
no
no
no
no
no
no
no
no
no
-
29
31
14
26
34
22
23
34
1
23.8 ± 10.7
45
32
29
97
50.8 ± 31.6
no
no
no
no
-
45
32
29
97
50.8 ± 31.6
yes
yes
yes
yes
yes
yes
yes
0
3
7
20
20
20
12
11.7 ± 8.6
9
26
7
30
29
20
26
21.0 ± 9.5
Mating interrupted at 2 hr:
Conspecific
9
Conspecific
23
Conspecific
0
Conspecific
10
Conspecific
9
Conspecific
0
Conspecific
14
Average
9.29 ± 8.0
Total
eggnests
Conspecific
Conspecific
Conspecific
Average
5
35
41
27.0 ±19.3
no
no
no
-
5
35
41
27.0 ± 19.3
Conspecific
Conspecific
Conspecific
Conspecific
Conspecific
Conspecific
Conspecific
Average
39
51
35
10
87
30
33
40.7 ± 23.8
no possibility
no possibility
no possibility
no possibility
no possibility
no possibility
no possibility
-
39
51
35
10
87
30
33
40.7 ± 23.8
Comparison vs. uninterrupted conspecific mating (Wilcoxon Signed Ranks):
Pre-remating eggnests, conspecific mating interrupted at 2 hours with possibility of remating:
Total eggnests, conspecific mating interrupted at 2 hours with subsequent remating:
Total eggnests, conspecific mating interrupted at 2 hours without subsequent remating:
Total eggnests, conspecific mating interrupted at 2 hours without possibility of remating:
Total eggnests, heterospecific mating interrupted at 2 hours with possibility of remating:
92
P≤
P≤
P≤
P≤
P≤
0.018
0.128
0.593
0.090
0.068
Table 4.4. Mated and unmated female M. septendecim responses to 2 minute playbacks
of male calling song. Females that produced at least one wing flick in response to
playbacks of male songs were scored as responding positively.
Day
Female
n
1
Mated
Unmated
22
17
Response
(+)
(-)
0
22
9
8
2
Mated
Unmated
22
16
0
8
22
8
≤ 0.001
3
Mated
Unmated
20
15
0
7
20
8
≤ 0.001
4
Mated
Unmated
18
14
0
7
18
7
≤ 0.001
93
P
(Fisher’s Exact 2- tailed)
≤ 0.001
Table 4.5. Mating interruption and female M. septendecim wing flick responses to
playbacks of male calls. Mating-interrupted females were likely to signal in response to
male calls, and all disruption treatments were equally likely to cause females to remain
responsive.
Treatment
Disrupt at 15 min.
Disrupt at 1 hr.
Disrupt at 2 hr.
Not disrupted
Number of
females responding
with wing flick signal
8
8
5
0
Number of
females
not responding
1
3
4
17
Comparison
P (Fisher’s Exact 2- tail)
15 min. vs. not disrupt
1 hr. vs. not disrupt
2 hr. vs. not disrupt
15 min. vs. 1 hr.
15 min. vs. 2 hr.
1 hr. vs. 2 hr.
≤ 0.001*
≤ 0.001*
≤ 0.002*
≤ 0.591
≤ 0.294
≤ 0.642
* indicates statistically significant difference.
94
Table 4.6. Census data from female attractiveness experiment. The cage containing
unmated females had a significantly greater number of visiting males than did the empty
cage (Wilcoxon Signed Ranks test, Z= -3.320, P≤ 0.001), and significantly fewer males
visited the cage containing mated females than the cage containing unmated females
(Wilcoxon Signed Ranks test, Z= 3.331, P≤ 0.001).
Time
Day 1
1:08
1:37
2:27
2:39
3:20
4:09
4:31
5:15
Day 2
9:00
9:31
10:01
10:30
10:50
11:07
11:17
11:30
11:42
Males found on
cage containing
unmated females
Males found on
cage containing
mated females
Males found on
empty
cage
1
2
2
0
4
3
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5
5
6
4
5
2
0
2
2
0
0
2
0
0
0
0
0
0
1
0
1
0
0
0
0
0
1
95
96
Control
Pre-remating
Total
Conspecific mating
Permitted to remate
Not
Permitted
To remate
Heterospecific
(M. cassini)
Mate
Figure 4.1. Numbers of eggnests laid by normally-mated female M. septendecim (control), mating-interrupted females allowed to
remate (pre-remating eggnests and total eggnests), mating interrupted females not premitted to remate, and females mated with
heterospecifics (M. cassini).
Comparison vs. uninterrupted conspecific mating (Wilcoxon Signed Ranks):
Pre-remating eggnests, conspecific mating interrupted with possibility of remating:
P ≤ 0.018
Total eggnests, conspecific mating interrupted with subsequent remating:
P ≤ 0.128
Total eggnests, conspecific mating interrupted without possibility of remating:
P ≤ 0.090
Total eggnests, heterospecific mating interrupted with possibility of remating:
P ≤ 0.068
0
10
20
30
40
50
60
70
CHAPTER V
SEXUAL SIGNALING IN PERIODICAL CICADAS, MAGICICADA SPP.
Abstract
We describe the nature and timing of a previously unknown female sexual
receptivity signal in Magicicada. Females produce “wing-flick” signals in response to male
calls with a timing that is species-specific. Males perceive both acoustical and visual
components of wing-flick signals. Chorusing males perceiving female signals alter their
behavior by increasing the number of calls between flights and localizing their mate search.
A male receiving repeated nearby responses to his calls approaches the signaling female
while calling and begins late-stage courtship behaviors leading to copulation. Using the
wing-flick signal as an assay of female receptivity, we demonstrate the functional
significance of the complex, two-part structure of male calls in M. septendecim, M.
tredecim, and M. neotredecim. These species’ calls include a main element and a
downslur; calls without downslurs are unlikely to provoke female responses against a
background chorus, but such call fragments may elicit responses in the absence of a
background chorus. Normally, only receptive females produce wing-flick signals;
however, a fungal pathogen evidently increases its odds of transmission by causing
infected males to signal. We also report a novel form of male-male acoustic competition:
97
Males engaged in close-range or protracted courtships obscure the downslurs of arriving
potential interlopers with “interference buzzes,” reducing the likelihood of a female
response and thus increasing the likelihood that interlopers will depart and search
elsewhere.
Introduction
Detailed studies of insect communication have provided valuable insights into topics
as diverse as insect systematics (e.g. Alexander 1957a; Walker 1962, 1963), speciation
(e.g. Henry 1985, Otte 1992, Butlin and Ritchie 1994, Alexander et al. 1997),
development and inheritance (e.g. Fulton 1933, Bigelow 1960, Olvido and Mousseau
1995, Shaw 1996), conflicts of interest and deception (e.g. Lloyd 1966, Lloyd 1979,
Spooner 1968, Otte 1974; Alexander et al. 1997), and sexual selection (e.g. Alexander
1975, Thornhill and Alcock 1983, Ritchie 1996). The “singing insects” (sound-producing
members of the Orthoptera and Cicadidae) have played a central role in this research. Many
key contributions have resulted from long-term synthetic studies of individual species or
species groups (e.g. Alexander and Moore 1962, Walker 1964, Lloyd 1966, Spooner
1968, Lloyd 1979, Otte and Alexander 1983, Otte 1994). One such group, the periodical
cicadas of eastern North America (Magicicada spp.1, Table 5.1), has attracted interest
because of its extraordinary combination of unique traits, including long life cycles (13 or
17 years), synchronized development, periodical adult emergences, near-perfect sympatry
and synchrony of multiple related species, dense populations and loud, conspicuous male
sexual behaviors. Periodical cicada aggregations served as the primary model for the
concept of the non-resource-based lek (Alexander 1975), and they offer an important
opportunity to compare and contrast mating aggregations in insects and vertebrates
1
Because the songs and behaviors of the 13- and 17- year cognate species are generally similar, they are
abbreviated herewith as M. –decim, M. –cassini, and M. –decula.
98
(Bradbury 1981, 1985). Although their remarkable characteristics have made Magicicada a
focus of research on ecology, life cycle evolution, and behavior (Williams and Simon
1995), central questions about Magicicada mating behavior and communication remain
unanswered.
Calling male periodical cicadas aggregate to form choruses that attract sexually
receptive females (Alexander and Moore 1958, 1962). Teneral (adult immature) and
ovipositing females are also found in these aggregations, although they may not be
specifically attracted to them. Magicicada chorusing consists of “sing-fly” behavior, or
short (ca. 3-15 s) bouts of singing alternated with local flights of variable length (up to
several meters; Alexander and Moore 1962, Alexander 1975, Dunning et al. 1979). Under
some conditions males appear to engage in local mate search behavior involving extremely
short flights or walking between calls. Upon locating a potentially receptive female, a
chorusing male may begin courtship, which involves a stereotypical series of behaviors and
distinctive acoustical and visual signals (Alexander and Moore 1962; Alexander 1968,
1975; Dunning et al. 1979; Table 5.2, Fig. 5.1). Courtship may be followed quickly by
copulation or may last hours and include long periods of waiting (Alexander 1968). These
observations have led to the suggestion that periodical cicada females may be extremely
choosy, imposing strong sexual selection on males (Alexander 1975) and that males locate
females by progressively narrowing their search as they shift from chorusing to local mate
search to courtship. The cues causing males to identify a suitable courtship target and
switch from chorusing (not directed at a specific individual) to courtship behavior (directed
at an individual female), or from extended courtship to copulation, have until now remained
unknown (Alexander 1968). The function of male calling song has also remained poorly
understood; Alexander (1975) noted that, because females are attracted to choruses of many
individuals yet no males within choruses employ silent searching tactics, the calling song
must play a role in individual males’ mating success.
99
We describe a visual and acoustical female receptivity signal, hereafter referred to as
a “wing-flick.” Receptive Magicicada females flick their wings with a quick motion in
timed response to male calls; males perceiving the signal respond by immediately dropping
out of the chorus and engaging in courtship behaviors directed at the signaling female. The
discovery of this signal has facilitated the development and testing of new hypotheses for
the adaptive nature of Magicicada chorusing behavior and the evolution of male calling and
courtship songs. The discovery also provides new information about the transmission of a
specialized fungal parasite of Magicicada, Massospora cicadina Peck. Although normal
males have never been observed making wing-flick signals, males infected with the fungus
sometimes produce wing-flick signals in response to the calls of other males; this
behavioral modification is presumably a fungal adaptation to increase spore transmission
(see also Alexander et al. ms). In addition, we report acoustical jamming signals produced
by Magicicada males in close-range competition for receptive females, a form of intrasexual
mating competition not previously reported in insects.
General materials and methods
Periodical cicadas in different regions emerge in different years; the populations of
different regions are called “broods.” We first observed female Magicicada signaling to
males in Brood I in 1995. We continued our studies of sexual signaling in 1996-1998
(Table 5.3) on both 13-and 17-year cicadas, concentrating on the M. -decim species but
including the others when sufficient numbers were available. We used a Macintosh
computer and Canary software (Cornell Bioacoustics Laboratory) for acoustical analyses,
and Sound Edit software (MacroMedia) for modification of songs and synthesis of artificial
sounds. Playback equipment consisted of a Sony WM-D6C cassette player or a Macintosh
powerbook computer connected to a Radio Shack SA-10 amplifier driving a 3” midrange
speaker for M. -decim calls or a tweeter for M. -cassini calls. We maintained playback call
intensity at natural levels, ca. 75 db at 20 cm. In all years, we kept cicadas in captivity, but
100
within their natural environment, by placing them in ca. 200 liter cages made by enclosing
living vegetation with black fiberglass screen or white nylon tulle. Some observations
were completed in larger 4x4x4 m “flight cages” placed over living woody vegetation.
Many playback experiments were conducted using 22x24x22 cm screen test chambers.
Magicicada nymphs usually emerge and undergo their final molt in the evening
(Maier 1982). The next morning, newly emerged “teneral” adults are identifiable by their
dull color, soft bodies (especially diagnostic in the ovipositor), and by their position low in
the vegetation. Although mated females commonly have a hardened white seminal plug in
the genital opening (White 1973), this plug is occasionally absent in mated females and is
therefore an unreliable indicator. To ensure that we used only unmated females in our
playback experiments, we collected teneral females early each morning and stored them in
single-sex cages.
Female responses to male calls
The nature of the female signal and the “rapprochement duet”
In 1995-1998, we observed interactions between male and female Magicicada. We
noted whether females produced wing-flick signals and how the responses were timed in
relation to male calls, using sonograms from audio-and videotape to measure timing. We
also recorded female M. –decim and M. -cassini responses to playbacks of male calling
song and how males responded to natural and simulated wing-flick signals.
Results. We have documented female wing-flick signals in all Magicicada species
except M. tredecula, which was not present in sufficient numbers to study in 1998. In all
species studied, a female signals in response to a calling conspecific male by moving her
wings in a single, quick motion which produces a broad-frequency sound of approximately
< 0.02 s duration. The motion and sound are similar to wing flutters produced in response
to disturbance, except that wing flutters consist of multiple wing movements. The signal’s
101
visual and acoustical components appear unspecialized but the timing of the signal in
relation to the male’s call is species-specific; the signal therefore consists of both the wingflick and the timing of its delivery. In M. septendecim, females produce the signal an
average of 0.387 ± 0.106 s after the end of the male calling phrase (Fig. 5.2). The delay in
M. cassini, 0.705 ± 0.112 s, is nearly twice as long (Fig. 5.3). These species’ 13-year
counterparts appear to have similar or identical signal timing. M. septendecula females
produce multiple wing-flicks in the brief silences between subphrases during the second
part of a male’s two-part call (Fig. 5.4).
In all species, a receptive female produces wing-flick signals in response to each of
the male’s calls as he approaches, engaging in a “rapprochement duet” with the calling male
(Fig. 5.1). This duet continues until the male switches to CII courtship calling. We have
not determined what stimuli cause the male to switch from the duet to CII calling; he
usually does so once within approximately 1-15 cm of the female, perhaps upon making
close visual contact. In M. -decim the female does not respond during the male’s CII song,
which is a continuous stream of shortened calling phrases without silent gaps; if the male
stops calling, the female responds with a wing-flick after the appropriate delay. M. -decim
males continue CII calling until reaching the female; once positioned next to the female, the
male switches to CIII courtship calling and attempts to copulate. The rapprochement duet
in M. -cassini is similar to that in M. -decim, except that females sometimes signal during
male CII calling. In M. septendecula there may not be a homologous CII courtship song;
after the rapprochement duet the male may proceed directly to CIII courtship and mounting.
Females do not signal during CIII in any Magicicada species studied. The context of the
signal, and male reactions to it, suggest its purpose: Sexually receptive females signal to
obtain mates.
102
The relationship of the female response to sexual receptivity in M.
septendecim
To demonstrate that immature females are not sexually receptive and do not signal,
in 1998, we six daily collections of approximately 25 teneral females each. Each day, we
played recordings of male song to each of the day-cohorts and watched for wing-flick
signals. To demonstrate that mated females do not wing-flick, in the 1997 study, we
allowed 22 individually-marked (Chapter 2) female M. septendecim to mate once and
divided them between two cages along with 22 marked, unmated females of the same age.
In each of four consecutive days we played recorded M. septendecim calling songs to the
females for two minutes and recorded the number of mated and unmated females
responding.
Results. Females first signaled 5-8 days after emerging (Table 5.4). This period
of nonresponsiveness is consistent with Magicicada “teneral periods” reported elsewhere
(Maier 1982; see also Karban 1981, Young and Josephson 1983) and with similar periods
in nine species of phaneropterine katydids (Spooner 1968). Variable weather during
maturation probably causes the inexactness of the relationship between age and maturity.
Since sclerotization involves chemical processes such as protein cross-linking, it is affected
less by time than by temperature and humidity (Chapman 1971 pp. 443-446). In the
experiment comparing mated and unmated females, none of the mated females responded to
playbacks, while at least half of the unmated females responded each day with wing-flick
signals (Table 5.5). This difference was significant in each of the four days (P ≤ 0.001;
Fisher’s Exact Two-Tailed Test).
Response of M. septendecim females to playbacks of heterospecific song
Magicicada aggregations contain as many as four different species (Alexander and
Moore 1962; Chapter 6), yet interspecific hybrids are rarely observed (Alexander and
103
Moore 1962; Dybas and Lloyd 1962, White 1973). To demonstrate the species-specific
nature of the wing-flick response, in 1996 we played recorded M. septendecim calling
phrases alternating with M. cassini calling phrases to 25 caged, unmated M. septendecim
females and noted their responses. We also included three unmated M. cassini females as
controls; we could not collect enough to include more. We tested the females in groups of
five, playing a series of 15 alternating M. septendecim and M. cassini calls (30 calls total)
and recording female responses.
Results. M. septendecim females routinely responded to the M. septendecim call
phrases in each trial. Only one of 25 M. septendecim females ever responded to a
heterospecific playback (Table 5.6). The three M. cassini females in the experiment
responded to conspecific calls and never responded to heterospecific playbacks.
Male responses to female signals
Effects of the signal on male behavior in M. septendecim
In 1996 and 1997, to document the effects of the female signal on male behavior,
we produced sounds and movements imitating female wing-flick signals for chorusing
male M. septendecim. Before we identified a suitable device for producing simulated
wing-flicks, we used a strip of paper that we flicked with our fingers. Later, we
discovered that toggling an ordinary household electric light switch was more convenient
for the experimenter. Because male responses to both artificial stimuli appear
indistinguishable, we combined trials with both methods in our statistical analyses.
A single M. -decim call phrase consists of a 1.5-4.0 s buzz (the “main element”) of
constant and nearly pure pitch followed by a brief (ca. 0.35 s) “terminal downslur” ending
about 500 Hz lower than the main element pitch. To document the importance of correct
timing in eliciting male M. septendecim courtship, we produced simulated wing-flicks in
response to the calls of males that had landed and begun calling on nearby vegetation,
104
timing our signals either (1) during the main element, (2) during the downslur, or (3) after
the downslur. We placed the flicking device within 25 cm of the male along the branch on
which he had just landed. We included control trials in which the experimenter approached
in the same fashion but did not produce simulated wing-flicks. We scored a male as
responding positively if he moved toward the stimulus and began late-stage courtship
behaviors such as CII or CIII calling, foreleg-vibrate, or mounting behavior. Each trial
ended when the male flew or walked away from the stimulus, or when the male remained
motionless longer than 20 seconds.
The above experiment simulated a scenario in which the male alights near a
signaling female. To examine male responses in a scenario involving weaker and/or
inconsistent female responses, we conducted trials in which we presented individual
chorusing males with either a single nearby or repeated distant simulated wing-flicks. We
produced single, nearby simulated wing-flicks in response to the first call following a flight
at a distance of approximately 25 cm from the male. For multiple distant simulated wingflicks, we produced the signals after each of the male’s calls from a distance of 1.3 m. We
again included control trials in which the experimenter approached in the same manner but
did not produce simulated wing-flick signals. In each trial we recorded the number of calls
the male made in his current bout, the nature of his next action (sit, walk, fly), and the
direction and distance of movement. We stopped monitoring males that paused for longer
than 20 seconds.
Results. Only wing-flicks produced after the downslur caused males to respond
positively. Males usually responded to such stimuli by walking toward the stimulus while
calling. In this behavior, termed “call-walking,” males stopped walking for ca. 1 second
immediately following each downslur. This pattern is distinct from chorusing behavior
which involves bouts of calling alternated with flights or silent walks. Males were equally
105
unresponsive to the control treatment and to simulated wing-flicks produced during the
main element or during the downslur (Table 5.7).
Males responded to single nearby and multiple distant simulated wing-flick signals
in a manner suggesting an attempt to localize the stimulus (Table 5.8). Both kinds of
stimuli caused males to increase the number of calls in the current calling bout compared to
control males; most males then flew to a new calling perch instead of call-walking toward
the stimulus. Whether walking or flying after the calling bout, males presented with the
single nearby or multiple distant simulated wing-flicks were more likely to move in the
direction of the stimulus than control males were. In control trials, males were more likely
to move away from the stimulus than toward it, suggesting that the presence of the
experimenter probably disturbs the cicada.
Components of the female signal
Because female wing-flick signals are often audible, we assumed early in our
investigation that males locate receptive females primarily by listening for wing-flick
signals. However, the importance of investigating separately the visual and acoustical
components of the female signal became apparent when we substituted an audiotaped
recording of a female wing-flick for the mechanically produced simulated flicks of our
earlier experiments. Male M. -decim did not give consistent or strong responses to these
playbacks, suggesting that they must perceive the visual component of the female signal
(actual signals and signals simulated with paper strips and electrical switches involve a
synchronized sound and quick movement, while the speaker produces only a sound).
Spectral analysis of wing-flick sounds indicates that they are broad-frequency sounds with
most of the energy above 4 kHz. It is possible that the different Magicicada species
perceive different aspects of female signals; although such high-frequency sounds are
within the range of maximal hearing sensitivity of M. -cassini, female wing-flicks may be
above the range of maximal hearing sensitivity for male M. –decim (Simmons et al. 1971;
106
Huber et al. 1980). We conducted experiments addressing the effects on male M. –cassini
and M. –decim of visual signals, acoustical signals, and signals containing both acoustical
and visual information on male chorusing behavior and male courtship behavior.
To examine the effects of timed flick sounds alone on male chorusing behavior, we
constructed a clicking device by attaching a 12 volt relay to the end of a 1-meter wooden
pole, covering all wires and dark (cicada-colored) parts of the relay with white masking
tape. In test trials, we identified an actively chorusing male M. septendecim or M. cassini,
placed the relay within 15 cm of him along the same branch, and clicked the relay in time to
his calls, following him as he moved and taking care not to make any timed movements
observable by the male. For controls, we placed the device in the same manner, but did not
click the relay. For each male, we measured the number of calls in each of two calling
bouts and the distance moved between the two calling bouts.
To investigate the effects of timed visual signals alone and the significance of other
visual cues such as the presence of a cicada-colored object, we approached chorusing male
M. septendecim with a model consisting of a black ballpoint pen cap or another model
made from an identical cap painted white. Without making sounds, we moved the model
once rapidly back and forth about 2 cm at the appropriate time for a female signal, or we
held it still. The black pen cap was similar in size and color to a real cicada, while the white
cap lacked only the appropriate color stimulus.
To determine whether combined visual and acoustical stimuli are more potent than
either alone, we used the following experimental design: From the surrounding chorus, we
selected 48 male M. septendecim and 48 male M. cassini one at a time and placed each in a
22x24x22 cm test chamber. Two walls of the chamber consisted of three layers of dark,
opaque cloth; the remaining walls consisted of fiberglass window screen. In one treatment,
we suspended a motionless model cicada inside the test chamber and responded to the
male’s calls with clicks produced behind the opaque chamber sides. In the second
107
treatment, the experimenter held the model inside the chamber and responded to the calling
male by moving the model slightly with the appropriate timing, without making clicks. In
the third treatment, the experimenter responded by producing clicks and by moving the
model. After the start of each trial, we recorded the male’s behaviors for the next two
minutes. Male M. septendecim and M. cassini were scored as responding positively to the
model if they exhibited any of the following behaviors during the 5 minute trial: CII or CIII
call, extrude genitalia, climb onto the model, foreleg vibrate, or attempt to copulate.
Results. Timed click sounds did not affect male M. septendecim behavior, but
male M. cassini given click sounds flew significantly shorter distances between calls (Table
5.9). Male M. septendecim courted plastic pen caps that moved in time with their calls, but
were less likely to engage in late-stage courtship with white colored caps than with black
colored caps (Fig. 5.5), indicating that visual stimuli alone are sufficient to provoke male
responses and that the stimulus includes the presence of an appropriately colored cicadalike object. The model that moved and clicked simultaneously was most attractive to M.
septendecim and M. cassini (Fig. 5.6). These results suggest that even though sounds or
movements may provoke courtship under certain circumstances, in both M. septendecim
and M. cassini, synchronized sounds and movements are most likely to be perceived as
female signals.
Behavioral bisexuality of males infected by the fungal pathogen
Massospora cicadina
Magicicada are sometimes infected by the fungus Massospora cicadina Peck
(Speare 1921, Soper 1974, Soper et al. 1976, Lloyd et al. 1982, White and Lloyd 1983).
Stage I (conidial phase) of this fungus is asexual; individual Magicicada so infected lose
their abdominal terminalia, exposing a chalky mass of conidiospores. Infected cicadas tend
to walk around on vegetation, spreading the spores by rubbing the exposed mass on
surfaces (Alexander et al. ms.). Individuals contacting these infective conidiospores in the
108
same emergence develop a Stage II (resting spore) infection. These individuals also lose
their abdominal terminalia, but they tend to make long flights, during which the friable
resting spores spill from the abdomen (Alexander et al. ms.). These resting spores are
believed to infect cicada nymphs of the next generation (Soper 1974, Soper et al. 1976).
In 1996, while observing free-flying Magicicada, we observed a Stage I
Massospora-infected male M. septendecim respond to a courting male twice with wingflick signals (Fig. 5.7). The courting male responded as expected by attempting to copulate
with the infected, signaling male, who rebuffed the courter by flapping his wings. These
observations were striking because, in hundreds of hours spent to that point observing
Magicicada in natural and controlled circumstances, we had never observed normal,
uninfected males making wing-flick signals in response to other males’ calls. The
observation of signaling by infected males led us to use playback experiments to compare
the infected males’ responsiveness to that of unmated adult females.
We placed M. septendecim four at a time in a cage and played 15 calling song
phrases to each group of cicadas. Trials began with a playback of male calling song at
normal pitch (1.39 kHz); then we first increased, then decreased the pitch in steps, and
trials ended with a playback of the calling song at normal pitch. We scored cicadas as
giving a positive response to a calling phrase if they responded with wing-flick signals to a
minimum of two phrases. We played normal and frequency-altered M. septendecim calling
songs to 8 wild-caught uninfected males, 6 Stage I Massospora-infected males, and 32
unmated females. We later tested the responsiveness of 5 Stage II Massospora-infected
male M. septendecim to playbacks of unaltered M. septendecim calling song.
Results. Stage I fungus-infected M. septendecim responded to playbacks with
appropriately timed wing-flicks, while no normal male or Stage II infected male ever
responded. Stage I infected males were generally less responsive than mature females
(Table 5.10). We have also observed Stage I infected M. tredecassini responding with
109
wing-flick signals to other males’ calls (Fig. 5.8). We expect that fungus-induced
bisexuality will be found in all Magicicada species.
Components of male calls
Alexander (1975) noted that “failure to find evidence of nonsinging males searching
through cicada choruses implies that the male’s song remains an essential part of his ability
to acquire females.” The discovery of the female sexual signal and its timing in relation to
the male’s call confirms this prediction and suggests a more detailed functional hypothesis
regarding the structural elements of M. –decim and M. -cassini calls, which we have
developed and tested primarily in M. septendecim.
The call phrases of M. -decim males are 1.5-4.0 seconds in length, ending with a
frequency downslur (Fig. 5.2). The nearly pure-tone main elements of M. -decim calling
songs overlap to produce a uniform chorus drone; as male density and activity increases, it
becomes difficult to perceive individual calls of individual males. In such situations the
downslur is the only portion of the male’s call that stands out from the background chorus.
Because females must be able to respond with a signal timed from the end of a male’s call,
we suspected that the downslur might function to enhance the end of the male’s call and
improve a calling male’s odds of being perceived by a nearby receptive female. Because
males have never been observed producing downslurs only, producing a “main element”
must also improve a male’s likelihood of provoking a female response. This hypothesis
leads to several predictions. First, if the main element alone is ever sufficient to elicit a
signal, its effectiveness should decrease as the background chorus increases in intensity.
Second, downslurs alone should not be sufficient to elicit wing-flick signals, unless the
background chorus is loud enough to sufficiently stimulate the female. Finally, neither the
downslur nor the main element alone should be as effective as the intact call in eliciting the
female response at any background level.
110
To determine which components of male M. septendecim calls most effectively
elicit a female response under different chorus conditions, we constructed a pure-tone
artificial calling phrase, a pure-tone phrase lacking a downslur, a pure-tone downslur
alone, and a simulated pure-tone background chorus. In previous experiments, we found
that females respond similarly to playbacks of recorded and pure-tone artificial calls. We
noted the responses of individually marked females in 19 groups of females during 146
playbacks of complete calls, calls lacking slurs, and slurs only at an intensity of 72-79 db
(intensity varied within the test chamber) over a background chorus of four different
intensities (0 db, 58-62 db, 63-77 db, 65-80 db).
Results. Although call fragments did elicit female responses under certain
conditions, females were more likely to respond to whole calls than to partial calls at all
background chorus intensities (Table 5.11). As the background chorus intensity increased,
female responsiveness to whole calls and main elements of whole calls declined, while
females became more responsive to slurs (Table 5.12; Fig. 5.9), such that at the highest
intensity, females were more responsive to slurs alone than to main elements alone.
Inter-male acoustical interference behavior
In 1996-1998 we observed a previously undescribed male sound in M. -decim and
M. -cassini species; this sound is composed of a short (ca. 0.25 s) “buzz” with a frequency
spectrum similar to that of the main element of the calling song but somewhat less puretoned. This sound is always produced during the downslur of another male’s call (Fig.
5.10). In part because the downslur is important in eliciting female wing-flick signals, we
suspected that a male engaged in close-range courtship uses this sound to obscure a newlyarrived competitor’s downslurs, thereby decreasing the likelihood that the courted female
will reveal her presence with a wing-flick signal before the competitor completes his short
calling bout and departs.
111
Alternatively, the sounds could be aborted calls caused by interruption or when the
stimuli inducing the call are insufficient to stimulate a complete call phrase. This
hypothesis does not predict that the buzzes should be consistent in length or timing in
relation to another male’s call, nor does it predict that the buzz should be produced under
limited circumstances. Another explanation for this behavior is that males might use this
sound to signal their sex to other males; males often court other males, so the sound may
discourage misdirected sexual attention. A similar explanation has been proposed for a
“flick-tick” signal produced by M. cassini males (Dunning et al. 1979). This hypothesis
predicts that the signal should be observed most often when males are crowded and
encounter one another commonly in the chorus.
We conducted three experiments to evaluate these hypotheses in M. septendecim.
In the first, designed to determine if crowded chorus conditions alone can induce buzzes,
we placed 30 chorusing males together in a 1x1x1 m cage, watched them interact, and
listened for the male buzzes. After 20 minutes of watching, we produced simulated wingflick signals in response to males on one side of the cage for one minute and then observed
the males for another five minutes. We repeated this experiment twice.
In the second experiment, intended to simulate the events involved in the
appearance of a potential interloper during courtship, we presented 22 individual males
with the following series of stimuli: First, we confined each male in a small screen test
chamber. We then played one to five minutes of recorded calling song at an intensity
designed to simulate a male calling at a distance of approximately 10 cm. Males often
began to call during this treatment, but if the playbacks were not sufficient to stimulate
calling, we produced simulated wing-flick signals in response to the playbacks using an
electrical switch held outside the chamber within view of the male; signals from the switch
always induced call-walking from the male. Once the male had begun calling, we stopped
the playbacks and responded to the male’s calls for 10-60 seconds, after which we ceased
112
responding. Once the male stopped his calling or courtship songs (some males had begun
CII or CIII calling during this duet), we resumed playbacks of calling song from the
speaker at the same volume and distance. We noted the context of any interference buzzes
produced by the male during the trial.
In the third test, we examined the effects of artificial pure-tone buzzes on female
responsiveness by creating a model call and buzz such that the call could be played with or
without the buzz. We confined four unmated female M. septendecim in the test chamber
and played a sequence of 30 call pairs; each pair consisted of (1) a normal call played from
one speaker followed by (2) a call played from the same speaker with a buzz played from a
second speaker and superimposed over the slur. We recorded the number of females
responding to each call and repeated the experiment 6 times, using different females. We
compared the number of females responding positively to each call pair using a Friedman
Two Way analysis of variance.
Results. In trial 1 of the first test, no males produced buzzes prior to the
production of simulated wing-flick signals, but at least 2 buzzes were heard once we began
to respond to the males with simulated wing-flick signals. In trial 2, one buzz was heard
during the first 20 minutes, but over 20 were heard once we began responding to the
males. Males called often and frequently landed within centimeters of each other during the
trials, and although some males harassed or attempted to court other males, these
interactions did not lead to male buzzes.
In the second test, males never produced buzzes in response to the initial series of
playbacks, never produced buzzes during artificial duets between the playbacks and the
experimenter, and never produced buzzes while duetting with the experimenter. However,
in 16 of the 22 trials, males began producing buzzes once the playbacks had resumed
following the termination of the male’s duet, usually in response to the first or second
playback call, but sometimes not until 3 or 4 calls had played. Males producing buzzes did
113
so only during the downslur of the recorded calling song phrases. In five of the trials, the
experimenter again began producing simulated wing-flick signals to the playback while the
male buzzed; in four of these five cases this caused the male to cease buzzing and begin
call-walking near the simulated wing-flick stimulus. In the fifth case the male walked while
buzzing after each playback call.
In the third test, the presence of artificial buzzes significantly decreased the
likelihood that a female would respond with wing-flick signals to an artificial calling song
(Table 5.13). Under the conditions of these experiments, buzzes halved the likelihood of a
female response.
Discussion
Wing-flick signals in Cicadidae
Communicative wing-flicking (sometimes called wing-tapping, -banging, clapping, -clacking, or -clicking) appears to be widespread in cicadas. We use the term
“wing-flick” for Magicicada because it connotes movement and sound, both of which are
perceived by males; other terms emphasize only the acoustic component of the signal. Male
wing-flicking during close-range courtship interactions with females has been reported in
Australian and New Zealand Kikihia, and Amphipsalta, (Dugdale and Fleming 1969, Lane
1995) North American Okanagana (Davis 1919, Alexander 1957b, Chapter 8), and
European Tibicina (Fonseca 1991), while males combine wing-flicks with long-range
calling song in Asian Cicadetta (Popov 1981), Australian and New Zealand Amphipsalta
(Dugdale and Fleming 1969, Lane 1995), and Western North American Platypediinae
(Moore 1968). Female wing flick signaling is known in North American Magicicada and
Okanagana, (Davis 1919, Chapter 8), Australian Cystosoma (Doolan 1981), Cicadetta
(Gwynne 1987), and Amphipsalta (Dugdale and Fleming 1969), and New Zealand
Amphipsalta, Kikihia, Maoricicada, Notopsalta, and Rhodopsalta (Lane 1995, Dugdale
114
and Fleming 1969), with the most detailed published reports of female wing-flick signaling
from Kikihia spp. (Lane 1995), Amphipsalta cingulata (Lane 1995), Cystosoma saundersii
(Doolan 1981), and Cicadetta quadricinctata (Gwynne 1987). In each species studied in
detail, female wing-flick signals elicit male courtship behavior and appear to have a specific
temporal relationship to the male’s song. In a few cases female wing-flick signals, while
present, are apparently not always prerequisites for mating: Although sometimes both
sexes of Okanagana canadensis and O. rimosa appear to use wing-flicks to signal their
presence, females most often signal receptivity simply by approaching stationary calling
males (Chapter 8).
Wing-flick signals and the Magicicada courtship sequence
In Magicicada, female wing-flick signals in response to male calling song cause
males to cease chorusing and begin courtship. A female Magicicada wing-flick signal
consists of a quick, timed movement of the wings that makes a broad-frequency sound
varying from a faint rustle to a loud snap; the signal’s timing in relation to the male’s call is
species-specific. A chorusing male perceiving a single wing-flick signal or repeated,
distant wing-flick signals increases his number of calls in the current bout and then resumes
chorusing behavior while moving in the direction of the stimulus, apparently to elicit
further responses and obtain better information about the female’s location. If the male
receives repeated nearby responses that allow him to localize the female, he ceases
chorusing behavior (calls alternated with flights) and call-walks toward the responding
female; the male may use short flights to cross gaps in the vegetation, but such short,
irregular flights are distinct from chorusing flights. In M. -decim and M. -cassini species,
the male may begin CII calling once within approximately 15 cm. of the female. Female
M. -decim do not wing-flick during the CII call, which contains no silent gaps; female M. cassini occasionally respond during males’ CII calling. After contacting the female or
while preparing to mount, males of all species begin CIII calling, which they continue until
115
the genitalia are engaged. Receptive females of all species do not wing-flick during CIII
and remain still throughout the entire courtship sequence. In nearly every mating that we
have observed, the cicadas have performed, in order, each of the behaviors of this
courtship sequence; exceptions include rare forced copulations that occur when females are
trapped by vegetation or in aggregations of males.
Males respond to both the acoustical and visual components of the signal. We do
not know to what extent male Magicicada can localize sound alone without a visual
component, but the wing-flick sounds alone do not elicit later courtship stages as rapidly or
consistently as do visual cues. The acoustical and visual stimuli sufficient to cause male
responses in all Magicicada species must be quite general, because the click and movement
of an ordinary electric light switch are highly attractive to courting males. Since male
cicadas directed the full repertoire of normal sexual behaviors towards active, signaling
models and ignored these same models when no signals were perceptible, male courtship
songs and the female wing-flick signals appear to be sufficient for sexual rapprochement in
these species.
Male-male acoustic interference competition
The discovery of the rapprochement duet in Magicicada improves our
understanding of male-male competitive interactions in choruses. Most significant is the
finding that male M. -decim and -cassini appear to use “interference buzzes” to acoustically
jam rival males’ signals in competition for mates. Interference buzzes are produced
specifically in the context of a close-range male-female duet or a prolonged courtship that
has been interrupted by the arrival of a calling, and potentially interloping, male competitor.
The courting male emits short buzzes coincident with the downslurs of his rival’s calls;
these buzzes obscure the downslur of the rival’s call and reduce the likelihood that the
female will perceive and respond to them. If the rival male does not elicit responses from a
female, he is likely to continue chorusing and depart. One potential objection to this
116
hypothesis is that the buzz itself could reveal the presence of the nearby receptive female.
For the buzz to function to reduce interloping, calling males either must be unable to
discern the buzz from the background chorus or they must be unable to hear sufficiently
while calling; otherwise, males should have evolved to respond to interference buzzes by
intensifying their local search. Male M. septendecim and M. cassini hearing sensitivity is
reduced 5-15 dB during calling by the action of muscles that reduce tension on the tympana
(Pringle 1954, Simmons et al. 1971); thus calling males may have difficulty perceiving that
their signals are being jammed by buzzing males. Also, buzzing males terminate the signal
precisely at the end of the interloper’s song, suggesting that buzzing behavior has been
selected to be produced only when it is undetectable by the potential interloper.
Other insects are known to make use of specialized male-male competitive signals.
For example, some male katydids produce sounds that appear to mimic female responses to
their own calling songs, apparently to confuse potential interlopers (Alexander 1975,
Grove 1959). A similar function could be served by male wing-clicking during calling in
cicadas such as Amphipsalta cingulata, in which males wing-click to their own calling
songs with a timing identical to that of female wing-flick responses (Lane 1995). The
Magicicada interference buzz appears uniquely complex, however, in that the signal
apparently deceives the courted female in order to achieve a competitive advantage over a
rival male.
The intense nature of Magicicada male mating competition likely results in few
females remaining in the mating pool long after the onset of sexual receptivity; the wingflick signal is so potent, and searching males so abundant, that newly-receptive females
may be detected almost immediately by one of the many chorusing males, in which case
wing-flick rapprochement duets may be brief. In addition, unless a teneral female’s
readiness to mate is activated all at once, fully formed, a newly adult female may be
expected to pass through a period of partial receptivity in which (1) she produces the wing-
117
flick signal very weakly and/or inconsistently and (2) she is more likely to reject males as a
result of fluctuations in her mating readiness. If so, males likely have evolved to respond
to very weak or intermittent signs of timed wing-flicking in females. Such a period of
partial or weak sexual receptivity could be involved in observations of lengthy courtships
involving repeated rejections before eventual copulation (Alexander 1968, Dunning et al.
1979) and the apparent coyness of Magicicada females. Subtlety of wing-flick responses
in newly-matured females, in addition to rapid detection of signaling females by chorusing
males, could explain why the Magicicada wing-flick signal has for so long remained
undiscovered.
Acoustic synchrony in M. –cassini .
The ability of M. –cassini males to synchronize their calls may be an adaptation
allowing individual males to maximize their probability of perceiving female responses. All
available evidence (Simmons et al. 1971; Huber et al. 1980; this study) suggests that male
M. –cassini can detect the acoustical component of female wing flick signals. Unlike M.
–decim calls, M. –cassini calls are broad frequency, and overlap the frequency range of
female wing flick signals (Fig. 5.3). We propose that male M. -cassini have evolved to
delay beginning calling phrases until a nearby conspecific begins, so that each male
maximizes his chances of hearing nearby female wing flicks without acoustical
interference. The net effect of an entire chorus pursuing this strategy is for males to
synchronize calling and flying; thus, acoustic detection of female signals may be an
explanation for the well-documented phenomenon of synchronized chorusing behavior in
M. –cassini (Alexander and Moore 1958).
Wing-flick signals and the parasitic fungus Massospora cicadina
Alexander et al. (in prep.) show that the fungus Massospora cicadina alters
Magicicada behavior in ways that likely increase spore transmission, and that the alterations
118
in behavior are different for the two fungal life history stages in ways that are appropriate
for the stage-specific modes of transmission. Our observations suggest that the first stage
of the fungus has apparently evolved to increase the odds of spore transmission between
males by causing infected males to produce wing-flick signals in response to the calls of
other males even while retaining normal male courtship behaviors. Thus, while females are
in jeopardy from sexual contact primarily with infected males, males are in jeopardy from
sexual contact with infected individuals of both sexes. Second stage Massospora-infected
males do not produce wing-flick signals, consistent with the theory that spores of this stage
do not infect cicadas of the same generation (Soper 1974, Soper et al. 1976). We do not
know the physiological mechanism or timing of infection by Stage I fungal spores, but the
behavioral bisexuality of infected males strongly supports the argument that spore
transmission between individuals of the same generation is crucial for the Stage I fungus.
The evolution of Magicicada chorusing behavior
Alexander (1975) noted that, in Magicicada, it is apparently the sound of the entire
chorus that attracts females, rather than the song of an individual male. Therefore, an
individual male Magicicada with chorusing behavior that is less effective in long-range
attraction but more likely to be detected by stationary receptive females would realize a
fitness advantage relative to those males with chorusing behaviors optimal for long-range
attraction. For Magicicada, the discovery that receptive females respond to the calls of
individual males with timed wing-flicks resolves the question, raised by Alexander (1975),
of why males do not silently parasitize the mate-attracting abilities of others by searching
without calling: Because male calls provoke female responses, males who search without
signaling sacrifice the opportunity to detect females by provoking responses.
Magicicada chorusing behavior lies at one extreme of the range of rapprochement
behavior found in cicadas. Males alternate unusually brief bouts of calling (and unusually
short call phrases, especially in M. -decim and M. -cassini) with short flights, a
119
rapprochement strategy that is centered around searching for responding females who have
moved to the chorus and become stationary. Only one other well-studied cicada species,
the Australian Tick-tock cicada (Cicadetta quadricinctata) has a similar rapprochement
system, although calling bouts in this species are still three to four times longer in duration
than those of Magicicada (Gwynne 1987, Alexander and Moore 1962). Gwynne (1987)
noted the similarities of male behavior in Magicicada and Cicadetta quadricincta and all but
predicted that female wing-flick signals would be found in Magicicada. At the other
extreme are species in which males produce a continuous song (with or without separate
phrases) from a single location for relatively long periods, such as in North American
Okanagana canadensis and O. rimosa (Chapter 8). This strategy emphasizes stationary
male advertisement over searching, and in such species females approach males for mating.
A similar range of rapprochement strategies is found in other singing insects such as
Phaneropterine katydids (Spooner 1968, Heller and von Helversen 1993). The Magicicada
rapprochement system is similar to those of fireflies (see Lloyd 1966, 1979) and substratevibrating leafhoppers (Hunt and Nault 1991, Hunt et al. 1992, Claridge 1985) and
lacewings (Wells and Henry 1992, Henry 1994).
Understanding the causes of taxonomic variation in rapprochement requires an
understanding of the relative costs and benefits of alternative signaling and searching
behaviors for males and females in each species under consideration (Alexander et al.
1997). Males, especially in species such as insects with little or no paternal investment, are
expected to be more risk-tolerant than females in rapprochement since their reproductive
success is most directly influenced by the number of mates obtained (Trivers 1972,
Alexander and Borgia 1979, Parker 1979) and hence by the effectiveness of mate-locating
activities. Because of the sexes’ unequal risk tolerance, male mate-locating activities are
influenced primarily by patterns of female dispersion (Emlen and Oring 1977), while
females are influenced by selection for effective parental investment. The relative costs and
benefits of alternative rapprochement behaviors for individuals of each sex will be affected
120
by interrelated factors including the kinds and pattern of predation, the spatial distribution
of appropriate habitats and resources (Emlen and Oring 1977, Cocroft and Pogue 1996,
Heller and von Helversen 1993), the sex-specific costs and benefits of searching vs.
remaining stationary, and the current strategies of the opposite sex.
Periodical cicada rapprochement behavior involves a remarkable combination of
related unusual features which have likely resulted from adaptation to high population
density. Because the phylogenetic relationship of Magicicada to other cicada genera is
poorly known, little information is available on the rapprochement behavior of the most
recent non-periodical ancestor of Magicicada. However, the uniqueness of the Magicicada
rapprochement system suggests that Magicicada must have evolved from ancestors with
relatively stationary, advertising males and relatively mobile females. The following
general hypothesis explains the unique features of Magicicada; similar selective pressures
have likely shaped related attributes of rapprochement systems of other organisms (e.g.
Heller and von Helversen 1993).
Upon the evolution of partial or complete periodicity (see Lloyd and Dybas 1966a,
b; Cox and Carlton 1991, Long 1993, Yoshimura 1997) high population density arising
from lack of ecological control by predators and parasites became a consistent feature of
Magicicada ecology. This presumably led to two important effects with dramatic
consequences for the Magicicada mating system: First, males and females began to
experience reduced risks associated with movement and/or signaling. Second, the greater
density of receptive females increased the chances that a male could encounter a receptive
female while moving through his environment. These factors together would have
improved the potential payoff to individual males of increasing time spent searching at the
expense of advertisement; increased male searching would have caused females to benefit
most from moving to areas of loud male chorus sound and remaining still once there, in
part because increased movement by males would reduce the ability of females to approach
121
individual males before their next flights. This strategy of moving to the chorus and
becoming stationary could have resulted in selection favoring a tendency in males to be
attracted to the choruses of other males, completing the evolution of the basic elements of
the Magicicada rapprochement system.
It is probably true that for any individual of a species in which females approach
calling males, unreceptive females can avoid calling males, and, by voluntarily approaching
calling males, sexually receptive females are able to mate on demand. For species in which
males locate females, however, females can reduce their rapprochement time by making it
easier for males to find them. Although in dense Magicicada emergences, it would seem as
though males would have little difficulty finding females, any individual female is likely
surrounded by a collection of immature cicadas, cicadas of the same sex, and mated
cicadas, as well as a small number of sexually receptive members of the opposite sex.
Female Magicicada may have evolved to signal their receptivity to ensure timely mating in
confusing, dense aggregations. Males would then have been selected to become efficient at
detecting these signals and minimizing their chances of mistakenly courting unreceptive
cicadas or other inert objects.
It is not difficult to imagine the advent of stationary females resulting in the
evolution or refinement of a female wing-flick signal to reduce rapprochement time,
although the female signal may have existed from the start in the nonperiodical ancestor.
An environment of locally aggregated, stationary females who respond to the call terminus
with a wing-flick would select males further for a local search strategy emphasizing short
calling bouts, short calling phrases, and short flights. The density of male aggregations
would allow the chorus sound itself to assume the female-attraction function of the male’s
call, allowing males to modify their calling and searching behaviors to be most effective for
local mate searches, possibly resulting in (1) quieter calls, (2) more pure-tone calls in the
122
M. –decim species, and (3) calls with terminal downslurs in the M. –decim and M. cassini
species or their common ancestor.
The best tests of the above scenario will likely come from comparative study using
data from pair-forming systems in other Cicadidae, but at present few species have been
studied in significant detail. One pattern within Magicicada offers initial support for the
theory: Two Magicicada species, M. septendecula and M. tredecula, are consistently less
abundant than other Magicicada; these species have longer calling bouts and a longer call
unit that contains multiple temporal “windows” for female wing-flick responses. Our
playback experiments confirm a different aspect of the hypothesis, that male calling song
has evolved under selection for distinctiveness in loud choruses. The two-part structure of
M. –decim and M. –cassini calling songs may provide a record of the evolution of song
distinctiveness from an ancestral call without a downslur: The downslur functions at least
in part to mark the end of a call against a background chorus dominated my main elements.
123
Literature Cited
Alexander, R. D. 1957a. The taxonomy of the field crickets of the eastern United States
(Orthoptera: Gryllidae: Acheta). Ann. Ent. Soc. Am. 50: 584-602.
Alexander, R. D. 1957b. Sound production and associated behavior in insects. Ohio
Journal of Science 57: 101-113.
Alexander, R. D. 1968. Arthropods. Pp. 167-216 in: T. A. Sebeok (ed). Animal
Communication. Techniques of Study and Results of Research. (Indiana University
Press: Bloomington, IN).
Alexander, R. D. 1975. Natural selection and specialized chorusing behavior in chorusing
insects. Pp. 35-77 in: D. Pimentel, ed. Insects, Science, and Society. (Academic
Press: New York).
Alexander, R. D., and T. E. Moore. 1958. Studies on the acoustical behavior of seventeenyear cicadas (Homoptera: Cicadidae: Magicicada). Ohio J. Science. 58: 107-127.
Alexander, R. D., and T. E. Moore. 1962. The evolutionary relationships of 17-year and
13-year cicadas, and three new species. (Homoptera: Cicadidae, Magicicada).
Misc. Publ. no. 121. Univ. of Michigan Museum of Zoology, Ann Arbor, MI.
Alexander, R. D., and G. Borgia. 1979. On the origin and basis of the male-female
phenomenon. Pp. 417-440 in: M. S. Blum and N. A. Blum, eds. Sexual selection
and reproductive competition in insects. (Academic Press: New York).
Alexander, R. D., D. C. Marshall, and J. R. Cooley. 1997. Evolutionary Perspectives on
Insect Mating. Pp. 1-31 in: B. Crespi, J. Choe, eds. The evolution of mating
systems in insects and arachnids. (Cambridge University Press).
Alexander, R. D., J. R. Cooley, and D. C. Marshall. (ms). A specialized fungal parasite
(Massospora cicadina) modifies the behavior of periodical cicadas (Homoptera:
Cicadidae: Magicicada).
Bigelow, R. S. 1960. Interspecific hybrids and speciation in the genus Acheta (Orthoptera:
Gryllidae). Can. J. Zool. 38: 509-524.
Bradbury, J. W. 1981. The evolution of leks. Pp. 138-169 in: R. D. Alexander and D.
W. Tinkle, eds.. Natural selection and social behavior. (Chiron Press: New York).
Bradbury, J. W. 1985. Contrasts between insects and vertebrates in the evolution of male
display, female choice, and lek mating. Fortschr. Zool. 31: 273-289.
Butlin, R. K., and M. G. Ritchie. 1994. Behaviour and speciation. In: P. J. B. Slater, and
T. R. Halliday, eds. Behaviour and Evolution. (Cambridge University Press).
Chapman, R. F. 1971. The Insects: Structure and Function, 2nd Edition. (Elsevier: New
York).
124
Claridge, M. F. 1985. Acoustic signals in the Homoptera: Behavior, Taxonomy, and
Evolution. Ann. Rev. Entomol. 30: 297-317.
Cocroft, R. B., and M. Pogue. 1996. Social behavior and communication in the
neotropical cicada Fidicina mannifera (Fabricius) (Homoptera: Cicadidae). J.
Kansas Entomol. Soc. 69(4) suppl.: 85-97.
Cox, R. T., and C. E. Carlton. 1991. Evidence of genetic dominance of the 13-year life
cycle in periodical cicadas (Homoptera: Cicadidae: Magicicada spp.). Am. Midl.
Nat. 125: 63-74.
Davis, W.T. 1919. Cicadas of the genera Okanagana, Tibicinoides, and Okanagodes, with
descriptions of several new species. J. N. Y. Entomol. Soc. 27: 179-223.
Doolan, J. M. 1981. Male spacing and the influence of female courtship behavior in the
bladder cicada, Cystosoma saundersii Westwood. Behav. Ecol. Sociobiol. 9:269276.
Doolan, J. M., and D. Young 1989. Relative importance of song parameters during flight
phonotaxis and courtship in the bladder cicada Cystosoma saundersii. J. Exp. Biol.
141 1989: 113-131.
Dugdale, J. S., and C. A. Fleming. 1969. Two New Zealand cicadas collected on Cook’s
Endeavour voyage, with a description of a new genus. New Zealand J. Sci. 12:
929-957.
Dunning, D. C., J. A. Byers, and C. D. Zanger. 1979. Courtship in two species of
periodical cicadas, Magicicada septendecim and Magicicada cassini. Anim. Behav.
27: 1073-1090.
Dybas, H. S., and M. Lloyd. 1962. Isolation by habitat in two synchronized species of
periodical cicadas (Homoptera: Cicadidae: Magicicada). Ecology 43: 444-459.
Emlen, S. T., and L. W. Oring. 1977. Ecology, sexual selection, and the evolution of
mating systems. Science (Wash. D. C.) 197: 215-223.
Fonesca, P. J. 1991. Characteristics of the acoustic signals in nine species of cicadas
(Homptera: Cicadidae). Bioacoustics 3: 173-182.
Fulton, B. B. 1933. Inheritance of song in hybrids of two subspecies of Nemobius
fasciatus (Orthoptera). Ann. Ent. Soc. Amer. 26: 368-376.
Grove, D. G. 1959. The natural history of the angular-winged katydid Microcentrum
rhombifolium PhD. Dissertation, Cornell University.
Gwynne, D. T. 1987. Sex-biased predation and the risky mate-locating behaviour of male
tick-tock cicadas (Homoptera: Cicadidae). Anim. Behav. 35: 571-576.
Heller, K., and D. von Helversen. 1993. Calling behavior in bushcrickets of the genus
Poecilimon with differing communication systems (Orthoptera: Tettigonioidea,
Phaneropteridae). J. Insect Behav. 6: 361-377.
Henry, C. S. 1985. Sibling species, call differences, and speciation in green lacewings
(Neuroptera: Chrysopidae: Chrysoperla ). Evolution 39: 965-984.
125
Henry, C. S. 1994. Singing and cryptic speciation in insects. Trends Ecol. and Evol.
9:388-392.
Huber, F., D. W. Wohlers, and T. E. Moore. 1980. Auditory nerve and interneurone
responses to natural sounds in several species of cicadas. Physiol. Entomol. 5: 2545.
Hunt, R. E., and L. R. Nault. 1991. Roles of interplant movement, acoustic
communication, and phonotaxix in mate-location behavior of the leafhopper
Graminella nigrifrons. Behav. Ecol. Sociobiol. 28: 315-320.
Hunt, R. E., J. P. Fox, and K. P. Haynes. 1992. Behavioral response of Graminella
nigrifrons (Homptera: Cicadellidae) to experimentally manipulated vibrational
signals. J. Ins. Behav. 5: 1-13.
Karban, R. 1981. Effects of local density on fecundity and mating speed for periodical
cicadas. Oecologia 51: 260-264.
Lane, D. H. 1995. The recognition concept of species applied in an analysis of putative
hybridization in New Zealand cicadas of the genus Kikihia (Insecta: Hemiptera:
Tibicinidae). Ch. 17 in: D. M. Lambert and H. G. Spencer, eds., Speciation and
the Recognition Concept: Theory and Application. (John Hopkins).
Lloyd, J. E. 1966. Studies on the flash communication system in Photinus fireflies. Misc.
Publ. no. 130. Univ. of Michigan Museum of Zoology, Ann Arbor, MI.
Lloyd, J. E. 1979. Sexual selection in luminescent beetles. Pp. 293-342 in M. S. Blum
and N. A. Blum, eds. Sexual selection and reproductive competition in insects.
(Academic Press: New York).
Lloyd, M., and H. S. Dybas. 1966a. The periodical cicada problem. I. Population
ecology. Evolution 20: 133-149.
Lloyd, M., and H. S. Dybas. 1966b. The periodical cicada problem. II. Evolution.
Evolution 20: 466-505.
Lloyd, M., J. White, and N. Stanton. 1982. Dispersal of fungus-infected periodical
cicadas to new habitat. Environmental Entomology 11:852-858.
Long, C. A. 1993. Evolution in periodical cicadas: A genetical explanation. Evol. Theory
10(4): 209-220.
Maier, C. T. 1982. Obervations on the seventeen-year periodical cicada, Magicicada
septendecim (Hemiptera: Homptera: Cicadidae). Ann. Ent. Soc. Amer. 75: 14-23.
Moore, T. E. 1968. The evolution of cicadas and their acoustical behavior. Thirteenth
Internat. Congr. Entomol. Abstracts of Papers: 173 (Moscow).
Otte, D. 1974. Effects and functions in the evolution of signaling systems. Ann. Rev.
Ecol. Syst. 5: 385-417.
Otte, D. 1992. Evolution of cricket songs. J. Orthop. Res. 1: 25-49.
126
Otte, D. 1994. The crickets of Hawaii: Origin, systematics and evolution (The
Orthopterists’ Society: Philadelphia, PA).
Otte, D., and R. D. Alexander. 1983. The Australian crickets (Orthoptera: Gryllidae).
Acad. Nat. Sci. Phil. Monogr. 22: 1-477.
Olvido, A. E., and T. A. Mousseau. 1995. Effect of rearing environment on calling-song
plasticity in the striped ground cricket. Evolution 49: 1271-1277.
Parker, G. A., 1979. Sexual selection and sexual conflict. Pp. 123-167 in: M. S. Blum
and N. A. Blum, eds. Sexual selection and reproductive competition in insects.
(Academic Press: New York).
Popov, A. V. 1981. Sound production and hearing in the cicada Cicadetta sinuatipennis
Osh. (Homptera: Cicadidae). J. comp. Physiol. 142: 271-280.
Pringle, J. W. S. 1954. A physiological analysis of cicada song. J. Exp. Biol. 34: 525560.
Ritchie, M. G. 1996. The shape of female mating preferences. Proc. Nat. Acad. Sci. USA
93: 14628-14631.
Shaw, K. L. 1996. Polygenic inheritance of a behavioral phenotype—interspecific genetics
of song in the Hawaiian cricket genus Laupala. Evolution 50: 256-266.
Simmons, J. A., E. G. Wever, and P. M. Pylka. 1971. Periodical cicadas: Sound
production and hearing. Science (Wash. D. C. ) 171: 212-213.
Soper, R. 1974. The genus Massospora entomopathogenic for cicadas. Part I. Taxonomy
of the genus. Mycotaxon 1: 13-40.
Soper, R., A. J. Delyzer, and L. F. R. Smith. 1976. The genus Massospora
entomopathogenic for cicadas. Part II. Biology of Massospora levispora and its
host Okanagana rimosa, with notes on Massospora cicadina on the periodical
cicadas. Ann. Ent. Soc. Amer. 69: 89-95.
Speare, A. T. 1921. Massospora cicadina Peck, a fungous parasite of the periodical cicada.
Mycologia 13: 72-82.
Spooner, J. D. 1968. Pair-forming acoustic systems of Phaneropterine katydids
(Orthoptera: Tettigoniidae). Anim. Behav. 16: 197-212.
Thornhill, R.A., and J. Alcock, 1983. The evolution of insect mating systems. (Harvard
University Press: Cambridge, MA).
Trivers, R. L. 1972. Parental investment and sexual selection. Pp. 136-179 in: B.
Campbell, ed. Sexual selection and the descent of man, 1871-1971. (Aldine:
Chicago).
Walker, T. J. 1962. The taxonomy and calling songs of United States tree crickets
(Orthoptera, Gryllidae, Oecanthinae). I. The genus Neoxabea and the niveus and
varicornis groups of the genus Oecanthus. Ann. Ent. Soc. Amer. 55: 303-322.
127
Walker, T. J. 1963. The taxonomy and calling songs of United States tree crickets
(Orthoptera, Gryllidae, Oecanthinae). II. The nigricornis group of the genus
Oecanthus. Ann. Ent. Soc. Amer. 56: 772-789.
Walker, T. J. 1964. Cryptic species among sound-producing ensiferan Orthoptera
(Grylidae and Tettigoniidae). Q. Rev. Biol. 39:345-355.
Wells, M. M., and C. S. Henry. 1992. The role of courtship songs in reproductive
isolation among populations of green lacewings of the genus Chrysoperla
(Neuroptera: Chrysopidae). Evolution 46: 31-42.
White J., 1973. Viable hybrid young from crossmated periodical cicadas. Ecology 54:
573-580.
White, J., and M. Lloyd. 1983. A pathogenic fungus, Massospora cicadina Peck
(Entomophthorales), in emerging nymphs of periodical cicadas (Homoptera:
Cicadidae). Environmental Entomology 12: 1245-1252.
Williams, K. S., and C. Simon. 1995. The ecology, behavior, and evolution of periodical
cicadas. Annu. Rev. Entomol. 40: 269-295.
Young, D. and R. K. Josephson. 1983. Pure-tone songs in cicadas with special reference
to the genus Magicicada. Journal of Comparative Physiology 152: 197-207.
Yoshimura, J. 1997. The evolutionary origins of periodical cicadas during ice ages. Am.
Nat. 149: 112-124.
128
129
17
17
13
17
13
M. septendecim (L.)
M. cassini (Fisher)
M. tredecassini Alexander and Moore
M. septendecula Alexander and Moore
M. tredecula Alexander and Moore
black with orange lateral band
> 3.00
> 3.00
> 3.00
black, rarely with weak ††
orange lateral band
black with orange lateral band
> 3.00
1.25 - 1.50
orange with black lateral band
or center
black, rarely with weak ††
orange lateral band
1.00 - 1.25
1.35 - 1.90
black
black
black
black
orange
orange
orange
Dominant call Pronotal
pitch (kHz)
extension
color
mostly orange
orange with black lateral band
or center
Abdominal sternites
7 - 14 †
7 - 14 †
2 - 6 §§
2 - 6 §§
1.5 - 4 §
1.5 - 4 §
1.5 - 4 §
upland
upland
midwestern,
southern
lowland
lowland
upland
mixed
upland
Habitat
eastern, plains
midwestern,
southern
eastern, plains
eastern, plains
midwestern,
southern
midwestern,
Ozark
Length of call Range
(seconds)
midday
midday
afternoon
afternoon
morning
morning
morning
Peak chorus
activity
§§
Roughly pure-tone, musical buzz terminating in a noticeable drop in pitch; no ticks. Usually 2-3 calls between flights.
Rapid series of ticks followed by high-pitched, broad-spectrum buzz that rises and then falls in intensity and pitch. Usually 1-2 calls between flights.
†
Repeated, rhythmic high-pitched, broad-spectrum tick-buzz phrases, followed by repeated phrases containing only ticks. Usually 1 call between flights.
††
Orange band, if present, often interrupted medially.
§
13
13
Life Cycle
(yrs.)
M. tredecim (Walsh and Riley)
M. neotredecim Marshall and Cooley
Species
Table 5.1. Traits distinguishing Magicicada species. "Pronotal extensions" are lateral extensions of the pronotum between the eyes and wing bases.
For additional descriptions see Alexander and Moore 1962.
130
Yes
Yes
1 phrase
between
flights
1 calling bout
between
flights
M. -cassini
M. -decula
Yes
2-3 phrases
between
flights
Yes
No
phrases similar to
those used in calling
and separated by
silent gaps
Yes
walking,
flying
between
phrases
walking,
flying
between
phrases
walking,
flying
between
phrases
Female
Wing
Flick
Behaviors
Yes
phrases similar to
those used in calling
and separated by
silent gaps
Courtship
CI Calling
phrases similar to
Under some those used in calling
conditions and separated by
silent gaps
No
Female
Wing
Flick
Synchrony
M. -decim
Species
Chorus
Calling
Table 5.2. The sexual sequence in Magicicada species.
none known
-
repeated
staccato
buzzes
repeated, shortened phrases of the same
type as in court I, with greater emphasis on
the slur, or terminal portion, without
Yes
intervening silences and with additional ticks
after the downslur
repeated
none known staccato
buzzes
walking
between
phrases
repeated
staccato
buzzes
CIII Calling
repeated, shortened phrases of the same
walking
type as in court I, with greater emphasis onAt end
between
the slur, or terminal portion, and without
of CII
phrases
intervening silence
CII Calling
Female
Wing
Flick
Behaviors
No
No
No
Foreleg
vibrate,
mount,
copulate
Foreleg
vibrate,
mount,
copulate
Foreleg
vibrate,
mount,
copulate
Female
Wing
Flick
Behaviors
Table 5.3. Study Sites, 1995- 1998.
Year
1995
1996
1997
1998
Brood
I
II
III
XIX
Life Cycle Location
17
Alum Springs
17
Horsepen Lake SWF
17
Siloam Springs SP
13
Harold Alexander WMA
County
Rockbridge
Buckingham
Brown, Adams
Sharp
131
State
VA
VA
IL
AR
Characteristics
Logged site
Logged site
Clearing
Cleared field
Table 5.4. Length of teneral period, in days, as measured by first female M.
septendecim responses to playback of male songs. Females were divided into groups
containing approximately 25 cicadas and placed in single-sex cages on living vegetation.
Females were exposed daily to playbacks of male song, and the day on which at least one
female in the cage responded with wing flick signals was recorded.
Group
A
B
C
D
E
F
Average
First response (days after emergence)
6
7
8
7
6
5
6.5
132
Table 5.5. Mated and unmated female M. septendecim responses to 2- minute playbacks
of male calls. Females responding positively produced at least one wing flick signal in
response to playbacks of male songs.
Day
Mating
Status
Mated
Unmated
n
22
17
Response
(+)
(-)
0
22
9
8
2
Mated
Unmated
22
16
0
8
22
8
≤ 0.001
3
Mated
Unmated
20
15
0
7
20
8
≤ 0.001
4
Mated
Unmated
18
14
0
7
18
7
≤ 0.001
1
P
(Fisher Exact Test)
≤ 0.001
133
Table 5.6. Wing flick (WF) responses of 25 female M. septendecim and 3 female M.
cassini to a series of 135 alternating male M. septendecim and M. cassini calls.
Species
M. septendecim
M. cassini
n
25
3
Number responding
t o M. septendecim call
t o M . c a s s i n i call
25
1
0
3
134
Table 5.7. Male M. septendecim responses to artificial wing flicks. We simulated
female wing flick signals by flicking a strip of paper or by clicking a heavy-duty light
switch at the appropriate time in relation to male call. We scored males as responding
positively to the artificial wing flicking if they moved toward the stimulus and began latestage courtship behaviors such as CII or CIII call, forleg-vibrate, or mounting behavior.
Fisher’s Exact 2- tailed tests are used to compare treatments to 46 controls in which the
clicking device was presented to the male, but no click was made; only 6 males responded
positively to such controls.
Control
n
46
(+)
6
(-)
40
vs.
Timing
n
(+)
(-)
P
End of call
During Call
During Slur
83
41
13
66
3
0
17
38
13
≤ 0.001
≤ 0.498
≤ 0.326
135
136
Multiple distant simulated wing-flick signals
Control
Call Number
1.94 ± 0.83 (n= 17)
Likelihood of flight
Fly after signal
13
Do not fly after signal 4
Distance of flight (cm)
35.76 ±26.55 (n= 13)
Direction
Toward Observer
0
Away from Observer 7
Single simulated wing-flick signal
Control
Call Number
2.28± 1.16 (n = 43)
Likelihood of flight
Fly after signal
36
Do not fly after signal 7
Distance of flight (cm)
25.75 ±19.4 (n = 40)
Direction
Toward Observer
3
Away from Observer 16
P
≤ 0.005 (Wilcoxon)
≤ 0.221 (Wilcoxon)
≤ 0.02 (Fisher’s Exact Test)
With signal
3.53 ± 1.26 (n= 19)
14
5
20.98 ±15.14 (n= 15)
4
2
≤ 0.003 (Fisher’s Exact Test)
≤ 0.223 (Fisher’s Exact Test)
≤ 0.355 (Wilcoxon)
38
15
22.60 ± 29.8 (n = 47)
13
7
P
≤ 0.001 (Wilcoxon)
With signal
3.70± 2.46 (n = 53)
Table 5.8. Male M. septendecim movement direction (flights and walks) following artificial wing flick signal or control (no signal)
treatment. Direction of movement was recorded as a value from 1-12 with the numbers representing the directions on a clock face,
with only the directions 11, 12, and 1 considered “toward stimulus” and 4, 5, and 6 “away from stimulus,” to avoid biased interpretation of ambiguous lateral movements. Males that paused for longer than 20 seconds were not monitored further.
Table 5.9. Experiments on male chorusing behavior. Clicking device placed near
chorusing male and clicked or not clicked (control). Number of calls in first and second
call bouts and length of first and second flights after treatment or control were measured.
Results analyzed with Kruskal-Wallis one-way analysis of variance.
M. septendecim
Click vs. control
First call bout (64/46)
First flight
(50/37)
Second call bout (45/34)
Second flight (39/32)
P
≤ 0.277
≤ 0.573
≤ 0.712
≤ 0.117
M. cassini
Click vs. control
First call bout (24/26)
First flight
(23/25)
Second call bout (18/22)
Second flight (15/19)
P
≤ 0.001
≤ 0.000
≤ 0.399
≤ 0.001
137
Table 5.10. Numbers of M. septendecim responding with wing flick signals to
frequency-modified playbacks of male song. Mature females made most reponses to
unmodified playback songs, as did Stage I Massospora cicadina infected males. Normal
males and Stage II infected males never responded to playbacks.
Song
Frequency (kHz)
2.35
1.91
1.65
1.48
Mature
Female (32)
0
2
13
16
Stage I
Male (6)
0
0
1
2
Stage II
Male (5)
-
Normal
Male (8)
0
0
0
0
1.39 (unmodified) 22
3
0
0
1.30
1.13
0.956
0.869
1
0
0
0
-
0
0
0
0
20
18
8
1
138
Table 5.11. Contrasts of female wing flick responses to playbacks of complete artificial
calling songs and partial artificial calling songs (main portion only, slur portion only) at
four background chorus intensities. P- values are for Fisher’s Exact 2- tailed tests. Whole
artificial calling song is always more effective in eliciting a female response than either
partial call. Background is a continuous 1.39 kHz pure tone; background levels: 0db, 5862 db, 63- 77 db, 65- 80 db, where maximum value is approximately equal to acoustic
intensity of playback call.
Contrast
Whole vs. Slur
Whole vs. Main
Main vs. Slur
Background
0 db
0.001*
0.001*
0.001*
58- 62 db
0.001*
0.001*
0.004*
139
63- 77 db
0.001*
0.001*
0.465
65- 80 db
0.005*
0.001*
0.012*
Table 5.12. Effects of background chorus intensity on female wing flick responses to
playbacks of complete artificial calling songs and partial artificial calling songs (main
portion only, slur portion only) at four background chorus intensities. Increasing
background chorus intensity significantly decreased the likelihood that whole calls and the
main portions of calls would elicit female responses and generally increased the likelihood
that slur alone would elicit responses. Background is a continuous 1.39 kHz pure tone;
background levels: 0db, 58- 62 db, 63- 77 db, 65- 80 db, where maximum value is
approximately equal to acoustic intensity of playback call.
Effect of increasing chorus intensity, Kruskal-Wallis one way analysis of variance:
Playback type
Whole
Main portion only
Slur portion only
Z
420
64
44
P
≤ 0.001
≤ 0.001
≤ 0.001
140
Table 5.13. Effects of M. -decim “interference buzz.” We confined four, unmated
female M. septendecim in a test chamber and played a sequence of 60 calls, alternating
normal calls and calls with buzzes, recording the number of females responding to each
call. We repeated the experiment 6 times.
Replicate
A
Buzz
Yes
No
Average number of
Females responding
to each call
1.13 ± 0.86
2.23 ± 0.77
B
Yes
No
0.57 ± 0.57
2.43 ± 0.68
≤ 0.00
C
Yes
No
0.80 ± 1.00
1.73 ± 1.08
≤ 0.001
D
Yes
No
1.27 ± 1.93
0.82 ± 0.58
≤ 0.006
E
Yes
No
0.93 ± 0.87
1.53 ± 0.78
≤ 0.045
F
Yes
No
0.47 ± 0.57
2.07 ± 0.98
≤ 0.00
141
P (Friedman
Two-Way
analysis of variance.)
≤ 0.001
142
male lands
and calls
wf
male
calls
male CI
wf
female
answers
male approaches
female
male CII
wf
male approaches
female, foreleg
vibrates, mounts,
copulates
male CIII
female
answers
Figure 5.1. Stylized sonogram of male call/female wing flick coutship duet for M. -decim. Male begins with cout I (CI) calls;
female answers each call with a wing flick. After several such interactions, male begins court II (CII) calling, which consists
of repeated phrases of the same general type as CI calling, but shortened and without intervening silence. After the male
ceases CII calling, the female wing flicks in response, and the male begins cour III (C III) calling, which consists of repeated
staccato buzzes. M. -cassini and M. -decula courtship sequences are similar, except that M. -decula lack clearly defined CII
calls.
male
"sing-fly"
behavior
male CI
female
answers
143
0
S
0
0.5
1.0
1.5
*
2.0
2.5
Figure 5.2. Sonogram of male M. -decim call and female wing flick response. Female response
(marked with asterisk) is a broad-frequency sound. Female produces signal 0.387 s after end of
male call (n= 235 examples, ± 0.110s). Wing flick sound enhanced and extraneous background
noise removed for clarity.
kHz
2
4
6
8
10
144
S
0
0
1
2
3
*
4
Figure 5.3. Sonogram of male M. -cassini call and female wing flick response. Female response
(marked with asterisk) is a broad-frequency sound. Female produces signal 0.705 s after end of
male call (n= 16 examples, ± 0.112s). Wing flick sound enhanced and background chorus removed
from 2 to 4 seconds for clarity
kHz
2
4
6
8
10
145
0
S
0
0.2
0.4
0.6
0.8
1.0
*
1.2
1.4
1.6
*
1.8
Figure 5.4. Sonogram of male M. -decula call and female wing flick responses. Female responses
(marked with asterisks) are broad-frequency sounds. Female produces signals during silent portions
of male call. Wing flick sound enhanced and extraneous background noise removed for clarity.
kHz
2
4
6
8
10
146
0.001
0.001
0.129
0.233
White/move (21) vs. white/still (20)
Black/move (22) vs. white/move (21)
Black/still (21) vs. white /still (20)
Call-Walk towards
Black/move (22) vs. black/still (21)
(male CI)
Artificial
signal
Artificial
signal
1.00
0.045
0.233
0.004
CII
1.00
0.345
1.00
0.108
CIII
Artificial
signal
1.00
0.001
1.00
0.001
Foreleg vobrate
Figure 5.5. Comparison of male responses to a model cicada constructed from a pen cap. Cap was either
black or white, and experimenter either moved at the appropriate time for a female wing flick signal, or held
it still. Male responses, including call-walk towards, CII call, CIII call, and Foreleg vibrate were noted.
Fisher's Exact 2- tailed Test P-values for pairwise comparisons of treatements plaed below a stylized
sonogram of courtship sequence in M. -decim.
147
Move
Move and Click
M. septendecim
Click
Move
M. cassini
Click
Move and Click
Figure 5.6. Male responses to actions of model. Males were scored as responding positively if they produced Court II or
Court III courtship songs, or if they attempted to mount and copulate with the model. Positive responses marked with
shading, negative responses unshaded. Within M. septendecim, the model that moved and clicked was more effective than
the model that clicked only (P ≤ 0.001, Fisher's Exact 2- tailed Test) or the model that moved only (P ≤ 0.001, Fisher's Exact
2- tailed Test). For M. cassini, the results were similar; the model that moved and clicked was more effective than the model
that clicked only (P ≤ 0.001, Fisher's Exact 2- tailed Test) or the model that moved only (P ≤ 0.001, Fisher's Exact 2- tailed
Test). Responses to click only and move only treatements did not differ in wither species.
0
2
4
6
8
10
12
14
16
18
148
0
0.5
1.0
1.5
*
2.0
Figure 5.7. Sonogram of response of M. -decim male with Stage I Massospora cicadina infection to an
M. -decim call. Male response (marked with asterisk) is a broad-frequency sound similar in timing and
acoustical properties to female wing flick signals. Wing flick sound enhanced and extraneous background
noise removed for clarity.
S
kHz 0
2
4
6
149
0
0
0.5
1.0
1.5
*
2.0
Figure 5.8. Sonogram of response of M. -cassini male with Stage I Massospora cicadina infection to an artificial
M. -cassini call. Male response (marked with asterisk) is a broad-frequency sound similar in timing and
acoustical properties to female wing flick signals. Wing flick sound enhanced and extraneous background
noise removed for clarity.
S
kHz
5
10
150
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0
63- 77
Background intensit (db)
58- 62
65- 80
Slur portion only
Main portion only
Whole call
Figure 5.9. Responses of female M. septendecim to playbacks of artificial, pure tone calls and portions of calls against an
artificial background chorus. Females were scored as responding positively if they produced wing flick signals in response
to playback. No actual chorus of M. septendecim was audible from the test chamber. Intensity of background chorus increased
in increments from 0 db to 80 db. Data presented as number of positive responses for a given call and background condition
given the total number of trials with those experimental conditions. Whole calls (square markers) were always more likely to
elicit responses than either main portion only (circles) or slur portion only (triangles). As background intensity increased, the
effectiveness of main portion only decreased, while the effectiveness of slurs alone increased.
Proportion responding
151
kHz
0
1
male
call
2
3
interference
buzz
4
5
Figure 5.10. Sonogram of male M. -decim call interference buzz of nearby male. Note how interference buzz
overlaps and acoustically obscures calling male's terminal downslur.
S
0
1
2
3
4
CHAPTER VI
REPRODUCTIVE CHARACTER DISPLACEMENT AND SPECIATION IN
PERIODICAL CICADAS, AND A NEW 13- YEAR SPECIES,
MAGICICADA NEOTREDECIM
Abstract
Acoustic mate-attracting signals of related sympatric, synchronic species differ, but
those of related allopatric species sometimes do not, suggesting that these signals may
evolve to “reinforce” premating species isolation when similar species become sympatric.
This model also predicts divergences restricted to regions of sympatry in overlapping
species, but such “reproductive character displacement” has rarely been confirmed. We
report such a case in the acoustic signals of a previously undetected, new 13-year periodical
cicada species, Magicicada neotredecim, here described. Where M. neotredecim overlaps
M. tredecim in the central U.S., the dominant chorus pitch (frequency) of M. neotredecim
increases from ca. 1.40 to 1.75 kHz, while that of M. tredecim slightly decreases from ca.
1.15 to 1.08 kHz. Female cicadas from sympatry (both species) and allopatry (only M.
neotredecim studied) respond best to song pitches similar to those of conspecific males.
M. neotredecim differs from 13-year M. tredecim in abdomen coloration, mtDNA, and
song pitch, but does not differ consistently from 17-year M. septendecim; thus, like other
152
Magicicada species, M. neotredecim appears most closely related to a geographicallyadjacent counterpart with the alternative life cycle. Speciation in Magicicada may often be
initiated by life cycle changes that create temporal isolation, and reinforcement could play a
novel role by fostering divergence in premating signals prior to speciation. The geographic
and phylogenetic relationships of the M. -decim species, and the pattern of reproductive
character displacement, suggest two theories of Magicicada life cycle evolution: “nurse
brood facilitation” and “canalization of climate-induced shifts.”
Introduction
The periodical cicada species (Magicicada spp.) of eastern North America have
extraordinary life histories and phylogenetic relationships (Marlatt 1923; Alexander and
Moore 1962; Lloyd and Dybas 1966a, b; Simon 1988; Williams and Simon 1995).
Magicicada live underground juvenile lives of either 13 or 17 years, after which they
emerge for a brief adult life of about three weeks. Six Magicicada species have been
described: In northern and plains states, three morphologically and behaviorally distinct
species coexist and emerge together once every 17 years (Fig. 6.1). These species are
reproductively isolated in part by distinctive male acoustic signals and female responses
(Alexander and Moore 1958, 1962). In the Midwest and South, three similar 13-year
species coexist. In this paper, we describe a fourth 13-year species, M. neotredecim. The
13- and 17-year populations are divided regionally into several temporally isolated
“broods” that emerge on different schedules. Each 13-year species is probably most
closely related to its 17-year counterpart; some of these species pairs can be distinguished
only by life cycle (Table 6.1). This pattern suggests that speciation in Magicicada may
involve a combination of geographic isolation and life cycle changes that create temporal, or
allochronic, isolation (Alexander and Moore 1962; Lloyd and Dybas 1966b; Lloyd and
White 1976). Allochronic speciation has been proposed for other organisms (e.g.
153
Alexander 1968; Alexander and Bigelow 1960; Tauber and Tauber 1977a,b), although one
case has recently been challenged (Harrison 1979; Harrison and Bogdanowicz 1995).
The male calling songs of sympatric Magicicada species are strikingly different,
while those of parapatric life cycle siblings (e.g. 17-year M. cassini and 13-year M.
tredecassini) are similar or indistinguishable (Alexander and Moore 1962). This pattern,
common in groups with long-range sexual signals, suggests a process in which costly
heterospecific sexual interactions lead to selection reinforcing trait differences that promote
premating isolation (Dobzhansky 1940; Blair 1955). The hypothesis of reinforcement also
predicts greater signal divergence in sympatry, a pattern termed “reproductive character
displacement,” when two species’ distributions partially overlap (Brown and Wilson
1956), but this second pattern is only rarely observed (Alexander 1967; Walker 1974;
Howard 1993). For example, only one case is known in the singing Orthoptera and
Cicadidae (Otte 1989). Our use of these terms, in which reinforcement is a process and
reproductive character displacement a resulting pattern, is similar to that of Loftus-Hills and
Littlejohn (1992) and Howard (1993), except that in our usage reinforcement can involve
selection against wasteful heterospecific sexual interactions of any kind, not merely
heterospecific mating as in Howard (1993). Reinforcement of premating isolation is of
interest as a mechanism by which barriers to interspecific gene flow are enhanced as a
direct result of species interactions in sympatry rather than as incidental outcomes of
evolution in allopatry (Dobzhansky 1940; Blair 1955; Howard 1993).
Here we report that the newly described 13-year periodical cicada species,
Magicicada neotredecim, shows reproductive character displacement in male song pitch and
female song pitch preferences in the central U.S. where it overlaps the 13-year species M.
tredecim (Fig. 6.1). M. neotredecim appears to have been derived by a life cycle change
from its 17-year counterpart M. septendecim (see also Martin and Simon 1988, 1990;
Simon et al. submitted). We propose new hypotheses of life cycle evolution and speciation
154
in Magicicada suggested in part by the geographic and phylogenetic relationships of M.
neotredecim to the remaining Magicicada species.
Methods and Results
Background
The Magicicada mating system involves male acoustical signals, female responses,
and complex acoustical courtship (Chapter 5). Periodical cicada males aggregate, forming
singing choruses attractive to sexually receptive females. Most Magicicada choruses
contain three species, a M. –decim1 species that produces a narrow band of sound
frequencies with a single dominant pitch between 1 and 2 kHz, and M. -cassini and M.
–decula species that each produce broad-spectrum sounds above 3 kHz (Fig. 6.2). While
observing 13-year Magicicada in northern Arkansas in 1998, we found choruses with two
peak frequencies in the M. -decim range (ca. 1.1 and 1.7 kHz), suggesting a new M. decim species (Figs. 6.2, 6.3). The pitch differences, easily detectable by ear, are unlikely
to result from temperature variations because the song types coexist in the same choruses,
because chorus pitch did not vary with air temperature (Fig. 6.4), and because song pitch is
not related to tymbal muscle temperature in the closely-related M. septendecim (Young and
Josephson 1983). The nature and location of this discovery suggested that the two
sympatric M. -decim might correspond to two forms of M. tredecim identified using
mitochondrial DNA (mtDNA) and abdomen coloration (Martin and Simon 1988) and found
to overlap in a zone reaching from Arkansas to Indiana (Simon et al. submitted). We
conducted acoustical, behavioral, and morphological analyses in this region of overlap to
test this possibility and to determine the role of song pitch in reproductive isolation of the
1
For convenience we refer to Magicicada sibling species groups using the following shorthand: M-decim:
M. septendecim (17), M. tredecim (13), and M. neotredecim (13); M -cassini: M. cassini (17) and M.
tredecassini (13); M. -decula: M. septendecula (17) and M. tredecula (13).
155
two species. Analysis of geographic song pitch variation led to the discovery of
reproductive character displacement in the new species, M. neotredecim.
Documenting Sympatric M. –decim Species with Songs and Morphology
Martin and Simon (1988) identified two forms of 13-year M. tredecim partly on the
basis of the amount of orange coloration on the abdominal sternites. To determine if the
sympatric M. -decim song types corresponded to these morphs, we measured the calling
songs of a random sample of 151 males from a mixed chorus in Sharp Co., AR, and tested
for association of song pitch and abdomen color. We collected the males from privatelyowned woods on County Rd. 51 approximately 0.25 mi. S. of County Rd. 62 (Lat.
36.256, Lon. -91.462), at a powerline right-of-way, just outside the northwest boundary
of the Harold E. Alexander Wildlife Management Area, Sharp Co., AR; we will refer to
this location as the “powerline” site.
We recorded male songs using a Sony Professional Walkman cassette recorder and
a Sony microphone mounted in a plastic parabola, or a Sony 8mm videocassette recorder
with built-in microphone. Because the songs of an individual male do not vary
significantly in dominant pitch (Fig. 6.5), we isolated a single call of each individual for
spectral analysis. For each recording we generated a power spectrum (plot of sound
intensity vs. frequency) and spectrogram (power spectrum vs. time) using Canary 1.1.1
(Cornell Bioacoustics Laboratory) on a Macintosh computer. Individual male M. –decim
songs consist of a 1-3 second steady-pitch and nearly pure-tone (sine wave) “main
element” followed by a quieter 0.5 second frequency “downslur” that ends approximately
500 kHz lower than the main element pitch (Fig. 6.3; Alexander and Moore 1958; Weber et
al. 1987). Because the comparatively minor sound energy in the downslur is distributed
across many frequencies, it contributes little to the spectral profile of the call; thus the main
element produces the dominant call pitch.
156
We scored the abdomen color of each individually-recorded male using the method
of Martin and Simon (1988), in which each male is assigned a value from 1 (50% black) to
4 (all orange). Because male calling song pitches fell into two distinct categories with just
one intermediate male out of 151 sampled, while the morphological traits were more
continuously distributed, we tested for association between song pitch and coloration by
dividing the male sample through an arbitrary intermediate value of 1.5 kHz and examining
abdominal color differences between the resulting groups using a Mann-Whitney U.
Results. Within the Sharp Co., AR, powerline site, M. -decim males producing
songs with low dominant pitch (mean 1.11 kHz) had the orange abdomen color (Table 6.2;
Appendix F) characteristic of Martin and Simon’s (1988) “mtDNA lineage B,” now
recognized to be the previously described Magicicada tredecim (Alexander and Moore
1962). The M. -decim males producing higher-pitched songs (mean 1.71 kHz) had the
darker abdomen color of Martin and Simon’s “mtDNA lineage A,” and constitute a new
species here named Magicicada neotredecim. The mixed chorus produced little sound at
pitches intermediate to those of M. tredecim and M. neotredecim (Figs. 6.2, 6.3, 6.6); this
and the strong correlation of song and abdomen color suggest that interbreeding is rare (see
also Simon et al. submitted). Among approximately 200 cicadas observed during our
study, we found just three unambiguous intermediates: One high-pitch male with a yellow
abdomen (category 4), one low-pitch male with a darker abdomen (category 2), and one
male with an intermediate calling song (1.43 kHz). These could be hybrids or merely
phenotypic outliers.
Magicicada neotredecim, new species: Holotype (male): Collected by Cooley and
Marshall on 18 May 1998 in Sharp Co., AR., in a powerline right-of-way on County Rd.
51 approximately 0.25 mi. S. of County Rd. 62 (Lat. 36.256, Lon. -91.462), outside the
Harold E. Alexander Wildlife Management Area. Dominant pitch of male song phrase
1.68 kHz. Deposited at University of Michigan Museum of Zoology (UMMZ), Ann
157
Arbor, MI with a recording of the male’s call stored on diskette. Allotype: Collected by
Cooley and Marshall on 18 May 1998 at same location as holotype. Deposited at UMMZ.
For additional description see Table 6.1.
Measuring Female Song Pitch Preferences in Sympatry
Sexually receptive female Magicicada produce “wing-flick” signals in response to
male songs (Chapter 5, Cooley and Marshall in prep.). We used this signal as an assay of
female mating receptivity to determine if female preference for song pitch was correlated
with abdomen color. Using Sound Edit Pro software(MacroMedia), we produced 14 puretone model calling phrases differing only in pitch (1.0 kHz to 2.3 kHz, in 0.1 kHz
increments) and lacking temporal structure. We modeled artificial songs after field
recordings of M. –decim songs (see example using artificial frog calls in Gerhardt 1992).
In previous experiments, we have found that females respond similarly to playbacks of
recorded and artificial songs. We estimated song pitch preferences of 91 M. -decim
females collected from the Sharp Co., AR, powerline site by playing the model songs to
individually-marked caged females in both haphazard and ordered sequences using a
Macintosh Powerbook computer connected to an amplified portable speaker. Females that
did not respond to any songs were dropped from the analysis. We positioned the speaker
25 cm away from the test cage and played the model songs at a volume simulating the
intensity of a calling male, as determined by measuring several males’ call intensities with a
sound level meter. The playback experiments were carried out between 11:00 and 16:00 in
bright overcast or sunny conditions against an acoustic background of a Magicicada chorus
containing M. neotredecim, M. tredecim, M. tredecassini and M. tredecula located in
woods ca. 8 meters away. For each female, we recorded whether she responded with
wing-flick signals and then calculated the weighted average of her responses to the 14
model songs using the following formula:
158
14
∑ (kHzcall )(responses to call )
i
i
i =1
(total responses)
We scored female abdomen color using the method described above for males. We
tested for association between weighted average pitch preference and abdomen color by
dividing the female sample through the same pitch as the male sample (1.5 kHz) and
examining the difference in abdomen color between the resulting groups using a MannWhitney U.
Results. The playback experiments showed evidence of two classes of females
(Fig. 6.6, Appendix G). At the powerline site, females responding maximally to lower
pitch songs (M. tredecim) were significantly more orange-colored than females responding
maximally to higher pitch songs (M. neotredecim; Table 6.2). About a third (26/91) of the
females did not respond to any call. This response rate is similar to that observed in
previous work with 17-year M. septendecim females in Virginia and Illinois (Chapter 5;
Cooley and Marshall in prep.), where only one M. –decim species is known.
Because the model songs contained no within-phrase temporal structure, differed
only in pitch, and accurately distinguished the species, we conclude that song pitch
differences may be an important cause of species-specificity in 13-year M. –decim mate
recognition. However, natural calls have a pattern of amplitude-modulation that results
from individual tymbal pulses (Young and Josephson 1983; Weber et al. 1987); we have
not ruled out a role for this temporal structure. Differences in such temporal patterning are
probably not the cause of pitch differences in M. -decim cicadas because variations in
tymbal contraction rate do not influence dominant pitch (Young and Josephson 1983). We
did not analyze temporal call structure because most of our field recordings contained too
much background noise, although this did not affect analysis of dominant pitch. Little is
159
known of the relative roles of temporal patterning and frequency content in cicada song in
general, although both function in Australian bladder cicadas (Cystosoma; Doolan and
Young 1989). In the singing Orthoptera, species-specificity of calling songs usually
results from differences in temporal patterning rather than pitch (but see Toms and Otte
1988 for an example in which pitch may be important), while dominant call pitch is
important in many Anuran amphibians (Littlejohn 1977; Gerhardt 1988).
Using chorus recordings to estimate species abundance.
For mixed choruses, we developed a method for using the relative intensities of the
two species-specific M. –decim dominant chorus pitches to estimate relative proportions of
the species. This approach assumes that (1) M. neotredecim and M. tredecim are uniformly
distributed at a given site and (2) both species show the same relationship between male
abundance and chorus intensity. To test the first assumption, we recorded a continuous
chorus sample along a 200m woodside trail in the Harold E. Alexander Wildlife
Management Area, Sharp Co., AR, while pointing the parabola/ microphone assembly into
the treetops along one side at a 45° angle. We recorded one side while walking in one
direction and then recorded the other side while returning. From portions of this recording
taken at seven meter intervals, we generated power spectra and measured the intensities of
the M. neotredecim and M. tredecim frequency bands; this yielded 47 samples because of
gaps in the forest on one side (350m total). If the M. neotredecim and M. tredecim at the
site were not uniformly distributed with respect to one another on a local scale, we
predicted that this series of samples would show significant variation among locations in
the relative intensities of the two species’ chorus bands. To test the second assumption, we
compared the distribution of call pitches of a random sample of males collected from the
mixed-species Sharp Co., AR chorus to the distribution of song frequencies in the chorus
power spectrum, using a Kolmogorov-Smirnov test to determine if the shapes of the two
distributions differed significantly.
160
Results. Although the proportions of the two M. –decim species vary on a scale
of miles (e.g. Fig. 6.7 insets), the species do not appear to assort significantly within a
location. In the 350m continuous recording, the proportion of M. -decim chorus sound
produced by the rarer species (M. neotredecim) remained between 10% and 36% (mean =
19.0%, SD = 6.0, n = 47), and the chorus intensities of the two species were not
significantly negatively correlated (Pearson coefficient = -0.229, P= 0.121) as would be
expected if the species formed exclusive aggregations. Therefore a recording taken at a
given position is likely to give a reasonable estimate of the local chorus sound. This result
matches our subjective impressions; we never detected by ear variations in the M. -decim
chorus sound within a location. If males of the two species sing with similar consistency,
intensity, and call duration, the sound intensity at a given species-specific pitch should be
related to the local density of males producing that pitch. The random sample of the Sharp
Co., AR powerline population verified this relationship: The standardized histogram of call
pitches of individually-recorded males was indistinguishable from the standardized
quadratic chorus power spectrum (Kolmogorov-Smirnov test, P > 0.05.; Fig. 6.6).
Therefore the power spectrum of a mixed chorus can be used to estimate the relative
proportions of M. tredecim and M. neotredecim present.
Although the downslurs of the higher-pitched M. neotredecim songs sometimes
overlap the higher-frequency portions of the lower-pitched M. tredecim songs, downslurs
from even a dense M. neotredecim chorus do not obscure sparse populations of M.
tredecim because M. -decim downslurs are brief and their sound energy is distributed
across a broad band of frequencies. M. tredecim at the Sharp Co., AR, powerline site
produced a distinct peak in the chorus power spectrum (Fig. 6.6) despite constituting just
8% of the study population. However, extremely rare M. tredecim could be overlooked in
a dense M. neotredecim population; we attempted to minimize this possibility by recording
several minutes at each location and by sweeping the parabola in different directions. In
contrast, rare M. neotredecim are always easily detectable in a M. tredecim chorus because
161
M. tredecim males do not produce sounds at frequencies as high as those of M.
neotredecim.
Estimating Species Distributions using Geographic Variation in Songs
Once we had demonstrated that the two 13-year M. -decim forms are species
differing in song pitch and that the species’ relative abundance can be estimated from
chorus recordings, we analyzed the species’ distributions and geographic variation in song
pitch using recordings taken from 80 locations distributed throughout the 1998 Magicicada
emergence. Periodical cicadas in different regions emerge in different years, and the
cicadas of a region are referred to as a “brood” (Marlatt 1923; see individual brood maps in
Simon 1988). The 13-year cicadas emerging in 1998 belonged to the largest 13-year
brood, Brood XIX, which reaches from Maryland to Oklahoma. The widespread range of
this brood allowed a thorough characterization of 13-year M. -decim calling song
biogeography. The recording dates were as follows: Sharp, Fulton and Lawrence Co.,
AR, 12-25 May; remaining Arkansas sites, 28 May; Missouri sites 1-7 June; Randolph,
Monroe, Jersey, Sangamon and Piatt Co., IL 29-30 May; remaining Illinois sites 9-14
June; remaining central and southeastern state sites 31 May - 3 June; Maryland sites 29
May - 1 June.
Results. We found M. neotredecim in Missouri, Illinois, western Kentucky, and
northern Arkansas (Fig. 6.7, Appendix H; see also Simon et al. submitted). The
southernmost M. neotredecim populations overlap M. tredecim in a zone 50-150 km wide
reaching from northern Arkansas into southern Missouri, southern Illinois, and eastern
Kentucky. The remainder of Brood XIX contains M. tredecim and not M. neotredecim.
Simon et al. (submitted) document a similar pattern of overlap using abdomen coloration
and mtDNA and describe additional mixed populations in southern Indiana.
162
Geographic variation in dominant chorus pitch of M. neotredecim shows a pattern
of reproductive character displacement, a predicted outcome of reinforcement of premating
isolation (Dobzhansky 1940; Blair 1955; Howard 1993). M. neotredecim males have the
highest song pitch (ca. 1.7 kHz) where they coexist with M. tredecim. North of the
overlap zone, M. neotredecim dominant song pitch decreases to approximately 1.4 kHz in
Illinois and 1.5 kHz in Missouri (Fig. 6.8, 6.9). Most of the song change occurs over a
short distance immediately north of the zone of sympatry, and little variation occurs
elsewhere in allopatry. Only song pitch varies across the range of M. neotredecim;
abdomen color and mtDNA haplotype are comparatively uniform (Martin and Simon 1988,
1990).
Song pitch variation in M. tredecim is more subtle, only about 1/4th of that
observed in M. neotredecim. M. tredecim populations in deep sympatry with M.
neotredecim have very low-pitched songs, and M. tredecim peak song pitch slightly
increases south and east of the overlap zone from ca. 1.11 to 1.15 kHz (Figs. 6.9, 6.10).
However, the significance of this pattern is not clear because some allopatric M. tredecim
choruses in the southeast have songs as low-pitched as those in the overlap zone.
While most of the chorus samples included the songs of thousands of males, many
of the populations from Missouri and Alabama were recorded late in the emergence when
relatively few males remained. For these locations the chances of overlooking a rare type
were greater, although it seems unlikely that rare M. tredecim could have been missed in
each of the many samples from central and northern Missouri.
Additional Tests of Reproductive Character Displacement
To determine if female song pitch preferences change with male song pitch in M.
neotredecim, we measured weighted average song pitch preferences of 33 Magicicada
neotredecim females collected from a woodlot off Rt. 1300E 0.8 miles south of White
163
Heath, IL (Piatt Co.), beyond the northern limit of M. tredecim. We completed the
playback experiments at Lodge Park Co. Forest Preserve with a nearby background chorus
containing M. neotredecim, M. tredecassini, and M. tredecula. We used a Wilcoxon
signed-ranks test to determine if the average pitch preference of the Piatt Co. females
differed from that of the Sharp Co., AR females. Because of time constraints, we were
unable to study allopatric M. tredecim females.
To test the hypothesis that song pitch variations could be related to geographic
variation in body size, we compared the song pitches and body sizes of M. neotredecim
and M. tredecim males from sympatry at the powerline site with those of 17 M.
neotredecim males collected in Allerton Park, Piatt Co., IL, where no M. tredecim are
present. We used three characters to approximate size: right wing length, thorax width
between the wing articulations, and first abdominal sternite width between the sutures that
join it to the first abdominal tergite. We conducted pairwise comparisons among
populations using Mann-Whitney-U tests. For each population, we tested for an
association between size and song pitch using linear regressions.
Results. Female M. neotredecim preferences change with male song pitch: In
sympatry with M. tredecim, (Sharp Co., AR) female M. neotredecim were most responsive
to an average pitch of 1.75 kHz (n = 35), while in allopatry (Piatt Co., IL) female
preference averaged 1.31 kHz (n = 13; Wilcoxon signed ranks test, P < 0.002).
Size differences do not explain patterns in calling song variation because (1) M.
tredecim and M. neotredecim in sympatry overlapped broadly in size but not in song pitch,
(2) M. neotredecim populations from Illinois and Arkansas differed in song pitch but not in
size, and (3) we found no significant correlation between song pitch and any measure of
body size within species in any population (linear regression; Fig. 6.11).
164
Discussion
Reinforcement and Reproductive Character Displacement in M. neotredecim
Reinforcement is often viewed as a model of species formation (e.g. Butlin 1987,
1995; Liou and Price 1994; Kelly and Noor 1996), the only such model in which
prezygotic isolation evolves as a direct outcome of natural selection against hybridization.
In keeping with this view, Butlin (1987, 1989) suggests that the term “reinforcement”
should apply only to cases in which assortative mating is enhanced despite gene flow and
that the term “reproductive character displacement” should be used to refer to the
strengthening of prezygotic isolation when hybrids are sterile, because speciation is already
completed if gene flow is not possible. However, there are potential problems with this
distinction (see also Howard 1993). If the fundamental criterion for species status is
evolutionary independence (de Queiroz 1998), or low probability of future amalgamation,
then populations that have acquired sufficient postzygotic isolation to undergo
reinforcement are species even before they come into contact; the occurrence of
reinforcement is itself evidence of evolutionary independence. Furthermore, abundant
evidence of natural hybridization (Arnold 1997; Arnold and Emms 1998) shows that
species should not be identified, nor the completeness of speciation determined, by degree
of hybrid sterility. Practical difficulties with Butlin’s definitions arise because uncertainty
about past gene flow prevents the application of either term; for example there is little or no
evidence that M. neotredecim and M. tredecim currently hybridize, but this does not
necessarily mean that gene flow did not occur in the past. Consequently, we use the term
reinforcement to refer to a general process defined without respect to gene flow, speciation,
or relatedness of interactants, and we use the term reproductive character displacement to
identify the resulting pattern.
165
The process of reinforcement is the elaboration of differences between species’
sexual signals in response to selection against heterospecific sexual interactions (Grant
1972, Waage 1979, Howard 1993). If newly established contact between species involves
only slight range overlap, differences resulting from reinforcement may be manifest in the
form of clines, in which the two species are most extremely different in sympatry and less
distinct in allopatry (Brown and Wilson 1956). The clines reflect the relative contributions
of conspecific and heterospecific interactions to local optima of sexually-selected
characters. Individuals within the zone of contact face net directional selection for
distinctiveness from the other species, while conspecifics away from the zone of contact
face net selection for compatibility with their own species, with no benefit, and perhaps
some costs, to extreme individuals. Thus, in species undergoing reinforcement of
premating isolation, differences in species-specific sexual traits will become less extreme
away from the zone of contact where individuals face decreased pressures for distinction
from heterospecifics and increased selection for compatibility with conspecifics having no
history of contact with the other species involved.
A variety of observations support the hypothesis that the high song pitch of M.
neotredecim in sympatry with M. tredecim results from reinforcement with M. tredecim.
First, M. neotredecim in or adjacent to the area of overlap all have high-pitch songs (>1.6
kHz), the highest M. -decim dominant chorus pitch observed anywhere, while allopatric
M. neotredecim never do. Second, the dominant chorus pitches of M. tredecim and M.
neotredecim remain comparatively stable in allopatry, as does that of 17-year M.
septendecim, which varies only about 0.15 kHz in archived UMMZ recordings taken
throughout its widespread range (Table 6.3). Third, the typical song pitch of 17-year M.
septendecim, the likely sister group of M. neotredecim (see below), is indistinguishable
from that of allopatric M. neotredecim in Illinois (ca. 1.4 kHz); these species are also
indistinguishable in behavior, morphology, and mtDNA (Alexander and Moore 1962;
Simon et al. submitted). Fourth, song pitch apparently plays an important role in mate
166
recognition of both species. Fifth, M. neotredecim and M. tredecim songs in sympatry
differ just enough for average songs of each species to avoid frequency overlap, with M.
neotredecim downslurs ending at about the dominant song pitch of M. tredecim (Fig. 6.3).
Finally, the change in song pitch does not appear to be an incidental effect of a latitudinal
cline in body size: M. neotredecim populations in Arkansas and Illinois produce different
song pitches yet are indistinguishable in size, and song pitch does not correlate significantly
with body size within any population.
A potential challenge to the hypothesis of reinforcement in M. neotredecim arises
from the existence of some Missouri populations apparently well outside the range of M.
tredecim with a partially elevated dominant chorus pitch (ca. 1.5 kHz). This pattern could
be explained by the reinforcement model if (1) M. neotredecim colonized Missouri from
populations in Illinois that were themselves adjacent to the range of M. tredecim, or (2)
undiscovered M. tredecim populations exist in Missouri near the locations we sampled.
Future surveys should investigate the latter possibility. Because M. tredecim appears to
reach its northern limits on Mississippi River and Wabash River lowlands, it may be found
only in the vicinity of rivers elsewhere in the northern part of its range, increasing its
chances of being overlooked.
Also of interest is that the dominant chorus pitch of M. neotredecim does not appear
to be related to the relative abundance of the two 13-year M. -decim species in mixed
populations (linear regression, r2 = 0.053, P ≤ 0.317, n = 21); under certain conditions the
reinforcement model predicts this correlation because the strength of reinforcing selection
on one species should depend on the abundance of the other (Howard 1993; Noor 1995).
However, there are several possible reasons for this correlation to be absent even under the
reinforcement model. First, most M. neotredecim in populations sympatric with M.
tredecim have calls that are displaced on average just high enough to avoid frequency
overlap, suggesting that reinforcing selection ceases after the M. neotredecim song
167
phenotype reaches that point. If only a minor representation by M. tredecim is necessary to
drive this change in M. neotredecim, then a correlation between relative abundance and
degree of displacement would be detectable only in populations with very rare M.
tredecim; unfortunately we recorded few such populations. Second, the anticipated
correlation between relative abundance (in sympatry) and degree of displacement would be
weakened if mating signals are usually subject to stabilizing selection when not undergoing
reinforcement (e.g. Paterson 1993; Alexander et al. 1997); such selection would maintain
song displacement upon later changes in relative species abundance. This hypothesis could
be falsified in this case by finding evidence of lowered M. neotredecim song pitch where
M. tredecim is very rare and the weakly displaced M. neotredecim are unlikely to be
immediately descended from populations with undisplaced songs. Third, rapid,
continuous fluctuations in relative species abundance would also tend to weaken any
correlation between abundance and displacement.
A notable feature of the pattern of song displacement in Magicicada is its
asymmetry, although this phenomenon is not unusual in cases of reproductive character
displacement (e.g. Littlejohn 1965; Littlejohn and Loftus-Hills 1968; Fouquette 1975;
Waage 1979; Noor 1995). In general, because the strength of selection on each species
depends on factors that can differ between the species, symmetrical displacement is
probably unlikely (Grant 1972; Howard 1993). The selection each species imposes on the
other will be a function of (1) the benefits realizable from hybrid sexual interactions (if
hybrid offspring are fertile, the cost of hybridization may differ depending on which
species is which parent), and (2) the relative abundance of each species (the likelihood that
any individual will ever have a cross-species sexual encounter and face the danger of
hybridization; but see above). The selection for divergence imposed by heterospecifics
(through interspecific gene flow or the threat of it) should be stronger, and the stabilizing
selection imposed by conspecifics (through intraspecific gene flow) should be weaker for
the rarer of the species. Possible explanations for asymmetrical displacement in Magicicada
168
include: (1) greater numerical abundance of M. tredecim relative to M. neotredecim during
critical stages of the interaction, (2) greater M. tredecim female selectiveness upon initial
contact, and (3) greater constraints on evolution of lower song pitch.
The pattern of character displacement in M. neotredecim is similar to an example
from Gastrophryne (Anura: Blair 1955; Loftus-Hills and Littlejohn 1992) in east Texas.
Like M. neotredecim and M. tredecim, Gastrophryne carolinensis and G. olivacea in
sympatry are distinguishable by only average differences in morphological characters and
show the greatest nonoverlapping differences in dominant song pitch, a trait relevant to
species-specific mate recognition. As in the Magicicada example, evidence of hybridization
is weak and allopatric populations of the species show nearly or completely nonoverlapping
differences, a key observation suggesting that male calls were distinct, and mating errors
rare, even prior to sympatry. Like Loftus-Hills and Littlejohn (1992), we note that
reinforcement may occur even in the absence of interspecific mating as long as
enhancement of differences in male song or female preference increases the efficiency of
advertisement or mate-location.
The rarity of reproductive character displacement is a striking paradox (Alexander
1967; Walker 1974; Howard 1993; Alexander et al. 1997) given the ubiquity of song
distinctiveness of sympatric species and given other evidence of reinforcement (Coyne and
Orr 1989, 1997). Reproductive character displacement has been found in only one other
group of singing insects, crickets of the genus Laupala in Hawaii (Otte 1989). The paucity
of convincing examples has been interpreted differently by various authors; some point to a
lack of adequately studied cases (Walker 1974; Howard 1993), while others suggest that
sexual signal evolution may be driven mainly by within-species processes (West-Eberhard
1983; Paterson 1993). A somewhat neglected explanation for the paradox is that character
displacement may be quickly obscured by species range changes that disrupt the association
of song divergence with range overlap (Waage 1979): As discussed above, song changes
169
due to reinforcement might remain even after species’ ranges cease to overlap if mateattracting signals and female preferences for them are usually subject to stabilizing selection
(Paterson 1993; Alexander et al. 1997). Patterns of character displacement will also be lost
if sympatry becomes complete or if allopatric populations with the pre-contact song types
are lost.
Another possible explanation of the paradox is that, when male signals differ prior
to the establishment of sympatry, reinforcement may involve changes only in the stringency
of female mate acceptance criteria (Waage 1979; Gerhardt 1994; Kelly and Noor 1996) as
long as the costs of having a novel, minority receiver phenotype are offset by the
advantages of avoiding hybrid sexual interactions (Butlin and Ritchie 1994; see Halliday
1983). Such cases might tend to remain undiscovered because of an overemphasis on male
signals. This explanation is based in part on the premise that reinforcing selection should
occur initially and/or most strongly on females, who usually have the most to lose from
mating errors (Alexander et al. 1997). However, reinforcement of female traits alone
should be rare because any change in female behavior will tend to cause concerted
evolution in male signals: If selection on male signals derives mainly from efficiency in
mate attraction, males benefit from adaptations that increase their likelihood of securing a
female response, and male signals would evolve in concert with changes in female
preferences. Even if females in sympatry evolve only greater stringency of preferences,
with no change in the mean preference, variation among males in sympatry should be
reduced (relative to males in allopatry) as a result of stronger stabilizing selection.
Life Cycle Differences and Species Status
Designating M. neotredecim and 17-year M. septendecim as separate species may
provoke debate because many populations of M. neotredecim differ from M. septendecim
only in life cycle length and in geography (e.g. Lloyd 1984). However, if based upon
genetic differences, life cycle differences may provide genetic isolation and facilitate
170
evolutionary independence of population lineages, the fundamental criterion for species
status (Alexander and Moore 1962; de Queiroz 1998). The following evidence indicates
that Magicicada life cycle differences have a genetic basis: (1) in transplant experiments,
13-year cicadas reared north of 17-year cicadas emerged after a 13-year juvenile period
(Lloyd 1987); (2) the border between 13- and 17-year species appears historically stable
(but see Lloyd et al. 1983); and (3) the failure of small numbers of cicadas emerging offschedule to satiate avian predators and establish new populations (Marlatt 1923; Dybas
1969; Williams and Simon 1995) suggests persistent selection favoring additional genes
canalizing the majority life cycle phenotype. Life cycle differences substantially restrict
opportunities for gene flow, because M. neotredecim and M. septendecim emerge
simultaneously just once every 221 years. Moreover, during such co-emergences,
opportunities for interbreeding are limited further by the minimal geographic overlap
between 17- and 13-year populations and the fragmented nature of Midwestern populations
(Alexander et al. in prep.). Although some authors have suggested that 13- and 17-year
cicadas are widely sympatric (Marlatt 1923; Cox and Carlton 1991), more recent studies
(e.g. Lloyd et al. 1983) have found only limited sympatry. Alexander et al. (in prep.)
found that 17-year Brood III and 13-year Brood XIX overlap in Illinois in limited regions,
by no more than a few kilometers, and are most often separated by prairie gaps. Such
limited opportunities for gene flow are unlikely to prevent divergence of M. neotredecim
and M. septendecim, especially if differences in life cycle and geography impose divergent
selective pressures (Endler 1977; Rice and Hostert 1993). Absence of known life cycle
intermediates where M. neotredecim and M. septendecim meet and the stable position of
their contact zone suggest that gene flow is sufficiently restricted to permit divergence.
Simon et al. (submitted) discuss life cycle changes in Magicicada and note that this
process could bring M. neotredecim populations back into temporal synchrony and
panmixis with M. septendecim. Although life-cycle switching from M. septendecim is a
credible explanation for the origin of M. neotredecim, amalgamation of M. neotredecim and
171
M. septendecim by this process appears unlikely because (1) permanent life cycle changes
are apparently rare, (2) a random switching event has a low probability of establishing
synchrony between M. neotredecim and M. septendecim, and (3) a switching event that
synchronizes only part of M. neotredecim with M. septendecim will not affect the
remaining M. neotredecim.
The Origin of Magicicada neotredecim and Speciation in Periodical Cicadas
Cladistic analysis of M. –decim phylogeny and determination of ancestral life cycles
is not yet possible because the informative traits (morphological and molecular) within the
group are not shared with other Magicicada species or with other cicada genera; the same
problem obscures relationships among the Magicicada sibling groups. However, because
M. neotredecim and M. septendecim are in many locations distinguishable only in life cycle
and geography, and share the same mtDNA haplotype (Martin and Simon 1988, 1990),
one of these two species probably originated from ancestral populations of the other.
Together, these apparent cognates (descendants of an exclusive common ancestor or
speciation event) differ from M. tredecim in song pitch, abdomen coloration, and in
mtDNA by an estimated 2.6% of nucleotides, suggesting that their common ancestor
diverged from M. tredecim 1-2 million years ago (Martin and Simon 1990).
Biogeographic evidence suggests that populations of the M. septendecim lineage
gave rise to M. neotredecim, rather than vice versa, probably in the time since the last
glacial maximum. First, the comparatively restricted distribution of M. neotredecim,
surrounded on the west, north, and east by M. septendecim, suggests that M. neotredecim
may be the newer lineage; otherwise rapid and implausible range changes would be
required to explain the widespread distribution of M. septendecim. Second, the pattern of
M. neotredecim song variation, in which most M. neotredecim have undisplaced songs,
probably could not have persisted through range changes associated with Pleistocene
glacial cycles. As a species’ range shifts southward during glacial cooling, the populations
172
surviving in refuges are most likely founded from the southern margin of the earlier
distribution, while the most northern populations simply go extinct (Hewitt 1996). Had M.
neotredecim ever experienced such a range shift, all undisplaced M. neotredecim would
likely have been lost, because (1) M. neotredecim is completely bounded on the south by
M. tredecim, (2) southern M. neotredecim today all have displaced song pitches as a result
of overlap with M. tredecim, and (3) all northern, undisplaced M. neotredecim are found in
a region that 18,000 years ago may have contained spruce-fir forest and cooler, drier
conditions presently found only north of the range of Magicicada (Webb et al. 1993).
Thus, the pattern of reproductive character displacement, together with the similarity of
northern M. neotredecim and 17-year M. septendecim calling songs, is consistent with the
hypothesis that M. neotredecim originated recently via a permanent life cycle change from a
17-year ancestor (Martin and Simon 1988, 1990; Simon et al. submitted), although the
reverse hypothesis (17-year M. -decim evolving from Midwestern 13-year M. -decim)
cannot yet be ruled out.
Abdomen color and mtDNA data (Simon et al. submitted) indicate that the other
large 13-year brood (Brood XXIII), which inhabits most of the lower Mississippi Valley,
contains both M. neotredecim and M. tredecim, with M. neotredecim limited to northern
populations and partially overlapping M. tredecim. Archived UMMZ recordings from this
brood suggest that M. neotredecim has a displaced song pitch (ca. 1.7 kHz) where the two
species overlap; unfortunately we do not have recordings from any of the few populations
of pure M. neotredecim in Brood XXIII. Archived tapes further suggest that northern
populations of the remaining 13-year brood (Brood XXII), found only in southern
Mississippi and Louisiana, contain M. tredecim alone (dominant chorus pitch ca. 1.08 kHz;
Warren Co., MS, and Catahoula Parish, LA), although additional sampling will be
necessary to confirm the apparent absence of M. neotredecim from the brood. Simon et al.
(submitted) note that the presence of M. neotredecim or similar cicadas in Brood XXIII
could be explained in three ways: (1) by a second derivation from M. septendecim, (2) by a
173
single origin of Brood XIX M. neotredecim followed by the temporal migration of fouryear delayed M. neotredecim stragglers to Brood XXIII, or (3) by the formation of Brood
XXIII in part from four year delayed Brood XIX populations containing both M. -decim
species. Evidence that the developmental plasticity required by hypotheses #2 and #3
occurs in Magicicada includes observations of off-schedule cicadas emerging in numbers
difficult to explain by mutation (e.g. Dybas 1969; White and Lloyd 1979; Kritsky and
Simon 1996) and by the existence of temporally isolated broods of the same life cycle,
presumably formed as a result of short-lived regional climatic events that caused most
cicadas to delay or accelerate their emergence once (Alexander and Moore 1962; but see
Moore 1993).
Life cycle changes and speciation may be linked in Magicicada (see also Simon et
al. submitted). Each species appears most closely related to a cognate with the alternative
life cycle; it is therefore inescapable that 13- and 17-year life cycles have evolved repeatedly
in the group (Alexander and Moore 1962). The discovery that 17-year cicadas differ from
13-year cicadas in having a 4-year dormancy period (White and Lloyd 1975, Lloyd and
White 1976) increases the plausibility of the argument that Magicicada speciation often
involves permanent four year life cycle changes, if such a dormancy period could be
induced or deleted by few genetic changes. However, identifying plausible evolutionary
mechanisms for such life cycle evolution is a challenge because Magicicada populations
apparently must number many thousands per hectare to avoid being destroyed by avian
predators (Marlatt 1923; Beamer 1928; Alexander and Moore 1962; Dybas 1969; Williams
et al. 1993). In considering possible scenarios for the origin of M. neotredecim, we have
developed two models of life cycle evolution that may have general relevance for
Magicicada speciation, “nurse brood facilitation” and “canalization of climate-induced
plasticity.” These models are not exclusive alternatives; rather, the models are intended to
convey different ways in which the unique life-history attributes of Magicicada could be
involved in species formation.
174
“Nurse brood” facilitation and life cycle changes: Recent confirmation of
microgeographic overlap between 13- and 17-year broods in the Midwest (Alexander et al.
unpublished data; Lloyd et al. 1983), along with repeated observations of straggler cicadas,
suggests a mechanism in which speciation is initiated by temporal migration of life cycle
mutants from one brood to an overlapping brood of a different life cycle (Fig. 6.12). For
example, if mutations affecting life cycle length sometimes cause 17-year cicadas to emerge
with 13-year life cycles, such mutants could escape predation and establish an incipient
species (1) if they emerged in the life cycle overlap zone during an emergence of the
sympatric 13-year brood and (2) if the sympatric 13-year brood contained only dissimilar
13-year species in the area of overlap. Absence of a confusingly similar species in the 13year brood would probably be necessary to avoid loss of the temporal immigrants to
interspecific hybridization. The existing brood functions as a “nurse brood,” providing
predator protection for rare temporal mutants (Alexander amd Moore 1962); this process
has also been termed “induction” of one species into a brood lacking that species (Lloyd
and Dybas 1966b). One potential difficulty with this hypothesis is that multiple life cycle
mutants must appear in the same generation and location, but this problem may be mitigated
by the large size of periodical cicada populations, which can reach 3.7 million per hectare
(Dybas 1969), and by the possibility that minor genetic changes may influence life cycle
length (see above).
Nurse brood facilitation of life cycle mutants should lead to the formation of a new
incipient species whenever a given species in one brood does not already have a counterpart
in an overlapping brood of a different life cycle; this suggests an explanation for the
surprising tendency of Magicicada to repeatedly evolve the same life cycles (Alexander and
Moore 1962). In addition, reinforcement of premating isolation among sympatric species
of the same life cycle could interact with the nurse brood mechanism to produce new
species pairs. For example, reinforcement in M. neotredecim has incidentally caused some
populations of this species to differ from M. septendecim in dominant chorus pitch by at
175
least 300 Hz. If future 17-year life cycle mutants from song-displaced M. neotredecim
were to co-emerge with 17-year cicadas in the life cycle overlap zone (such as where 17year Brood X adjoins 13- year Brood XIX), these preexisting song pitch differences might
facilitate assortative mating and establishment of a new 17-year species (Fig. 6.13). The
general processes of life cycle change, nurse brood facilitation, and reinforcement could
explain patterns in Magicicada speciation including (1) pairs of cognate species differing
primarily in life cycle length and (2) initial synchronization of different species of the same
life cycle.
The hypothesis of nurse brood facilitation has the potential to explain the origin of
M. neotredecim from M. septendecim and its synchronization with the 13-year Brood XIX
(Fig. 6.14). However, for this hypothesis to be correct, a large region of Midwestern
Brood XIX must have contained only M. tredecassini and/or M. tredecula prior to the
arrival of M. neotredecim; otherwise most Midwestern M. neotredecim today would have
displaced (high-pitched) calling songs from past ecological interaction with M. tredecim.
Such a large region of prior range discordance between the 13-year species seems unlikely
because their distributions today are coincident throughout the rest of the 13-year range;
however such discordance does exist in 17-year cicadas in Oklahoma where the 17-year
brood (IV) contains only M. cassini (Dybas and Lloyd 1974). This hypothesis for the
origin of M. neotredecim predicts that Midwestern populations of M. tredecassini and/or
M. tredecula, which have not yet been adequately examined, are not more closely related to
their 17-year counterparts than to southeastern populations.
Canalization of climate-induced life cycle shift: The nurse brood mechanism can
explain only instances of life cycle evolution that occurred after broods of both life cycles
were already present. Unless 13- and 17-year life cycles appeared prior to the evolution of
periodicity and the need for predator satiation, additional mechanisms must be considered.
Without the nurse brood mechanism, life cycle change seems unlikely to be initiated by
176
mutation because life cycle mutants are unlikely to appear simultaneously in numbers
sufficient to satiate predators. We suggest a second model of Magicicada life cycle
evolution based on the assumption (discussed above) that unusual climatic stimuli can
trigger the expression of life cycle plasticity.
Developmental plasticity could facilitate life cycle evolution and speciation in
Magicicada if extreme climatic conditions sometimes induce periodical cicadas to switch life
cycle in numbers sufficient to satiate predators and if the conditions persist for generations.
Persistence of such extreme conditions would promote continued expression of the new
cycle, leading to selection that would (1) remove genes tending to cause expression of the
old phenotype and (2) favor alleles canalizing the new phenotype. Furthermore, if the
climate gradually returned to the initial conditions, canalizing selection could lead to the
evolution of cicadas that expressed the new life cycle even under the original conditions
(Fig. 6.15; Waddington 1953). The process would succeed only if the climate change was
substantial, sudden, and persistent. A sudden shift of lesser magnitude could create
overlapping broods or lead to local extinction if no populations remained large enough to
satiate predators. Gradual climate change could lead to canalization of the original cycle
and reduce the probability of a life cycle switch. A sudden change followed by an equally
sudden return to the original conditions would lead only to the formation of a temporallyisolated brood of the original life cycle. A general treatment of the role of phenotypic
plasticity in facilitating speciation can be found in West-Eberhard (1989).
If different Magicicada species possess similar life cycle plasticity, a climate shift
causing a change in the life cycle of one species will likely change sympatric species in
synchrony. In this manner, simultaneous, parallel speciation events could occur in
Magicicada (Figs. 6.13, 6.14). The hypothesis that different Magicicada species share
similar developmental plasticity is supported by the observation that both species (M.
177
septendecim and M. cassini) present in Brood XIII near Chicago produced millions of
four-year premature stragglers in the same year, 1969 (Lloyd 1984).
If M. neotredecim was formed from populations of M. septendecim by canalization
of a climate-induced life cycle, we might expect to find that the 13-year M. -cassini and M.
-decula sympatric with M. neotredecim are also more closely related to their 17-year
counterparts, because their ancestors would have been subjected to the same climate
fluctuations. There should also be paleoclimatological or other data consistent with recent,
dramatic climate change in Missouri and Illinois. The M. neotredecim / M. tredecim
overlap zone is concordant with the hypothesized southern boundary of the Prairie
Peninsula (Cox and Carlton 1991); perhaps prior to the climate change the border between
13- and 17-year cicadas was located there, in which case prior discordance between the 13year species’ ranges is not necessary to explain the undisplaced calling song of northern M.
neotredecim. One weakness of the canalization hypothesis is that it offers no explanation
other than chance for the synchronization of the new Midwestern 13-year cicadas with the
remainder of Brood XIX. Simon et al. (submitted) propose a related hypothesis in which
successive life cycle changes occur, each one farther north, and each facilitated by the
existence of the adjacent Brood XIX by a mechanism similar to the nurse brood process
discussed above.
178
Literature Cited
Alexander, R. D. 1967. Acoustical communication in arthropods. Ann. Rev. Entomol. 12:
495- 526.
Alexander, R. D. 1968. Life cycle origins, speciation, and related phenomena in crickets.
Q. Rev. Biol. 43: 1-41.
Alexander, R. D, and R. S. Bigelow. 1960. Allochronic speciation in field crickets, and a
new species, Acheta veletis. Evolution 14: 334- 346.
Alexander, R. D., and T. E. Moore. 1958. Studies on the acoustical behavior of seventeenyear cicadas (Homoptera: Cicadidae: Magicicada). Ohio J. Sci. 58: 107- 127.
Alexander, R. D., and T. E. Moore. 1962. The evolutionary relationships of 17-year and
13-year cicadas, and three new species. (Homoptera: Cicadidae, Magicicada).
Misc. Publ. no. 121. Univ. of Michigan Museum of Zoology, Ann Arbor, MI.
Alexander, R. D., D. C. Marshall, and J. R. Cooley. 1997. Evolutionary Perspectives on
Insect Mating. Pp. 4-31 in: B. Crespi and J. Choe, eds. The evolution of mating
systems in insects and arachnids. (Cambridge University Press).
Alexander, R. D., A. F. Richards, D. C. Marshall, and J. R. Cooley. (In prep). The
microdistributional relationships of 17- and 13-year cicadas in Illinois (Magicicada):
Broods III, 1997, and XIX, 1998.
Arnold, M. 1997. Natural hybridization and evolution. (Oxford University Press: New
York).
Arnold, M. L., and S. K. Emms. 1998. Paradigm lost: Natural hybridization and
evolutionary innovations. Pp. 379-389 in: D. J. Howard and S. H. Berlocher, eds.
Endless Forms: Species and Speciation. (Oxford University Press: New York).
Beamer, R. H. 1928. Studies on the biology of Kansas Cicadidae. Univ. Kansas Sci.
Bull. 18: 155- 263.
Blair, W. F. 1955. Mating call and stage of speciation in the Microhyla olivacea- M.
carolinensis complex. Evolution 9: 469-4 80.
Brown, W. L., and E. O. Wilson. 1956. Character displacement. Syst. Zool. 5: 49- 64.
Butlin, R. K. 1987. Speciation by reinforcement. TREE 2: 8- 1 3.
Butlin, R. K. 1989. Reinforcement of premating isolation. Pp. 158-179 in D. Otte and J.
A. Endler, eds. Speciation and its consequences. (Sinauer: Sunderland, MA).
Butlin, R. K. 1995. Reinforcement: an idea evolving. TREE 10: 432- 434.
179
Butlin, R. K., and M. G. Ritchie. 1994. Behaviour and speciation. In: P. J. B. Slater, and
T. R. Halliday, eds. Behaviour and Evolution. (Cambridge University Press).
Cooley, J. R., and D. C. Marshall. (In prep). Sexual signaling and mating behavior in
periodical cicadas (Magicicada spp.).
Cox, R. T., and C. E. Carlton. 1991. Evidence of genetic dominance of the 13-year life
cycle in periodical cicadas (Homoptera: Cicadidae: Magicicada spp.). Am. Midl.
Nat. 125: 63- 74.
Coyne, J. A., and H. A. Orr. 1989. Patterns of speciation in Drosophila. Evolution 43:
362- 381.
Coyne, J. A., and H. A. Orr. 1997. “Patterns of speciation in Drosophila” revisited.
Evolution 51: 295- 303.
de Queiroz, K. 1998. The general lineage concept of species, species criteria, and the
process of speciation. Pp. 57- 79 in: D. J. Howard and S. H. Berlocher, eds.
Endless Forms: Species and Speciation. (Oxford University Press: New York).
Dobzhansky, T. 1940. Speciation as a stage in evolutionary divergence. Am. Nat. 74: 312321.
Doolan, J. M., and D. Young. 1989. Relative importance of song parameters during flight
phonotaxis and courtship in the bladder cicada Cystosoma saundersii. J. Exp. Biol.
141: 113- 131.
Dybas, H. S. 1969. The 17-year cicada: A four-year mistake? Bull. Field Mus. Nat. Hist.
40: 10- 12.
Dybas, H. S., and M. Lloyd. 1974. The habitats of 17-year periodical cicadas (Homoptera:
Cicadidae: Magicicada spp.). Ecol. Monogr. 44: 279- 324.
Endler, J. A. 1977. Geographic variation, speciation and clines. (Princeton University
Press: Princeton, NJ).
Fouquette, M. J. 1975. Speciation in chorus frogs. I. Reproductive character displacement
in the Pseudacris nigrita complex. Syst. Zool . 24: 16- 22.
Gerhardt, H. C. 1988. Acoustic properties used in call recognition by frogs and toads. Pp.
455-483 in B. Fritsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W.
Walkowiak, eds. The evolution of the amphibian auditory system. (Wiley: New
York).
Gerhardt, H. C.. 1992. Conducting playback experiments and interpreting their results.
Pp. 59-77 in: P. K. McGregor, ed. Playback and studies of animal communication.
(Plenum Press: New York).
Gerhardt, H. C. 1994. Reproductive character displacement of female mate choice in the
grey treefrog, Hyla chrysoscelis. Anim. Behav. 47: 959- 969.
Grant, P. 1972. Convergent and divergent character displacement. Biol. J. Linn. Soc. 4:
39- 68.
180
Halliday, T. R. 1983. The study of mate choice. Pp. 3- 50 in: P. Bateson, ed. Mate
Choice. (Cambridge University Press).
Harrison, R. G. 1979. Speciation in North American field crickets: Evidence from
electrophoretic comparisons. Evolution 33: 009- 1023.
Harrison, R. G., and S. M. Bogdanowicz. 1995. Mitochondrial DNA phylogeny of North
American field crickets: perspectives on the evolution of life cycles, songs, and
habitat associations. J. Evol. Biol . 8: 209- 232.
Hewitt, G. M. 1996. Some genetic consequences of ice ages, and their role in divergence
and speciation. Biol. J. Linn . Soc. 58: 247- 276.
Howard, D. J. 1993. Reinforcement: Origin, dynamics, and fate of an evolutionary
hypothesis. Pp. 46-69 in: R. G. Harrison, ed. Hybrid zones and the evolutionary
process. (Oxford University Press: New York).
Kelly, J. K., and M. A. Noor. 1996. Speciation by reinforcement: A model derived from
studies of Drosophila. Genetics 143: 1485- 1497.
Kritsky, G., and S. Simon. 1996. The unexpected 1995 emergence of periodical cicadas
(Homoptera: Cicadidae: Magicicada) in Ohio. Ohio J. Sci. 96: 27- 28.
Liou, L. W., and T. D. Price. 1994. Speciation by reinforcement of premating isolation.
Evolution 48: 1451- 1459.
Littlejohn, M. J. 1965. Premating isolation in the Hyla ewingi complex (Anura: Hylidae).
Evolution 19: 234- 243.
Littlejohn, M. J., and J. J. Loftus-Hills. 1968. An experimental evaluation of premating
isolation in the Hyla ewingi complex (Anura: Hylidae). Evolution 22: 659- 663.
Littlejohn, M. L. 1977. Long-range acoustic communication in anurans: An integrated and
evolutionary approach. Pp. 263-294 in D. H. Taylor and S. I. Guttman, eds. The
reproductive biology of amphibians. (Plenum Press: New York).
Lloyd, M. 1984. Periodical cicadas. Antenna 8: 79- 91.
Lloyd, M. 1987. A successful rearing of 13-year periodical cicadas beyond their present
range and beyond that of 17-year cicadas. Am. Midl. Nat. 117: 362- 368.
Lloyd, M., and H. S. Dybas. 1966a. The periodical cicada problem. I. Population
ecology. Evolution 20: 133- 149.
Lloyd, M., and H. S. Dybas. 1966b. The periodical cicada problem. II. Evolution.
Evolution 20: 466- 505.
Lloyd, M., and J. White. 1976. Sympatry of periodical cicada broods and the hypothetical
four-year acceleration. Evolution 30: 786- 801.
Lloyd, M., G. Kritsky, and C. Simon. 1983. A simple Mendelian model for the 13- and
17-year life cycle of periodical cicadas, with historical evidence for hybridization
between them. Evolution 37: 1162- 1180.
181
Loftus-Hills, J. J., and M. J. Littlejohn. 1992. Reinforcement and reproductive character
displacement in Gastrophryne carolinensis and G. olivacea (Anura: Microhylidae):
A reexamination. Evolution 46: 896- 906.
Marlatt, C. L. 1923. The periodical cicada. U.S.D.A. Bureau of Entomology Bulletin 71,
183 pp.
Martin, A. P., and C. Simon. 1988. Anomalous distribution of nuclear and mitochondrial
DNA markers in periodical cicadas. Nature (Loondon) 336: 237- 239.
Martin, A. P., and C. Simon. 1990. Differing levels of among-population divergence in
the mitochondrial DNA of periodical cicadas related to historical biogeography.
Evolution 44: 1066- 1080.
Moore, T. E. 1993. Acoustic signals and speciation in cicadas (Insecta: Homoptera:
Cicadidae). Pp. 269-284 in: D. R. Lees and D. Edwards, eds. Evolutionary
patterns and processes. (Academic Press: London).
Noor, M. A. 1995. Speciation driven by sexual selection in Drosophila. Nature (London)
375: 674- 675.
Otte, D. 1989. Speciation in Hawaiian crickets. Pp. 482-526 in: D. Otte and J. A. Endler,
eds. Speciation and its consequences. (Sinauer : Sunderland, MA).
Paterson, H. E. H. 1993. Evolution and the recognition concept of species. (Johns
Hopkins University Press: Baltimore, MD).
Rice, W. R., and E. E. Hostert. 1993. Laboratory experiments on speciation: What have
we learned in 40 years? Evolution 47: 1637- 1653.
Simon, C. 1988. Evolution of 13- and 17-year periodical cicadas (Homoptera: Cicadidae:
Magicicada ). Bull. Entomol. Soc. Am. 34: 163- 176.
Simon, C., J. Tang, S. Dalwadi, G. Staley, J. Deniega, and T. Unnasch. Submitted.
Restricted gene flow between 13-year cicadas and coexisting “17-year cicadas with
13-year life cycles.”
Tauber, C. A., and M. J. Tauber. 1977a. A genetic model for sympatric speciation through
habitat diversification and seasonal isolation. Nature (London) 268: 702- 705.
Tauber, C. A., and M. J. Tauber. 1977b. Sympatric speciation based on allele changes at
three loci: evidence from natural populations in two habitats. Science (Wash. D. C.)
197: 1298- 1299.
Toms, R. B., and D. Otte. 1988. New genera, species and records of east and southern
African tree crickets (Orthoptera: Gryllidae: Oecanthinae). Ann. Transvaal Museum
34: 469- 533.
Waage, J. K. 1979. Reproductive character displacement in Calopteryx (Odonata:
Calopterygidae). Evolution 33: 104- 116.
Waddington, C. H. 1953. Genetic assimilation of an acquired character. Evolution 7: 118126.
182
Walker, T. J. 1974. Character displacement in acoustic insects. Am. Zool. 14: 1137- 1150.
Webb T. III, P. J. Bartlein, S. P. Harrison, and K. H. Anderson. 1993. Vegetation, lake
levels, and climate in eastern North America for the past 18,000 years. Pp. 415-467
in: H. E. Wright, Jr., J. E. Kutzbach, T. Webb III, W. F. Ruddiman, F. A. StreetPerrott, and P. J. Bartlein, eds. Global climates since the last glacial maximum.
(University of Minnessota Press, Minneapolis, MN).
Weber, T., T. E. Moore, F. Huber, and U. Klein. 1987. Sound production in periodical
cicadas (Homoptera: Cicadidae: Magicicada septendecim, M. cassini). Pp. 329-336
in: Proc. 6th Auchen. Meeting, Turin, Italy, 7-11 Sept. 1987.
West-Eberhard, M. J. 1983. Sexual selection, social competition, and speciation. Q. Rev.
Biol. 58: 155- 183.
West-Eberhard, M. J. 1989. Phenotypic plasticity and the origins of diversity. Ann. Rev.
Ecol. Syst. 20: 249- 278.
White, J., and M. Lloyd. 1975. Growth rates of 17- and 13-year periodical cicadas. Am.
Midl. Nat. 94: 127-1 43.
White, J., and M. Lloyd. 1979. 17-year cicadas emerging after 18 years: a new brood?
Evolution 33: 1193- 1199.
Williams, K. S., and C. Simon. 1995. The ecology, behavior, and evolution of periodical
cicadas. Ann. Rev. Entomol. 40: 269- 295.
Williams, K. S., K. G. Smith, and F. M. Stephen. 1993. Emergence of 13-yr periodical
cicadas (Cicadidae: Magicicada): phenology, mortality, and predator satiation.
Ecology 74: 1143- 1152.
Young, D., and R. K. Josephson. 1983. Pure-tone songs in cicadas with special reference
to the genus Magicicada. J. Comp. Physiol. 152: 197- 207.
183
184
> 3.00
> 3.00
black, rarely with weak ††
orange lateral band
black, rarely with weak ††
orange lateral band
black with orange lateral band
black with orange lateral band
17
13
17
13
M. cassini (Fisher)
M. tredecassini Alexander and Moore
M. septendecula Alexander and Moore
M. tredecula Alexander and Moore
> 3.00
black
black
black
black
orange
orange
orange
Pronotal
extension color
7 - 14 †
7 - 14 †
2 - 6 §§
2 - 6 §§
1.5 - 4 §
1.5 - 4 §
1.5 - 4 §
Length of call
(seconds)
§§
Roughly pure-tone, musical buzz terminating in a noticeable drop in pitch; no ticks. Usually 2-3 calls between flights.
Rapid series of ticks followed by high-pitched, broad-spectrum buzz that rises and then falls in intensity and pitch. Usually 1-2 calls between flights.
†
Repeated, rhythmic high-pitched, broad-spectrum tick-buzz phrases, followed by repeated phrases containing only ticks. Usually 1 call between flights.
††
Orange band, if present, often interrupted medially.
§
1.25 - 1.50
orange with black lateral band
or center
17
M. septendecim (L.)
> 3.00
1.00 - 1.25
1.35 - 1.90
Dominant call
pitch (kHz)
mostly orange
orange with black lateral band
or center
Abdominal sternite color
(each)
13
13
Life Cycle
(years)
M. tredecim (Walsh and Riley)
M. neotredecim Marshall and Cooley (new
species)
Species
Table 6.1. Traits distinguishing Magicicada neotredecim and other Magicicada species. "Pronotal extension" is the lateral
extension of the pronotum behind the eye. For additional description and color photographs see Alexander and Moore (1962)
Table 6.2. Relationship of abdomen color to male call pitch and female pitch preference
in M. -decim at the Sharp Co. AR “powerline” site. The male and female M. -decim
samples were each divided on the basis of calling song pitch and pitch preference (at 1.5
kHz). The abdomen colors of these groups are significantly different.
Males
Call
Category
< 1.5 kHz
> 1.5 kHz
n
27
124
Average male
dominant call
pitch in kHz
(mean ± SD)
1.11 (±0.07)
1.70 (±0.07)
Abdomen
color score
(mean ± SD)
3.67 (±0.55)
2.00 (±0.44)
Species
designation
M. tredecim
M. neotredecim
Abdomen color differs significantly in each group (Mann-Whitney U= 3348.0; P< 0.001)
Females
Preference
Category
< 1.5 kHz
> 1.5 kHz
n
13
35
Average female
call pitch
preference in kHz
(mean ± SD)
1.22 (±0.11)
1.75 (±0.14)
Abdomen
color score
(mean ± SD)
3.23 (±0.83)
2.20 (±0.72)
Species
designation
M. tredecim
M. neotredecim
Abdomen color differs significantly in each group (Mann-Whitney U= 366.5; P< 0.001)
185
Table 6.3. Dominant frequencies of Magicicada septendecim chorus recordings in the
UMMZ sound library.
Brood
I
II
II
III
IV
VIII
XIII
XIV
State
VA
CT
VA
IL
KS
PA
IL
OH
Year
1995
1962
1996
1997
1998
1976
1973
1957
Frequency
1.41
1.39
1.40
1.35
1.34
1.32
1.30-1.31
1.29-1.32
186
M. septendecim (17)
- abdomen black and orange
- call pitch 1.25-1.50 kHz
- mtDNA lineage A
M. cassini (17)
M. septendecula (17)
M. neotredecim (13)
M. tredecim (13)
- abdomen black and orange
- call pitch 1.25-1.90 kHz
- mtDNA lineage A
- abdomen mostly orange
- call pitch 1.00-1.25 kHz
- mtDNA lineage B
M. tredecassini (13)
M. tredecula (13)
Figure 6.1. Distribution of the seven periodical cicada (Magicicada)
species, summarized from county-level maps in Simon (1988) and from
1993-1998 field surveys. The 17-year species are sympatric except in
peripheral populations: M. cassini alone inhabits Oklahoma and Texas,
while only M. septendecim is found in some northern populations
(Dybas and Lloyd 1974). Two 13-year species (M. tredecassini and
M. tredecula) are sympatric across the entire 13-year range, while the
remaining 13-year species, M. tredecim and the new species
M. neotredecim, overlap only in the central US. County-level maps
overestimate distribution limits, hence the overlap between the 13- and
17-year populations is probably exaggerated. We have found only
limited 13/17 overlap during recent surveys in Illinois (1997-1998
emergences of Broods III and XIX, respectively) and have adjusted this
map accordingly. The overlap of M. tredecim and M. neotredecim is
plotted from recent field surveys (this paper, Simon et al. submitted).
Characters distinguishing M. –decim siblings are noted; the M. -cassini
and M. -decula siblings are distinguishable only by life cycle.
"Call pitch" is dominant pitch of male call phrase; mtDNA lineage
refers to types described in Martin and Simon (1990).
187
M. septendecim
M. cassini + M. septendecula
A.
Power
kJ/Hz
1
2
3
4
Pitch (kHz)
5
6
7
B.
M. tredecassini + M. tredecula
M. tredecim
M. neotredecim
Power
kJ/Hz
1
2
3
4
Pitch (kHz)
5
6
7
Figure 6.2. Power spectra of typical 17- year chorus containing M. septendecim,
M. cassini, and M. septendecula (A); and 13- year chorus containing
M. tredecim, M. neotredecim, M. tredecassini, and M. tredecula (B). Note that
M. -cassini and M. -decula frequency characteristics differ little in the two
choruses, but that the presence of two M. -decim species in the 13- year chorus
is readily apparent. 17 - year chorus recording S. of Fandon, McDonough Co.,
IL, 7 June 1980 (Brood III); 13- year chorus along US 51 near county line in
southern Jackson Co., IL 1 June 1976 (Brood XXIII).
188
8
189
0.5
1.0
Pitch 1 . 5
(kHz)
2.0
2.5
2.0
Time (s)
4.0
6.0
Figure 6.3. Spectrogram (power spectrum vs. time) showing a two-banded, mixedspecies chorus of male calls with one call of each species standing out against
the background chorus. Individual calls end with a downslur. Comparatively
faint slurs of background chorus males overlap and are not visible. Intervening
time between calls has been removed.
0.0
M. tredecim
M. neotredecim
190
1
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
71.1
79.9
81
81.5
81.7
82
83
Air temperature, °F
81.5
83.3
84.9
85.5
87.1
87.4
M. tredecim
M. neotredecim
Fig. 6.4. Dominant pitch of a mixed M. neotredecim/M. tredecim as a function of air temperature.
No effect of temperature is evident.
Chorus pitch (kHz)
191
1
2
0
2
4
6
8
10
12
16
Time (seconds)
14
18
20
22
24
26
28
30
Figure 6.5. Sonogram of 30-second recording of an individual male M. neotredecim
in Sharp County, AR, illustrating consistency of individual male calling song pitch.
Pitch
(kHz)
192
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
2.3
Pitch (kHz)
0
100
0.1
0
200
300
400
500
0.2
0.3
0.4
0.5
600
Figure 6.6. Power spectrum (shaded area) of mixed M. neotredecim and M. tredecim chorus at
"powerline" study site, Sharp Co., AR, with accompanying frequency histograms of male song
pitches (black bars) and female pitch preferences (white bars) of individuals collected
randomly at the site. M. tredecim (low song pitch) constitute only 8% of the population.
Proportion
0.6
Power (nJ/Hz)
193
TN
AL
KY
GA
SC
NC
VA
100 km
N
MD
Figure 6.7. Proportions of M. neotredecim (black) and M. tredecim (white) estimated from chorus recordings
of the 1998 emergence of Magicicada 13-year Brood XIX.
AR
MO
IL
IL
MO
N
AR
Dominant pitch
of chorus (kHz)
1.68 - 1.77
1.58 - 1.67
1.48 - 1.57
1.38 - 1.47
100 km
Figure 6.8. Geographic variation in dominant chorus pitch of M. neotredecim,
showing reproductive character displacement (higher pitch songs) in and near the
region of overlap with the low-pitched M. tredecim. Lighter shaded circles indicate
higher pitched songs. Shaded region is approximate M. tredecim range. Weak
choruses not plotted.
194
Song
pitch
(kHz)
1.8
1.7
1.6
1.5
1.4
M. neotredecim
1.3
1.2
M. tredecim
1.1
1.0
N
S
Figure 6.9. Schematic representation of calling song
displacement in Brood XIX M. tredecim and M. neotredecim.
Light portions of pie graphs represent proportion of
M. tredecim present in choruses; dark portions represent
proportion of M. neotredecim. In the north, choruses contain
only M. neotredecim, while in the south, choruses contain
only M. tredecim. Song pitch is displaced in the relatively
narrow zone of overlap betweeen these species.
195
196
AR
MO
IL
TN
AL
KY
GA
SC
100 km
NC
1.15 - 1.17
1.12 - 1.14
1.10 - 1.11
1.07 - 1.09
Dominant pitch
of chorus (kHz)
VA
MD
Figure 6.10. Geographic variation in dominant chorus pitch of M. tredecim, suggesting weak reproductive character displacement.
Lighter shaded circles indicate more displaced (lower pitched) songs. Shaded area is approximate M. neotredecim range. Note
that range of variation is only 1/4th of that shown in Fig. 6.7. Weak choruses not plotted.
N
*
A.
B.
1.9
Right wing length (mm)
40
*
1.8
Call pitch (kHz)
1.7
1.6
*
1.5
*
1.4
1.3
*
35
*
*
30
1.2
*
1.1
*
1.0
0.9
25
Arkansas
M. tredecim
Illinois
Arkansas
M. neotredecim
M. tredecim
D.
C.
8.0
First sternite width (mm)
12
Thorax width (mm)
Illinois
M. neotredecim
11
10
7.5
7.0
*
6.5
*
*
6.0
9
Arkansas
M. tredecim
Illinois
Arkansas
M. neotredecim
M. tredecim
Illinois
M. neotredecim
Figure 6.11. Box plots of male calling song pitch (A), right wing length (B), thorax
width (C), and first sternite width (D) of M. tredecim (Sharp Co., AR; n = 26) and
M. neotredecim (Sharp Co., AR, n = 61; and Piatt Co., IL, n = 17). Asterisks indicate
outliers; open circles indicate extreme outliers. Song pitch of M. neotredecim is
significantly different in Arkansas (sympatry with M. tredecim) and Illinois (allopatry
with M. tredecim; P < 0.001, Mann-Whitney-U), indicating reproductive character
displacement, while size measurements show no significant differences within
M. neotredecim. M. tredecim is different from M. neotredecim in all size traits
(for each, P < 0.001; Mann-Whitney-U).
197
198
13 A'
Life cycle
overlap zone
South
13 C
13 B
13 A
Figure 6.12. Formation of an incipient Magicicada species (13-year A') by "nurse brood
facilitation" of life cycle mutants. Three 17-year species each have a similar 13-year
counterpart, but the 13-year counterpart of species 17 A is not present where the life
cycle types overlap geographically. In this situation 13-year life cycle mutants from
17 A (curved arrow) can establish a new species (13 A') in the overlap zone if they
emerge synchronously with the 13-year "nurse" brood (13 B + 13 C). Success of
life cycle mutants from 17 A is facilitated because 13 B and 13 C provide the rare
mutants with numerical protection from predators. Rare life cycle mutants from 17 B
and 17 C are likely to be lost to interspecific hybridization with the similar and abundant
13 B and 13 C. Degree of prior divergence of 13 A and the parent species 17 A will
influence the evolutionary outcome (mixture, hybrid zone, or reinforcement) of later
contact between 13 A' and 13 A.
17 C
17 B
17 A
North
199
A
a
a'
A
II
a
a'
13
A''
A''
17
A
a'
a
13
A
17
III
IV
a'
a
13
Our data suggest that steps I and II have occurred. Steps III and IV are plausible by
extension of the model.
I. 17-year cicadas (A) mutate (to a') and join an overlapping, co-emerging 13-year brood
II. Reinforcement occurs between a' and a in the 13-year brood
III. Individuals of 13-year a' mutate and join the 17-year brood, forming A"
IV. Reinforcement occurs between A and A" in the 17-year brood
Figure 6.13. A model of Magicicada speciation using nurse brood facilitation and
reinforcement of premating isolation to explain pairs of similar life cycle cognates.
Vertical dimension reflects degree of character divergence.
17
13
17
I
Hypothesis 1
17
Present
13
Observed pattern
Hypothesis 2
17
13
13
13
M. -decim
M. -decim
17
13
13
C
17
13
17
M. -decim
M. -cassini
M. -decula
?
13
M. -cassini and M. -decula
M. -cassini and M. -decula
17
17
13
B
13
M. -decim
M. -decim
M. -cassini
M. -decula
17
13
M. -cassini and/or M. -decula
17
C
17
C
13
A
M. -decim
M. -cassini
M. -decula
17
C
Past
13
M. -decim
13
M. -cassini and/or M. -decula
Figure 6.14. Two hypotheses for the origins of Midwestern 13- year cicadas. In the first hypothesis
(left column), only M. septendecim underwent a life cycle change. At time (A), a region (marked C )
contained 13- year M. -cassini and/or 13- year M. -decula, but lacked 13- year M. -decim. At time
(B), some 17- year M. -decim in the shaded area of slight overlap between 13- and 17- year cicadas
underwent a permanent life cycle change and became a new 13- year species, M. neotredecim,
synchronous with the existing 13- year brood to the south, and escaping predation by the "nurse
brood" mechanism. Under the alternative hypothesis (right column), three species simultanesouly
underwent a life cycle change. At time (A), the Midwestern region (marked C ) presently inhabited
by M. neotredecim was inhabited by 17- year cicadas. During time (B), large numbers of 17- year
cicadas in the shaded area underwent a permanent life cycle change, becoming 13- year cicadas.
The process leading to mass changes in life cycle possibly involved environmentally- induced
canalization of formerly plastic life cycles. Under both hypotheses, time (C) is the present, in which
two 13- year M. -decim species exist, but any new 13- year M. -cassini or M. -decula species, if they
exist, are cryptic. See text for details.
200
Proportion of
life cycle
phenotypes
Climate
parameter
A
B
C
D
Time
Figure 6.15. A model of Magicicada life cycle evolution via canalization of
a climate-induced life cycle shift. Graph shows temporal change in a climate
parameter such as temperature. Pie charts indicate proportion of cicadas
emerging in 17 years (light) and 13 years (dark). During stage A, all cicadas
emerge on a 17-year cycle but are capable of expressing life cycle length
plasticity and emerging in 13 years under unusual climatic conditions. During
stage B, the climate changes suddenly and dramatically such that the majority of
cicadas are induced to express the 13- year cycle. During stage C, climatic
conditions slowly ameliorate, imposing canalizing selection for the majority life
cycle phenotype of 13- years because 17-year stragglers are never abundant enough
to survive predation. By stage D, the population has evolved to express the new
13-year cycle even under the original conditions.
201
CHAPTER VII
ASYMMETRY AND MATING SUCCESS IN A PERIODICAL CICADA,
MAGICICADA SEPTENDECIM (HOMOPTERA: CICADIDAE)
Abstract
Phenotypic symmetry has been portrayed as an index of environmental or genetic
quality. However, like any trait, symmetry results both from a heritable genetic
predisposition and the influence of the developmental environment, and thus for species in
which choosy females gain only genetic benefits, symmetry will be of no greater utility as a
choice criterion than any other indicator trait. When choosy females receive material
benefits, any functional impairment imposed by phenotypic asymmetry may reduce the
benefits available to females accepting asymmetrical mates. This paper (1) discusses the
possible causes of asymmetry and clarifies the specific situations in which symmetry-based
mate choice is likely, and (2) presents research on female mate choice and symmetry in the
periodical cicada, Magicicada septendecim (L.). Male mating aggregations and lack of
mating gifts or parental care make this species an ideal target for studying female choice for
202
superior mates, and thus for a relationship between male symmetry and male mating
success.
Introduction
For bilaterally symmetric organisms, phenotypic symmetry results from heritable
precision in ontogenetic canalization. Deviations from symmetry result from (1) genetic
quality (because genotypes susceptible to perturbation or with poorly interacting
constituents lead to asymmetries), (2) environmental quality (because conditions during
development influence production of a symmetrical phenotype), or (3) interactions between
these two sources. In any population, zero-centered, normally distributed deviations from
perfect bilateral symmetry are termed “fluctuating asymmetry” (FA; Van Valen 1962).
Asymmetries that are “handed” (“antisymmetry;” e.g., in humans, one hand is generally
dominant; some are right- while others are left- handed) or directional (e.g., all humans
have an asymmetrical heart in which the right side is responsible for pulmonary circulation)
have different population distributions. The terms “fluctuating asymmetry,”
“antisymmetry,” and “directional asymmetry” refer only to population distributions and not
the underlying causes of asymmetry.
Symmetry has been used to address diverse topics, from mate choice to
environmental degradation (Clarke 1993, Møller and Swaddle 1997). However, the
widespread application of symmetry theory is not without controversy; some of the most
recent and hotly debated exchanges in symmetry theory (see Møller 1983a, Palmer 1996,
Møller and Thornhill 1997a, 1997b, Houle 1997, Leamy 1997, Markow and Clarke 1997,
Palmer and Strobeck 1997, Pomiankowski 1997, Swaddle 1997, Whitlock and Fowler
1997, Clarke 1998, Houle 1998) result from an oversimplification of the multiple causes of
asymmetry and the resulting measurable patterns. For example, some claims that mate
choice favors individuals exhibiting developmental stability are based on correlations
between mean population developmental stability (as measured by asymmetry) and mean
203
population fitness (Møller 1997 but see Clarke 1998), yet do not take into consideration
whether individuals receive any benefits at all from using symmetry as a mate choice
criterion. Previous studies also suffer from weak tests of their hypotheses; they succeed in
uncovering evidence consistent with their hypotheses, but offer no strong tests of the
alternatives (Møller and Höglund 1991, Møller 1993b, c, Møller and Pomiankowski 1993,
1994, Møller and Zamora-Muñoz 1997), a problem exacerbated by potential observer
biases (see Houle 1998).
Causes of asymmetry
Environmental stress. The term “developmental stability” refers to an
organism’s ability to develop a normal phenotype in spite of environmental perturbations.
For bilaterally symmetric organisms, deviations from symmetry due to flaws, damage,
developmental inaccuracies, or an inability to withstand hostile aspects of the environment
are indicators of “environmental stress” (see Møller 1993 a, b). In any population,
normally-distributed, non-adaptive asymmetries with a mean value of zero fit the criteria of
fluctuating asymmetry (FA; Van Valen 1962); thus the pattern of fluctuating asymmetry is
one estimate of the processes involved in developmental stability. Although fluctuating
asymmetry is a population index, if a population exhibits FA, then the asymmetries of its
members are individual manifestations of developmental instability (and thus individual FA
even though the term itself was defined as a measure of population characteristics; see
Møller 1993d, Møller and Swaddle 1997 p. 11).
Symmetry is not, however, synonymous with developmental stability.
Antisymmetries or directional asymmetries are often adaptive or result from the pleiotropic
effects of adaptations. Adaptively asymmetrical structures, such as the human heart, should
be just as subject to developmental stress as bilaterally symmetric characters (Møller and
Swaddle 1997), but such characters are rarely used in studies of developmental stability.
Perturbations might have different impacts depending on when in the developmental
204
process they occur or how effectively subsequent development compensates for them. For
instance, the closer such an event happens to the ontogenetic origin of a cell stem line, the
more serious could be its effects (Emlen et al. 1994; see Polak 1994). Environmentallycaused deviations from a genome’s developmental program are identifiable only given
some information about how that genotype has been selected to develop in the absence of
perturbation. Such direct information is virtually unobtainable, and in most cases, we have
only a population average phenotype, or, for a specific genotype, a reaction norm, either of
which might be affected by population- or environment- specific stresses. Studies of
developmental stability thus rely on characters for which it is safe to assume that normal
development produces symmetrical phenotypes, because deviations from normal
development are readily identifiable as fluctuating asymmetry (Møller and Swaddle 1997).
Environmentally-induced asymmetries are not, by definition, heritable; apparent
heritabilities of environmentally-induced asymmetries result from “common environment”
effects (e.g., if parents and offspring develop in similarly stressful environments).
Environmentally-induced asymmetries are in addition to or in spite of genetic endowment,
and are deviations from the most symmetrical phenotype that a given genotype is capable of
producing. Evidence for environmentally-induced asymmetry consists of decreased
symmetry in laboratory animals under elevated environmental stress (Parsons 1961;
reviewed in Møller and Pomiankowski 1993), and some evidence of elevated asymmetry in
animal populations living in marginal habitats (see Parsons 1994).
Genetic quality. Discussions of asymmetry often invoke genomic effects,
including the concept of “balanced genotypes” (Lerner 1954, Waddington 1957) and the
impact of “genetic balance” on development. Both are indices of how well the constituent
genes of a genome interact, such that observable asymmetries result from the pleiotropic
effects of the genome in its entirety. For example, heterozygous individuals sometimes
exhibit a high degree of symmetry (Mitton 1994), and hybrids, often especially
205
heterozygous, tend to be highly symmetrical (Parsons 1990). Heterozygosity may lead
directly to increased symmetry, or may do so by conferring resistance to parasites or other
stresses promoting asymmetry.
Genetic effects on asymmetry also include inheritable predisposition/ resistance to
asymmetrical development and adaptations that compensate for environmental
perturbations. A number of studies demonstrate that sensitivity to environmental
perturbations is inheritable, although some are more thorough than others in ruling out
common-environment effects (Palmer et al. 1994, Houle 1997, Leamy 1997, Markow and
Clarke 1997, Palmer and Strobeck 1997, Pomiankowski 1997, Swaddle 1997, Whitlock
and Fowler 1997). The many possible causes of asymmetry make any broad
generalizations about heritability and symmetry questionable (Whitlock 1996, Woods et al.
1998, but see Møller and Thornhill 1997a, 1997b; Møller and Swaddle 1997).
If symmetrical males enjoy mating or survival advantages, the ability to develop
symmetrically would be subject to selection, with poor variants at a disadvantage (e.g.,
Møller and Pomiankowski 1994, Palmer and Strobeck 1997). This selection would cause
the elimination of genetic variants promoting asymmetry (specific examples of “bad
genes”), or place a premium on the evolution of modifier or masking genes (specific
examples of “good genes”) that enforce symmetry regardless of (or in spite of) genomic
qualities or environmental effects. Such modifier genes are another possible genetic
influence on phenotypic symmetry. Reduced levels of asymmetry in sexual ornaments,
compared to levels of asymmetry in other morphological characters, suggest that such
genetic effects exist and are favored by selection (Møller 1990, Møller and Höglund 1991,
Møller 1993c, d, e, Møller and Pomiankowski 1993, 1994, Møller 1995).
206
Symmetry and Mate Choice
Symmetry has been linked to male dominance in fallow deer (Malyon and Healy
1994), attractiveness in humans (Gangstead et al. 1994, Møller et al. 1995), and mating
success in barn swallows (Møller 1993b, 1994) and zebra finches (Swaddle 1996).
Symmetrical male Japanese scorpionflies generally tend to win both inter- and intraspecific fights for control of mating gifts (Thornhill 1992a, b), the most symmetrical male
dung flies and houseflies are also the most successful breeders (Liggett et al. 1993, Møller
1996), and the most symmtrical dragonflies (Coenagrion puella) have the highest lifetime
mating success (Harvey and Walsh 1993). Although not all studies show a relationship
between symmetry and mating success (see Oakes and Barnard 1994, Tomkins and
Simmons 1998), at least some mate choice mechanisms appear to favor symmetry or its
correlates.
Phenotypic symmetry is tempting as a focus of mate choice studies, because
symmetry gives the appearance of being a universal indicator of genetic quality: Nearly all
phenotypes exhibit some form of symmetry, and symmetry at least partly reflects genomic
qualities. As a result, symmetry appears to occupy a special status as an accurate and
uncheatable quality indicator (Møller and Swaddle 1997). However, like all indicator
traits, phenotypic symmetry is subject to a paradox: (1) As the number of genes involved in
affecting the trait increases (in the extreme, if expression of the trait reflects the overall
genome), the less accurately the trait identifies mates able to provide inheritable genetic
benefits, because meiosis and independent segregation ensure that large, coadapted pieces
of the genome are not passed on intact (thus offspring quality will depend on the
“goodness” of the match between the genes in the sperm and egg that join to create the
zygote); yet (2) The smaller the number of genes with a predominant influence on the trait,
the more accurately the phenotype identifies an inheritable genetic basis for the trait, but the
greater the likelihood that population-wide variation will tend to be homogenized by
207
successive generations of selection. The solution to this paradox is that indicator traits
generally are not stable unless they are costly or condition-dependent, reducing the
likelihood that expression of advantageous phenotypes can become fixed by selection
(Zahavi 1977, Halliday 1978, Borgia 1979, West-Eberhard 1979, Andersson 1982,
Hamilton and Zuk 1982, Parker 1982, Price et al. 1993, Kodric-Brown and Brown 1984).
Condition-dependence implies that traits similar to category (1) above are most likely to
become the basis for stable mate choice systems, because condition-dependence involves
the performance of the entire organism, and thus involves the entire genome.
Considering symmetry as a condition-dependent trait opens up new possibilities for
understanding the mechanisms by which symmetrical mates are chosen. Although studies
of neural networks suggest that relatively simple mechanisms could detect and favor
symmetrical mates (Johnstone 1994, Enquist and Arak 1994), if symmetry is conditiondependent, it is also possible that females have no evolved mechanisms for detecting
symmetry and incidentally choose on the basis of overall condition or performance. For
example, in Japanese scorpionflies (Panorpa japonica), male pheromonal attractiveness
correlates with wing symmetry (Thornhill 1992c). These results could suggest that some
males have both more symmetrical wings and more attractive pheromones, and that females
have mechanisms for detecting these differences. If so, however, after several generations
all males would be highly symmetrical and attractive. Alternatively, since these insects
must fly to scavenge dead arthropods for food, and since dietary substances may be
included in pheromones, there is a functional link between wing symmetry, foraging
success, and sexual attractiveness. Thus, females could choose not on the basis of a single
trait or a small suite of traits, but on the basis of the performance of the entire male
organism. Unlike choice on the basis of a particular character (which would quickly
exhaust that character’s heritable variation), choice on the basis of performance require no
complex or special apparatus for assessing male quality. Females exercising performancebased choice give their offspring access to any genetic advantages of their mate; those
208
requiring male post-zygotic investment may receive superior resources as well (See
Heywood 1989, Hoelzer 1989, and Price et al. 1993).
Thus, considering phenotypic symmetry per se as an indicator trait may be an
oversimplification; symmetry may instead be a correlate of factors that influence mating
success. For highly asymmetrical individuals, lack of mating success may stem from
impaired performance of some behavior necessary for survival or mating (Liggett et al.
1993, Møller and Pomiankowski 1994, Møller 1997, Møller and Swaddle 1997). For
example, the enhanced mating success of symmetrical flying insects or birds (Harvery and
Walsh 1993, Møller 1994, MacLachlan and Cant 1995, Møller 1996) could be explained
by aerodynamics. Instead of choosing the most symmetrical males, females might be likely
to mate with the best fliers, most, but not all, of whom will incidentally be symmetrical;
this is a form of “indirect” mate choice (Chapter 1). Symmetry may also contribute to the
ease with which sexual ornaments are recognized (Johnstone 1994, Enquist and Arak
1994), so that some sexual ornaments may be compound characters whose effectiveness
depends both on the symmetry and on the extremity of the display (see Møller 1990, Møller
and Höglund 1991, Møller 1992, Møller 1993c, d, e, Møller and Pomiankowski 1993,
1994). To the extent that symmetry enhances performance, selection for performance will
favor mechanisms that enforce symmetry (Clarke 1994, Møller and Swaddle 1997). As
long as females make relative comparisons, and choose only on the basis of direct or
indirect male competitive interactions, variation in male quality is inexhaustible, and mate
choice on the basis of performance is possible because males could use any means at their
disposal to enhance their competitive abilities (this hypothesis is similar to some solutions
of the “lek paradox;” see Pomiankowsi and Møller 1995, Hill 1994).
Mate choice and symmetry in periodical cicadas
Magicicada mating aggregations and complex courtship suggest a long history of
sexual selection. Alexander (1975) proposed that Magicicada male choruses are non209
resource based leks, and that females restricting mating activities to the choruses receive
some benefit unobtainable by mating elsewhere. Males are not known to supply material
benefits or mating gifts, so potential benefits to females choosing mates are limited to (1) a
superior genetic complement for offspring (genetic benefits) or (2) reduced costs from
choosing mates most able to appear at a time and location congruent with female interests
(phenotypic benefits). Thus, whether females choose on the basis of genetic or phenotypic
benefits, Magicicada seem plausible candidates for species in which male performance
attributes, affected by symmetry, influence mating success. Because it is not yet known
what females gain by visiting male aggregations, a study of mate discrimination and
fluctuating asymmetry stands the greatest chance of finding any relationship if it makes use
of characters that could be important in both direct and indirect mate choice (Chapter 1).
The absence of any relationship between symmetry and male mating success would be
evidence that females do not choose mates on the basis of symmetry or any of its
correlates.
Magicicada are highly mobile insects; males engage in acoustical courtship,
producing sound with a pair of ribbed organs called tymbals. Sexually receptive females
join male singing aggregations and produce timed visual and acoustical receptivity signals
in reponse to the calls of a nearby male (Chapter 5, Cooley and Marshall ms.). A male
perceiving such signals will court, approach, mount, and copulate with the signaling
female. Courting males face severe competition from other nearby males, who seek out the
signaling female, so this mating system puts a premium on males’ ability to fly to choruses,
call and court females within the choruses, walk to receptive females, and be more effective
at these tasks than their competitors. Thus, wing, tymbal, and leg asymmetries are likely
candidates for having effects on male mating success.
210
Methods and Results
I studied Brood II Magicicada in a large open field near Horsepen Lake State
Wildlife Management Area, Buckingham County, VA, in May 1996. Vegetation in the
field consisted primarily of oak, maple, and tulip tree stump sprouts from logging
approximately two years prior; the surrounding forest contained primarily oaks and maples,
interspersed with patches of planted pines. The morning after their final molt, newly
emerged cicadas (“teneral” cicadas) are readily identifiable by their dull color, soft bodies,
and by their low positions in the vegetation. I collected unmated, teneral female cicadas
early in the mornings several days before beginning experiments and allowed them to
mature in single-sex storage cages made by enclosing living vegetation in 200 liter
hardware cloth cages. Immediately prior to starting the experiments, I captured active,
chorusing males.
I erected two large, cylindrical, 3000 liter white nylon tulle cages over stump
sprouts, trimming vegetation and creating access openings so that all parts of the cages
could be reached and observed with minimum perturbation. I placed 10 mature females
and 20 recently captured males into each cage and carefully examined them at least once per
hour. Because normal matings last longer than two hours (Chapter 4), I am confident that
no matings were unobserved. Sexually receptive females signal to calling males (Chapter
5), causing all males in the vicinity to rush to find and mate with the signaling female.
Therefore, the experimental design of this study probably tends to encourage male-male
scramble competition and accentuate quality differences among males, although the
densities in these cages were not unnaturally high (Chapter 2). I removed all mating pairs
upon first observation and immediately preserved them in 70% ethanol. When all females
had been removed, or at the end of the day, the remaining, unpaired males were preserved
in a separately marked jar. The experiment was repeated the following day using different
cicadas. I collected a total of 27 mating pairs; three unmated male cicadas were lost, and all
211
other cicadas were accounted for. Each preserved male was given a numerical code that,
without a key, provided no information about mating status. I investigated three classes of
asymmetries and their relationships to mating success: (1) Left-right character size
asymmetries, (2) Asymmetries in tymbal rib number, and (3) The presence or absence of
anomalous wing vein characters.
Left-right size asymmetries
Methods and results. A trained technician, unaware of the design or purpose
of the study (see Houle 1998), made measurements of the right and left sides of three
forewing characters, two hindwing characters, and two leg characters (Fig. 7.1). For
measurements, the technician used a calibrated ocular micrometer mounted in the eyepiece
of a Zeiss binocular dissecting microscope and clamped each wing or leg flat between two
pieces of microscope slide glass using a homemade holding jig (Appendix I). All
measurements on each cicada were repeated four times on separate occasions, and I
recoded the specimens so that the technician was unaware of remeasuring the same
specimens. I instructed the technician to skip measurements on damaged cicadas.
For each side of each character, I excluded the most extreme value of the four
repetitions, making the measurements conservative estimates of asymmetry. For each
character, individuals for which there were not three remaining measurements were
eliminated from consideration. I analyzed the repeated measurements using an ANOVA
design to ensure that variation came from actual asymmetry differences, not from
measurement repetitions (see Palmer and Strobeck 1986, Swaddle et al. 1994, Palmer
1994). Mixed-model ANOVAs had the following structure: Factors were repetition (1-3)
and side (left/right), and the dependent variables were the seven characters. In only one
case (hindwing character H40) did repetition explain a significant amount of the variation in
the data. Kolmogorov-Smirnov tests confirmed this relationship, so this character was
discarded (Table 7.1).
212
Using the average values calculated above, I calculated percent asymmetry for each
character in each individual according to the following formula:
2*
Right − Left
Right + Left
= Scaled % Asymmetry
Because this index of asymmetry is scaled to the size of the individual, it allows
comparisons of the relative levels of asymmetry among individuals. I used percent
asymmetry values for all subsequent analyses and comparisons discussed below (Appendix
J).
I calculated average sizes for the right and left sides of the six remaining characters
for each specimen. To determine whether any asymmetry was truly fluctuating asymmetry
or whether it represented species-typical right/left morphological asymmetry (e.g., the right
side is typically different from the left), I used a Lilliefors’ test of normality of scaled,
signed asymmetries and Sokal and Rohlf’s test of skew and kurtosis significance (Table
7.2). Only characters F15 and FFEM pass both these tests; H46 passes the kurtosis/skew
test, but not the Lilliefors’ test. These analyses indicate that some of the characters do not
exhibit normal distributions typical of fluctuating asymmetry (Table 7.2). I also calculated
sizes for each character by averaging all right/left values, to allow evaluation of any
relationship between size and mating success. The numbers of specimens used in the
analysis for each character are noted in Table 7.3.
I examined correlations among asymmetries of the different characters by
constructing a Pearson pairwise correlation matrix with Bonferroni corrections (Table 7.4).
Wing character asymmetries tend to be correlated, but correlation patterns do not
immediately suggest uniformity of expressed asymmetry across all characters within an
individual. Correlations among character sizes are stronger (i.e. there is some indication of
213
allometric relationships among characters; Table 7.5), but there is no relationship between
character size and degree of asymmetry (Table 7.6).
For each individual, I examined the relationship between both the symmetry and
average size of each independent character and mating status, using Kruskal-Wallis OneWay ANOVAs (Table 7.7). Mated males had significantly more symmetrical fore-femurs
(Character FFEM) than unmated males. I then calculated a cumulative asymmetry score for
each individual by summing the unsigned (absolute-value) asymmetries of all characters.
Any individuals that lacked symmetry measurements for all of the characters were not used
in this analysis. I calculated an alternative cumulative asymmetry score by summing only
the unsigned asymmetries of the two characters exhibiting fluctuating asymmetry, F15 and
FFEM. These cumulative measurements of asymmetry should tend to (1) expose genomic
effects by allowing repeated measures of developmental stability; and (2) maximize any
overall asymmetry differences among individuals. Thus, they should be sensitive assays
for whether asymmetries of any combinations of characters influence mating ability (cf.
Møller and Swaddle 1997 p.2). Using Kruskal-Wallis One-Way ANOVAs, I found no
relationship between cumulative asymmetry and mating status, although cumulative
fluctuating asymmetry scores for mated males were significantly lower than for unmated
males (Fig. 7.2), and was probably influenced by character FFEM.
Tymbal rib asymmetries
Methods and results. The number of tymbal ribs in M. septendecim varies from 11 to
13, and rarely, the right and left tymbals have different numbers of ribs. On each
numerically coded specimen, the number of ribs on both tymbals was counted twice.
Because tymbal ribs sometimes branch and merge, only sclerotizations that remained
separate from any other structures for more than 3/4 of their total length, and only
structures that formed an acute angle with the nearest sclerotization were counted as ribs.
214
There was no relationship between tymbal rib count asymmetries and mating success.
(Table 7.8).
Wing venation asymmetries
Methods and results. For the same reasons that asymmetrical males may be
inferior mates, females may benefit by avoiding males with atypical wings. Because wing
venation in Magicicada septendecim can be highly variable, the technician scored each
cicada for the presence/absence of extra wing veins and whether the anomalies were
bilaterally symmetrical. These venation anomalies are distinct from “wing crumpling,” or
incomplete expansion of the wings at the final molt, noted in some crowded populations
(White et al. 1979). None of the cicadas used in this study had crumpled wings. Sixteen
of the 77 males used in this study had unusual or atypical wing venation, either in the form
of extra or missing wing veins. Four of the sixteen mated, while twelve did not. To
determine whether mated males were underrepresented in the group of abnormal-winged
males, I created a computer algorithm that resampled the data 10,000 times. Each iteration
of the algorithm randomly selected a 16 male sample from the 27 mated and 50 unmated
males in the study and then tabulated the number of mated males in the sample. After
10,000 iterations, the algorithm constructed percentiles, which describe the likelihood of
finding only four mated males in any 16- member random sample from the original data.
Percentiles are reported in Table 7.9, and the code for the resampling algorithm is included
in Appendix K. The number of mated and unmated males among the atypically-veined
males was not significantly different from that expected by chance.
Discussion
Magicicada seem ideal for mate choice studies because their mating aggregations,
complex courtship, and male-male competitive behaviors suggest a history of sexual
selection in which females discriminate against some potential mates. Male Magicicada
215
septendecim likely provide females no resources (other than sperm) necessary for
reproduction, because females prevented from mating eventually oviposit normalappearing, unfertilized eggs (pers. obs.; Graham and Cochran 1954); thus, females’
potential benefits from discriminating among mating partners would necessarily be genetic
or involve phenotypic benefits other than donated resources.
Mate rapprochement behaviors in Magicicada require males to fly, walk, and fend
off competitors. Although slight wing and hindleg asymmetries should affect general
mobility, they evidently do not affect the final outcome of courtship or correlate with
mating success. Overall asymmetry was also not a predictor of mating success, ruling out
general effects of genotypic or phenotypic quality. However, foreleg symmetry correlates
with male mating success, suggesting some special functional significance of this character.
Walking cicadas appear to use their forelegs to probe surfaces in front of them before
venturing onto them. Cicadas with damaged or missing foreleg claws have reduced
mobility because they tend to paw at the surface in front of them and remain motionless
(pers. obs.); less extreme damage or deformation could have less noticeable effects. The
conditions in these experiments probably accentuated male-male scramble competition:
Several males could simultaneously perceive a female’s signals. If slight foreleg
asymmetries lead to slight reductions in mobility or maneuverability, then males with the
most symmetrical forelegs should outcompete less symmetrical cicadas as they race to make
contact with signaling females. If such differences are relevant under natural conditions,
then male-female rapprochement duets and male-male competition would lead to indirect
mate choice (Chapter 1), such that males with the most symmetrical forelegs would be the
most successful. Although this study suggests that “indirect” mate choice mechanisms
cause females to be more likely to mate males with symmetrical forelegs, there is no
evidence that females have mechanisms specifically evolved for favoring symmetrical
males. One way to test whether differential male success in these experiments results from
differential male abilities independent of any female actions would be to repeat the
216
experiments, replacing the females with an experimenter making artificial signals (Chapter
5) and noting whether the males first reaching the artificial stimulus have the most
symmetrical forelegs.
Studies, such as the present one, relying on naturally occurring variation to evaluate
the relationship between symmetry and mating success, assume either (1) a positive
correlation between symmetry and mating success; or (2) that the sample of individuals
used accurately reflects potential mating partners. The first assumption is not valid for
most insects, because it requires females to make use of “best-of-n” (BN) mate choice
criteria, in which higher quality mates receive incrementally more matings (see Alexander et
al. 1997, Janetos 1980, Chapter 1). Insects’ life history patterns and limited opportunities
for learning make “threshold” choice mechanisms, in which all potential mates meeting
certain criteria are equally acceptable, a more likely alternative (Alexander et al. 1997,
Chapter 1). Females exercising threshold choice discriminate among potential mates only if
unacceptable ones are present, calling into question assumption (2). There is no evidence
that the present study included a class of unacceptable males, because extreme males were
distributed proportionally among those that did and those that did not mate. The males
used in this study were collected from those engaged in mate search behaviors, the pool
from which unconfined females normally draw mates. Consequently, from the present
study, there is no evidence that periodical cicada females have specially- evolved
mechanisms for acquiring especially symmetrical mates from the pool available to them.
Just as foreleg asymmetries seem to influence male success within small cages,
other gross asymmetries affecting mobility could prevent males from entering choruses and
encountering females. In most cicada emergences a small class of males has deformed
wings from faulty ecdysis; these males have impaired flight and/or signaling capabilities.
Comparing symmetry characters of grossly misshapen and normal males would be difficult
at best, because deformed males lack the landmarks necessary for accurate measurements.
217
Such a comparison might also be trivial, because deformed males are unable to join treetop
chorusing centers where they are most likely to encounter receptive females and thus
undoubtedly receive, on average, fewer matings than normal males. Again, no specifically
evolved mate choice mechanism need be invoked to explain such discrimination. Because
the present study did not include deformed males and confined males in proximity to
females, its conclusions are limited to understanding how symmetry affects male mating
success within mating aggregations and it provides no information on the differential
abilities of males to locate and enter these aggregations.
Given asymmetry’s multiple causes and the diverse interests of individuals
choosing mates, there is little reason to expect symmetry to be a universal mate choice
criterion. Although we have no reason to expect that phenotypic symmetry is an automatic
indicator of quality, the qualities for which symmetry is a proxy may lead to predictable
gains for choosy individuals. These predictions lead to testable hypotheses addressing
how females might, actively or incidentally, create or participate in situations favoring
symmetrical males:
1. Unless females have specially evolved mechanisms for directly assessing
symmetry, females must choose males on the basis of functional correlates
of symmetry.
2. For those species without male-donated resources, male parental care, or other
phenotypic benefits available from differential mating, females can benefit
only from choosing among males whose asymmetry differences have some
genetic basis.
3. For those species where males provide access to resources, females can benefit
from choosing if there is a genetic basis to symmetry or a functional link
between symmetry and male performance (which may have heritable
components), measured as ability to gather and contribute resources.
These statements lead to testable hypotheses linking sexual selection and asymmetry
theory. These testable hypotheses underscore the importance of explicitly considering the
possible selective advantages of choosiness and the applicability of specific patterns of
asymmetry to hypothetical evolved mechanisms for mate choice in the basis of symmetry.
218
Literature Cited
Alexander, R. D., 1975. Natural selection and specialized chorusing behavior in chorusing
insects. Pp. 35-77 in: D. Pimentel, ed. Insects, Science, and Society. (Academic
Press: New York).
Alexander, R. D., D. C. Marshall, and J. R. Cooley, 1997. Evolutionary Perspectives on
Insect Mating. Pp. 1- 31 in B. Crespi, J. Choe, eds. The evolution of mating
systems in insects and arachnids. Cambridge University Press.
Andersson, M. 1982. Sexual selection, natural selection, and quality advertisement. Biol.
J. Linn. Soc. 17: 375- 393.
Borgia, G. 1979. Sexual selection and the evolution of mating systems. Pp. 19- 80 in: M.
S. Blum and N. A. Blum, eds. Sexual selection and reproductive competition in
insects 19-80. (Academic Press: New York).
Borror, D. J., C. A. Triplehorn, and N. F. Johnson. 1989. An introduction to the study of
insects. (Saunders: Philadelphia).
Clarke, G. M. 1993. Fluctuating asymmetry of invertebrate populations as a biological
indicator of environmental quality. Environmental Pollution 82: 207-11.
Clarke, G. M. 1994. The genetic basis of developmental stability. I. Relationships
between stability, heterozygosity, and genomic coadaptation. Pp. 17-25 in: T. A.
Markow, ed. Developmental Instability: Its origins and Evolutionary Implications.
(Kluwer, The Netherlands).
Clarke, G. M. 1998. Developmental stability and fitness: The evidence is not quite so clear.
Am. Nat. 152: 762- 766.
Emlen, J. M., D. C. Freeman, and J. H. Graham. 1994. Nonlinear growth dynamics and
the origin of fluctuating asymmetry. Pp. 79-96 in: T. A. Markow, ed.
Developmental Instability: Its origins and Evolutionary Implications. (Kluwer, The
Netherlands).
Enquist, M., and A. Arak. 1994. Symmetry, beauty, and evolution. Nature (London) 372:
169-172.
Gangstead, S. W., R. Thornhill, and R. A. Yeo. 1994. Facial attractiveness,
developmental stability, and fluctuating asymmetry. Ethology and Sociobiology
15: 73-85.
Graham, C., and A. B. Cochran. 1954. The periodical cicada in Maryland in 1953. J.
Econ. Ent. 47:242-244.
Halliday, T. R. 1978. Sexual selection and mate choice. Pp. 180-231 in: J. R. Krebs and
N. B. Davies, eds. Behavioral Ecology: An Evolutionary Approach. (Blackwell:
Oxford).
219
Hamilton, W. D., and M. Zuk. 1982. Heritable true fitness and bright birds: A role for
parasites? Science (Wash. D. C.) 218: 384-387.
Harvey, I. F., and K. J. Walsh. 1993. Fluctuating asymmetry and lifetime mating success
are correlated in males of the damselfly Coenagrion puella (Odonata:
Coenagrionidae). Ecological Entomology 18:198-202.
Heywood, J. S. 1989. Sexual selection by the handicap mechanism. Evolution 43: 13871397.
Hill, G. E. 1994. Trait elaboration via adaptive mate choice: sexual conflict in the evolution
of signals of male quality. Ethology, Ecology, and Evolution 6: 351- 370.
Hoelzer, G. A. 1989. The good parent process of sexual selection. Anim. Behav. 38:
1067- 1078.
Houle, D. 1997. Comment on “A meta-analysis of the heritability of developmental
stability” by Møller and Thornhill. J. Evol. Biol. 10: 17-20.
Houle, D. 1998. High enthusiasm and low R- squared. Evolution 52: 1872- 1876.
Janetos, A. C., 1980. Strategies of female mate choice: A theoretical analysis. Behav.
Ecol. Sociobiol. 7: 107-112.
Johnstone, R. A. 1994. Female preferences for symmetrical males as a by-product of
selection for mate recognition. Nature (London): 372: 172-175.
Kodric-Brown, A., and J. H. Brown. 1984. Truth in advertising: The kinds of traits
favoured by sexual selection. Am. Nat. 124: 309-323.
Leamy, L. 1997. Is developmental stability heritable? J. Evol. Biol. 10: 21-29.
Lerner, I. M. 1954. Genetic Homeostasis. (MacMillan: New York).
Liggett, A. C., I. F. Harvey, and J. T. Manning. 1993. Fluctuating asymmetry in
Scatophaga stercoraria L.: Successful males are more symmetrical. Anim. Behav.
45: 1041- 1043.
Malyon, C. and S. Healy. 1994. Fluctuating asymmetry in antlers of fallow deer, Dama
dama, indicates dominance. Anim. Behav. 48: 248- 250.
Markow, T. A., and G. M. Clarke. 1997. Meta-analysis of the heritability of
developmental stability: A giant step backward. J. Evol. Biol. 10: 31-37.
McLachlan, A., and M. Cant. 1995. Small males are more symmetrical: Mating success in
the midge Chironomous plumosus L. (Diptera: Chironomidae) Anim. Behav. 50:
841- 846.
Mitton, J. R. 1994. Enzyme heterozygosity, metabolism, and developmental stability. Pp.
49-67 in: T. A. Markow, ed. Developmental Instability: Its origins and
Evolutionary Implications. (Kluwer, The Netherlands).
Møller, A. P. 1990. Fluctuating asymmetry in male sexual ornaments may reliably reveal
male quality. Anim. Behav. 40: 1185-1187.
220
Møller, A. P. 1992. Female swallow preference for symmetrical male sexual ornaments.
Nature 357: 238-240.
Møller, A. P. 1993a. Fluctuating asymmetry. Nature (London): 363: 217
Møller, A. P. 1993b. Female preference for apparently symmetrical male sexual ornaments
in the barn swallow Hirundo rustica. Behav. Ecol. Sociobiol. 32: 371- 376.
Møller, A. P. 1993c. Patterns of fluctuating asymmetry in sexual ornaments predict female
choice. J. Evol. Biol . 6: 481- 491.
Møller, A. P. 1993d. Developmental stability, sexual selection and speciation. J. Evol.
Biol. 6: 493- 509.
Møller, A. P. 1993e. Developmental stability, sexual selection, and the evolution of
secondary sexual characters. Etologia. 3: 199- 208.
Møller, A. P. 1994. Sexual selection in the barn swallow (Hirundo rustica ). IV. Patterns
of fluctuating asymmetry and selection against asymmetry. Evolution 48: 658670.
Møller, A. P. 1995. Patterns of fluctuating asymmetry in sexual ornaments of birds from
marginal and central populations. Am. Nat. 145: 316- 27.
Møller, A. P. 1996. Sexual selection, viability selection, and developmental stability in the
domestic fly, Musca domestica. Evolution 50: 746- 752.
Møller, A. P. 1997. Developmental stability and fitness: A review. Am. Nat. 149: 916932.
Møller, A. P., and J. Höglund. 1991. Patterns of fluctuating asymmetry in avian feather
ornaments: Implications for models of sexual selection. Proc. R. Soc. Lond. B.:
245: 1-5.
Møller, A. P., and A. Pomiankowski. 1993. Fluctuating asymmetry and sexual selection.
Genetica 89: 267- 279.
Møller, A. P., and A. Pomiankowski. 1994. Fluctuating asymmetry and sexual selection.
Pp. 269-281 in: T. A. Markow, ed. Developmental Instability: Its origins and
Evolutionary Implications. (Kluwer, The Netherlands).
Møller, A. P. , M. Soler, and R. Thornhill. 1995. Breast asymmetry, sexual selection,
and human reproductive success. Ethology and Sociobiology 16: 207-219.
Møller, A. P. and J. P. Swaddle. 1997. Asymmery, Developmental stability, and
Evolution. (Oxford University Press).
Møller, A. P., and R. Thornhill. 1997a. A meta-analysis of the heritability of
developmental stability. J. Evol. Biol. 10: 1-16.
Møller, A. P., and R. Thornhill. 1997b. Developmental instability is inheritable. J. Evol.
Biol. 10: 1-16.
221
Møller, A. P. and C. Zamora- Muñoz. 1997. Antennal asymmetry and sexual selection in
a cerambycid beetle. Anim. Behav. 54: 1509- 1515.
Oakes, E. J. and P. Barnard. 1994. Fluctuating asymmetry and mate choice in paradise
whydahs, Vidua paradisaea: an experimental manipulation. Anim. Behav. 48: 93743
Palmer, A. R. 1994. Fluctuating asymmetry analyses: A primer. Pp. 335-364 in: T. A.
Markow, ed. Developmental Instability: Its origins and Evolutionary Implications.
(Kluwer, The Netherlands).
Palmer, A. R. 1996. Waltzing with Asymmetry. BioScience 46: 518- 531.
Palmer, A. R., C. Strobeck, and A. K. Chippindale. 1994. Bilateral variation and the
evolutionary origin of macroscopic asymmetries. Pp. 203-220 in: T. A. Markow,
ed. Developmental Instability: Its origins and Evolutionary Implications. (Kluwer,
The Netherlands).
Palmer, A. R. and C. Strobeck 1986. Fluctuating asymmetry: Measurement, analysis,
patterns. Ann. Rev. Ecol. Syst. 17: 391- 421.
Palmer, A. R., and C. Strobeck. 1997. Fluctuating asymmetry and developmental stability:
heritability of observable variation vs. heritability of inferred cause. J. Evol. Biol.
10: 39-49.
Parker, G. A. 1982. Phenotype-limited evolutionarily stable strategies. In: Kings’ College
Sociobiology Group, eds., Current Problems in Sociobiology 173- 202.
(Cambridge: Cambridge University Press).
Parsons, P. A. 1961. Fly size, emergence time, and sternopleural chaeta number in
Drosophila. Heredity 16: 455- 473.
Parsons, P. A. 1990. Fluctuating asymmetry: An epigenetic measure of stress. Biol. Rev.
65: 131- 145.
Parsons, P. A. 1994. Developmental variability and the limits of adaptation: Interactions
with stress. Pp. 247-255 in: T. A. Markow, ed. Developmental Instability: Its
origins and Evolutionary Implications. (Kluwer, The Netherlands).
Polak, M. 1994. Parasites increase fluctuating asymmetry of male Drosophila
nigrospiracula: Implications for sexual selection. Pp. 257-267 in: T. A. Markow,
ed. Developmental Instability: Its origins and Evolutionary Implications. (Kluwer,
The Netherlands).
Pomiankowski, A. 1997. Genetic variation in fluctuating asymmetry. J. Evol. Biol. 10:
51-55.
Pomiankowski, A., and A. P. Møller. 1995. A resolution of the lek paradox. Proc. Roy.
Soc. Lond. B. 260: 21-29.
Price, T. D., D. Schluter, and N. E. Heckman. 1993. Sexual selection when the female
directly benefits. Biol. J. Linn. Soc. 48: 187- 211.
222
Simon, C. 1983. Morphological differentiation in wing venation among broods of 13- and
17- year periodical cicadas. Evolution 37: 104- 115.
Swaddle, J. P. 1996. Reproductive success and symmetry in zebra finches. Anim.
Behav. 51: 203-210.
Swaddle, J. P. 1997. On the heritability of developmental stability. J. Evol. Biol. 10: 5761.
Swaddle, J. P., M. S. Witter, and I. C. Cuthill. 1994. The analysis of fluctuating
asymmetry. Anim. Behav. 48: 986-989.
Thornhill, R. 1992a. Fluctuating asymmetry and the mating system of the Japanese
scorpionfly, Panorpa japonica. Anim. Behav. 44: 867-879.
Thornhill, R. 1992b. Fluctuating asymmetry, interspecific aggression, and male mating
tactics in two species of Japanese scorpionflies. Behav. Ecol. Sociobiol. 30: 357363.
Thornhill. R. 1992c. Female preference for the pheromone of males with low fluctuating
asymmetry in the Japanese scorpionfly (Panorpa japonica : Mecoptera) Behavioral
Ecology 3: 277- 283.
Tomkins, J. L., and L. W. Simmons. 1998. Female choice and manipulations of forceps
size and symmetry in the earwig Forficula auricularia L. Anim. Behav. 56: 347356.
Van Valen, L. 1962. A study of fluctuating asymmetry. Evolution 16: 125-142.
Waddington, C. H. 1957. Strategy of the Genes. (Oliver and Boyd: London).
West-Eberhard, M. J. 1979. Sexual selection, social competition, and evolution. Proc.
Am. Phil. Soc. 123: 222-234.
White J. 1973. Viable hybrid young from crossmated periodical cicadas. Ecology 54: 573580.
White, J., M. Lloyd, and J. H. Zar. 1979. Faulty eclosion in crowded suburban periodical
cicadas: Populations out of control. Ecology 60: 305-315.
Whitlock, M. 1996. The heritability of fluctuating asymmetry and the genetic control of
developmental stability. Proc. Roy. Soc. Lond. B. 263: 849-854.
Whitlock, M. C., and K. Fowler. 1997. The instability of studies of instability. J. Evol.
Biol. 10: 63-67.
Woods, R. E., M. J. Hercus, and A. A. Hoffmann. 1998. Estimating the heritability of
fluctuating asymmetry in field Drosophila. Evolution 52: 816-824.
Zahavi, A. 1977. The cost of honesty. J. Theor. Biol. 67: 603-605.
223
Table 7.1. P- values for tests of measurement error (significant
differences among repetitions of the same measurement) and handedness
(significant differences between right and left sides of an individual). For
Smirnov test, a character is considered to fail if any of the repetitions was
significantly different from any other.
Character
F14
F15
F27
H40
H46
FFem
Hfem
ANOVA
Repetition
Side
0.312
0.994
0.097
0.651
0.413
0.598
0.010
0.615
0.186
0.186
0.793
0.133
0.496
0.391
Smirnov
Repetition
Pass
Pass
Pass
Fail
Pass
Pass
Pass
224
Table 7.2. Tests for normality of signed, scaled asymmetries.
(Skew)
(Kurtosis)
Character
Lilliefor’s P Sokal G1
Sokal G2
Conclusion
F14
0.001
Pass
Fail
Not FA
F15
0.566
Pass
Pass
FA
F27
0.001
Fail
Fail
Not FA
H46
0.010
Pass
Pass
Possibly FA
FFem
0.067
Pass
Pass
FA
Hfem
0.003
Fail
Pass
Not FA
Cumulative 0.000
Fail*
Fail*
Not FA
Cumul. FA 0.402
Fail*
Fail*
Not FA
*Since these are cumulative measures of absolute deviations from symmetry, they
have no negative values; data were standardized to allow tests.
225
Table 7.3. Numbers of cicadas used in analysis
of symmetry and mating success.
Character
F14
F15
F27
H40
H46
FFEM
HFEM
Cumulative
Mated
males used
25
23
22
26
24
24
24
22
Unmated
males used
43
47
43
40
42
45
40
40
226
227
F14
1.000
0.411*
0.180
-0.068
0.258
0.054
0.016
1.000
0.483*
-0.257
-0.027
0.069
0.205
F15
1.000
-0.121
-0.114
-0.042
0.211
F27
* Indicates Bonferroni corrected significance P ≤ 0.05.
F14
F15
F27
(H40)
H46
FFEM
HFEM
1.000
-0.098
-0.121
-0.169
(H40)
H46
1.000
-0.255
0.111
Table 7.4. Pearson pairwise correlation matrix for character asymmetry.
1.000
0.073
FFEM
1.000
HFEM
228
F14
1.000
0.715
0.600
0.636
0.702
0.529
0.515
1.000
0.613
0.586
0.780
0.680
0.663
F15
1.000
0.581
0.636
0.561
0.557
F27
1.000
0.630
0.429
0.600
(H40)
*All values significant, Bonferroni corrected significance P ≤ 0.05.
F14
F15
F27
(H40)
H46
FFEM
HFEM
Table 7.5. Pearson pairwise correlation matrix for character size.
1.000
0.511
0.664
H46
1.000
0.752
FFEM
1.000
HFEM
229
0.031*
-0.141
-0.008
-0.069
0.003
-0.111
-0.184
0.089*
0.073
0.092
0.204
0.030
0.092
F15
-0.021*
0.078
-0.064
0.105
0.011
F27
0.045*
0.055
0.204
0.064
(H40)
Symmetry
* Indicates Bonferroni corrected significance P ≤ 0.05.
F14
F15
F27
(H40)
H46
FFEM
HFEM
Size
F14
-0.116*
-0.108
-0.045
H46
-0.084*
-0.118
FFEM
-0.106*
HFEM
Table 7.6. Pearson pairwise correlation matrix with Bonferroni probabilities for character sizes and asymmetry.
Table 7.7. Kruskal-Wallis One-Way ANOVA analysis of mating status and signed,
scaled asymmetry (left columns) and average character size (right columns). Mated and
unmated males differed only in levels of asymmetry for character FFEM.
Symmetry
U
P
Size
U
P
≤ 0.685
≤ 0.028*
514
673
≤ 0.367
≤ 0.979
287
≤ 0.838
Characters not meeting definition of FA (Table 7.2):
F14
613
≤ 0.508
747
F27
733
≤ 0.539
674
HFEM
495
≤ 0.112
613
≤ 0.442
≤ 0.991
≤ 0.789
FA characters (Table 7.2):
F15
479
FFEM
880
Possibly FA character (Table 7.2):
H46
266
≤ 0.531
230
Table 7.8. Results of tymbal rib count asymmetry analysis.
Asymmetrical
Wings
Symmetrical
Wings
TOTAL
2
25
27
Did not mate 5
45
50
TOTAL
70
77
Did mate
7
Fisher exact test (two-tail)
P < 1.000
231
Table 7.9. Results of resampling algorithm to evaluate whether males with wing
asymmetries were as likely to mate as normal males. The algorithm draws a random
sample of 16 males from a population of 27 mated and 50 unmated males and determines
whether the mating frequency in the random sample matches the mating frequency of the 16
males with asymmetrical wing venation. These results demonstrate that the mating
frequency (4/16) in the observed sample of 16 asymmetrical males is indistinguishable
from the mating frequency in a sample drawn randomly from the cicadas in this study.
Number of
“mated” males
in sample
0
≤1
≤2
≤3
≤4*
≤5
≤6
≤7
≤8
≤9
≤10
≤11
≤12
≤13
≤14
≤15
≤16
Cumulative
frequency
in 10,000
iterations
0.0008
0.01
0.043
0.12
0.27
0.47
0.67
0.83
0.93
0.98
0.99
1
1
1
1
1
1
*Actual value among males with asymmetrical wing venation.
232
F 14
F 27
F 15
Forewing
H 40
H 46
Hindwing
F FEM
H FEM
Foreleg
Hindleg
Figure 7.1. Characters used in analysis of phenotypic symmetry.
F 14: Forewing character #14 from Simon 1983.
F 15: Forewing character #15 from Simon 1983.
F 27: Forewing character #27 from Simon 1983.
H 40: Hindwing character #40 from Simon 1983.
H 46: Hindwing character #46 from Simon 1983.
F fem: Fore- femur length.
H fem: Hind femur length.
Wing diagram adapted from Simon 1983, leg structure diagrams adapted
from White 1973 and Borror et al. 1989.
233
234
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
Mated Unmated
Cumulative Asymmetry
Mated Unmated
Cumulative FA
Figure 7.2. Comparison of cumulative asymmetry scores for mated and unmated males. Left columns: Cumulative
asymmetry (sum of unsigned asymmetries for each character); no significant difference between mated and unmated males
(Kruskal-Wallis One-Way ANOVA;U = 198, P ≤ 0.655). Right columns: Cumulative fluctuating asymmetry (sum of unsigned
asymmetries for characters F 15 and Fore-Femur); mated males have significantly smaller cumulative FA scores than
unmated males (Kruskal-Wallis One-Way ANOVA;U = 667, P ≤ 0.022). Bars are standard errors.
Scaled, unsigned asymmetry
CHAPTER VIII
THE GEOGRAPHIC DISTRIBUTION AND MATING BEHAVIORS OF
TWO CICADA SPECIES (OKANAGANA SPP.) IN NORTHEAST
MICHIGAN
Abstract
I present the results of a 3 –year biogeographic survey of Okanagana canadensis
and O. rimosa in northeastern Michigan. The abundance of these species varied from year
to year, although in all years some individuals of each were present within the survey area.
In some specific locations, O. rimosa were present in all years, confirming reported
observations that this species is not strictly periodical, even on a microgeographic scale. I
also describe both species’ sexual behaviors, which involve stationary, calling males and
approaching females. These behaviors are in contrast to those of Magicicada spp., the
periodical cicadas of North America, in which males actively search for sexually receptive
females. I discuss how behaviors such as those found in Okanagana could have evolved
into the unusual mating behaviors typical of Magicicada.
235
Introduction
Okanagana canadensis Provancher, and the closely related O. rimosa (Say) are
cicadas common in northern Michigan. Life cycles of these species have been reported as
four years (Kirby and McPhee 1963) 8-10 years (Huber et al. 1979), or, based upon
studies of nymphal development, 9 years (Soper Delyzer, and Smith 1976). Okanagana
species are not periodic, because in localities where they are known to occur, some adults
can be found in almost any year, although the species’ abundance varies annually.
This study involves two parts: 1. Biogeographic survey of O. canadensis and O.
rimosa emergences in northern Michigan, 1996- 1998 and 2. Behavioral studies of O.
canadensis and O. rimosa. The biogeographic survey is part of a long-term effort to
provide a permanent record of detailed distributional information and resolve ambiguities
about these species’ life cycles. The behavioral studies complement ongoing research on
the behaviors of other cicadas, including the periodical cicadas of North America
(Magicicada spp.). The sister-group relationship of Okanagana and Magicicada has been
inferred on the basis of gross similarity of appearance and habits1 (see Simon 1979),
although no molecular data support this relationship (Simon pers. comm.). Deciphering
the taxonomic relationships of North American cicadas is made difficult by the uniqueness
of Magicicada. The sexual behaviors of Magicicada, (in which males search for signaling
females; Chapter 5) appear unique2, while the sexual behaviors of Okanagana, (in which,
as reported here, females approach calling males) are similar to those found in other cicada
genera, such as Tibicen (Alexander 1960, Alexander and Moore 1962, unpub.
observations), or Diceroprocta (Alexander and Moore 1962). Regardless of the taxonomic
1
Both Okanagana and Magicicada lack tymbal covers and oviposit in live twigs, while members of another
large North American genus, Tibicen, have covered tymbals and oviposit in dead wood or bark. See
Appendix L.
2
However, Alexander (1960) and Alexander and Moore (1962) suggest that some of the chorusing
behaviors that seem unique to Magicicada are also found in species of the genera Okanagana, Tibicen, and
Diceroprocta in dense emergences.
236
relationships among these genera, the sexual behaviors of present-day Okanagana may be
similar to those of ancestral Magicicada, and comparisons and contrasts of the behavior and
biology of Okanagana and Magicicada allow inferences about how sexual selection shaped
the mating system of each.
Methods and Results
Biogeographic surveys
Because of cicadas’ conspicuous, species-specific calls, mapping species
distributions over large geographic areas is possible (Marshall et al. 1996, Alexander et al.
in prep.). The distributions of O. canadensis and O. rimosa were determined by driving no
more than 35 miles per hour along two lane roads with the car windows open and marking
the species present on a high quality map, using methods similar to those used for other
species of cicadas (Fig. 8.1; see Marshall et al. 1996, Alexander et al. in prep.).
Results. Surveys of O. canadensis and O. rimosa on July 16 1996 and August 1,
1996 (Fig. 8.2), July 25 1997 (Fig. 8.3), and June 20 1998 (Fig 8.4) illustrate that in
some locations, O. canadensis were present in all three years of the study, and in 1998,
there was a marked increase in the abundance of both O. canadensis and O. rimosa. The
increase in abundance in the 1998 survey is not attributable to the earlier date of the survey
because there was a decrease in local abundance of O. canadensis in some specific
sampling locations, such as at the intersection of Bridge View Road and US 2. O.
canadensis were observed primarily in cedars (occasionally aspen), with singing males
located low (usually less than 8 feet off the ground n= 20) and on the central trunk or at the
base of a large branch. O. canadensis seemed to prefer rocky, high ground in which
northern white cedar was present. In contrast, O. rimosa were often present in sandy
areas, associated with quaking aspen, maple, or oak, and seldom found on cedars. The
calling habits of O. rimosa are more variable than those of O. canadensis; male O. rimosa
237
call from a variety of locations, from grass stems to treetops, and sing from perches at the
ends of branches as well as from the bases of branches or trunks.
Upper Peninsula. On July 16, 1996, a substantial (ca. 20) emergence of O.
canadensis was discovered at the interchange of US 2 and Bridge View Road, near the
ramp from eastbound US 2 to Southbound I-75, in Mackinac County, Michigan. A search
of the area yielded no nymphal skins and no fresh oviposition scars; however, some scars
reminiscent of cicada eggnests were noted on small branches of white cedar. O. canadensis
were present at this location on July 24, 1996, July 25, 1996, August 1, 1996, July 25,
1997 and June 20, 1998.
On July 16, 1996 O. canadensis were noted at I-75 and M-134, and at I-75 and Bay
Road, and in the town of Rudyard, near where H-40 crosses a small river and railroad
tracks. They were not noted north of Rudyard on that date.
On July 17, 1996, no cicadas were noted between Rudyard and Rexford, or along
M-28 from Rudyard to Munising. However, both O. canadensis and O. rimosa were heard
at the Seney National Wildlife Refuge, Schoolcraft County. No further Okanagana were
noted along M-77 to US 2, or along US 2 from Blaney Park to Brevort. At Brevort and at
the intersection of US 2 and Gudmonson Road, dense populations of O. canadensis were
noted, with isolated patches at points in between.
The survey of Okanagana emergences continued on July 25, 1996, and the same
route as follows was revisited on August 1, 1996, July 25 1997 and June 20, 1998. East
of Rudyard on M-48, many O. canadensis were present in aspens at Munuscong Creek
before the intersection of M-48 and Fish Road. No cicadas were heard at this location in
1997 or 1998, but O. canadensis were present in 1997 and 1998 along the fork of
Munuscong Creek crossing M-48 just east of its intersection with Fish Road. In 1998, east
of the intersection of M-48 and 5- mile road, O. rimosa were almost continuously present
238
in wooded areas along M-48 until the intersection with Blair Road; O rimosa were also
present along M-48 just east of the intersection with Pennigton Road, and O. canadensis
were present at the curve where M-48 joins Fairview Road. O. rimosa were also present
in the wet areas associated with East Branch Creek. No cicadas of any species were heard
along this stretch in 1996 or 1997.
In 1996 and 1998, O canadensis were present along M-48 at Stalwart, and on high
ground just before the intersection of M-48 and Prentiss Bay Road; O. rimosa were present
in these locations in 1998. In 1998, both species were present in small numbers in wooded
areas between Prentiss Bay Road and North Caribou Drive; O. rimosa may have been
present at the curve in M-48 just before intersection of M-48 and North Caribou Drive in
1996 as well, but the only male heard stopped singing after first being heard, so the record
is not reliable.
In 1996- 1998, O. canadensis were almost continuously present along M-134 from
high ground midway between Albany Island and Cadogan Point to Whitefish Point, and
from McKay Bay though Cedarville and continuing west of Cedarville until Nun’s Creek
Road. In 1996, O. rimosa were present high in trees at Steele River. In 1998, O.
canadensis were present in great abundance along this stretch, with few forested areas
silent. O. rimosa were also occasionally present in 1998 in this location. No individuals
of either species were heard in any year from a line extending directly north from the
western edge of Search Bay west to Ponchartrain Shores, but both species were heard
between Ponchartrain Shores and I-75 in 1996. In 1998, O. canadensis were abundant
along this stretch. On August 1, 1996, both O. canadensis and O. rimosa were singing
along this stretch, whenever trees were present.
In 1997, O. canadensis were present in small numbers along H 63 between M-48
and St. Ignace and were especially abundant near the Carp River. In 1998, O. canadensis
were continuously present along this stretch, and O. rimosa were present near Planz Lake.
239
Lower peninsula. On July 16, 1996, and June 20 1998 O. rimosa were heard
singing in treetops at the University of Michigan Biological Station, Cheboygan County,
Michigan. In 1996 and 1997, few O. rimosa were present along Riggsville Road between
I-75 and Bryant Road, along Bryant Road between Riggsville Road and Douglas Lake
Road, or along Douglas Lake Road between Bryant Road and the east branch of the Maple
River. In 1998, O. rimosa were present in great abundance (> 1000/mile of road) in these
locations, with emergence hole densities approaching 50 per square meter. No O.
canadensis were found in any year at these sites.
Behavioral observations
Methods
Okanagana canadensis. I conducted intensive behavioral observations of
uncaged cicadas near the interchange of US Highway 2 and Interstate 75, Mackinac
County, Michigan on July 16, 24, 25, and August 1, 1996, July 27 1997, and on June 20,
1998. I made most observations on O. canadensis in an overgrown field bounded by US 2
(eastbound), an entrance ramp to Interstate 75 (southbound), and Bridge View Road.
Vegetation in this field consisted of shrubs and small northern white cedars (Thuja
occidentalis). For some observations, I placed cicadas in mesh cages, made by folding a
piece of white nylon tulle approximately 1 x 2 m over small cedar seedlings.
On July 25, 1996, I captured and caged 9 male O. canadensis using a dead female
Magicicada septendecim specimen as a lure, and I captured 8 additional males by other
means. I captured two females who landed on the male cage and placed them in a separate
cage, along with four additional females captured by other means. On August 1, I collected
four males and one female cicada using similar methods. I made some observations in the
field, by placing female cicadas in the male cage and videotaping resultant sexual
interactions for later analysis, but I observed no matings in the field. After initial field
240
observations, I separated cicadas by sex, placed them on recently cut tree branches, and
transported them to mesh bag cages placed on an elm sapling bordering a pasture at Richard
D. Alexander’s farm near the intersection of Bethel Church Road and Schneider Road,
Freedom Township, Washtenaw County, MI. Transportation deprived cicadas of fresh
vegetation for a maximum of six hours.
In 1997, using similar methods, I captured and transported to Washtenaw County 4
male and 6 female Okanagana canadensis, except instead of placing the cicadas in mesh
cages, I observed them in a large (4m x 4m x 4m) screen cage (a “flight cage”) placed over
living vegetation. After stocking the cage, I made observations from a truck parked
nearby, because O. canadensis not engaged in sexual behavior are easily startled by the
presence of humans. I videotaped all sexual interactions. Videotaping required an
observer to be out of the truck, but the vehicle was parked in such a way that, from the
location of the cages, the observer could not be silhouetted against the horizon. When
cicadas were not being observed or videotaped, they were separated by sex.
Okanagana rimosa. Because of low cicada densities, I was unable to capture
female O. rimosa in 1996 and 1997. In 1998, O. rimosa emerged in locally high densities
(approaching 50 emergence holes per square meter, comparable to Magicicada); I collected
and observed them in an open, sandy field dominated by bigtooth aspen (Populus
grandidentata) and other shrubby deciduous vegetation near the University of Michigan
Biological Station Streams Laboratory (Fig. 8.5). For most observations, I placed cicadas
in a large flight cage erected over a small quaking aspen (Populus tremuloides).
Measurements of mating duration. In 1996- 1998, I measured the
copulation durations of 4 O. canadensis and 4 O. rimosa mating pairs by observing cicadas
continuously during courtship and mating, recording the time at which the male first
engaged his genitalia and the time at which he separated completely. In 1995, during the
Brood I Magicicada emergence at Alum Springs, Rockbridge County, Virginia, I measured
241
copulation durations of 43 mating pairs of Magicicada septendecim by placing mature,
unmated females and recently caught males in 200 liter mesh cages. To establish starting
and stopping times for each observed copulation, I monitored cages continuously or
scanned them every 10-15 minutes, or ad lib scanned when I heard male courtship sounds.
Mating durations for M. septendecim are thus, in some cases, estimates ± 30 minutes.
Results
Okanagana canadensis. Males typically call one per tree. Male O. canadensis
tend to perch no more than 8 feet off the ground, sometimes at ground level, (approx. 20
noted, many directly observed, drive by observations of many others for height and tree
species information) on the main trunk of small trees, usually cedar (see Davis 1919).
Males call with wings slightly open and abdomen extended, sometimes facing down, or, if
on a side branch, towards the main trunk. Male calls are broad-frequency, slightly metallic
lisps of variable duration. A sonogram of a portion of a male O. canadensis calling bout is
included in Fig. 8.6. In 1997, 22 male O. canadensis calling bouts averaged 2:29 minutes,
and 13 periods of silence between calling averaged 2:22 minutes (Table 8.1); however,
bout lengths are highly variable, and some of the shorter bouts may be courtship behaviors.
Males do not engage in “sing-fly” behavior (Alexander and Moore 1962, see also Gwynne
1987), but rather remain on the same calling perch for long periods. For 5 males observed
in the field, an estimate of the minimum time spent at a calling perch is 17± 16 minutes
(Table 8.1), but males’ lack of movement, and difficulty in determining arrival times at
calling perches suggest this is an underestimate.
With few exceptions (see below), calling male O. canadensis were easily startled
and difficult to capture (pers. obs; see Davis 1919). Males emit an alarm call when
handled, confined, or harassed. Both sexes flew silently from their calling perches at the
slightest provocation; alternatively, some calling males ceased calling and remained
immobile and silent on their perches. Immobile males were difficult to find because of their
242
mottled coloring, and they remained immobile even if their calling tree was thrashed and
shaken. These behaviors suggest that male O. canadensis can adopt two alternative
predator avoidance strategies, escape flight or inconspicuous camouflage.
Okanagana rimosa. Male O. rimosa behavior is generally similar to that of O.
canadensis, except males appear less selective about calling perches, perching on deciduous
trees (often higher than 20 feet), grass stems, or on the ground. A sonogram of a portion
of a male O. rimosa calling bout is included in Fig. 8.7. In 1998, although individual male
O. rimosa could be found calling on trees, shrubs, or even on the ground as in previous
years, there were also single trees that had dense Magicicada – like choruses, except that
males in the trees remained still between calling bouts and did not engage in “call-fly”
behavior, and males did not synchronize their calls, as in Magicicada cassini and M.
tredecassini (Alexander and Moore 1958). The sound intensity at the base of one such tree
was 66db, compared to 80+ db for Magicicada. Males’ lack of movement unless perturbed
made estimation of residence time at calling perches difficult. During 102 minutes of
videotaped sexual interactions during the dense 1998 emergence, in which on average two
males were visible at any given time, only 9 spontaneous, undisturbed flights were
observed, suggesting that even at high densities, calling males tend to remain on their
perches.
The presence of dense aggregations in one tree, while a seemingly similar tree
nearby is devoid of males, is reminiscent of the aggregative choruses of Magicicada. It is
likely that O. rimosa males actively flew to such trees as opposed to emerging underneath
them; although these trees were appropriate species for oviposition, and emergence holes,
shed nymphal skins, and ovipositing females were present at each tree, the density of
emergence holes and shed skins underneath one such tree could not account for the total
number of adult cicadas in the tree. Furthermore, at the study site, nymphal skins and
emerging nymphs were found in the forest bordering the open field, but most calling
243
cicadas were along the forest edge or in trees in the field, suggesting net movement away
from the forest center, although differences in emergence phenology between the forest and
the field cannot be ruled out. Some trees along the forest edge showed evidence of
“flagging,” or twig death due to dense oviposition (Marlatt 1923).
Like O. canadensis males, male O. rimosa emit an alarm call when harassed.
However, although female O. rimosa were easily startled, calling male O. rimosa were
readily approachable and were more efficiently captured by hand than with a net.
Unfortunately, all observations of predator avoidance behavior in O. rimosa were made in
1998, during a locally dense emergence; thus, the apparent differences between O. rimosa
and O. canadensis could reflect density effects: When sparse, both species could employ
the wary behaviors reported for O. canadensis, while higher densities could prompt
behaviors more similar to those reported for O. rimosa. Nevertheless, the association of
high densities and “predator-foolhardy” behaviors is similar to the lack of predatoravoidance behaviors in dense Magicicada emergences (Karban 1982, Williams et al. 1993,
Williams and Simon 1995). Predators were not absent, however: Local residents reported
large gull flocks feeding on cicadas (K. Wehrly, pers. comm.; see also Soper, Delyzer, and
Smith 1976), and the sarcophagid fly, Colcondamyia auditrix Soper (Soper, Shewell, and
Tyrell 1976) an internal parasite of O. rimosa, may also have been present. Of several
hundred cicadas examined, two females had extensive damage to their abdominal tergites,
suggesting the emergence of fly larvae.
Among several hundred O. rimosa examined in 1998, two males and two females
had visible infections of Massopora levispora Soper, a fungal pathogen. Fungal infections
were similar in appearance to Stage I (conidiospore-producing) M. cicadina infections of
Magicicada (Soper 1963; Soper, Delyzer, and Smith 1976; White and Lloyd 1983). No
cicadas producing resting spores were noted, but unlike Stage II (resting spore producing)
M. cicadina infections of Magicicada, second-stage infections of M. levispora do not
244
visibly alter cicadas (Soper 1963; Soper, Delyzer, and Smith 1976). First-stage infections
of M. levispora, like those of M. cicadina, may cause male hosts to be behaviorally
hermaphroditic (Chapter 5): Male Okanagana mount any stationary object they can locate
(see below), and normal males do not tolerate being mounted by others; however, I
observed one intriguing case in which a fungus infected male O. rimosa repeatedly
appeared to tolerate mounts by different males. This male may, due to fungal infection,
have been accepting mounts and acting in the interests of the fungus; he did not appear
otherwise lethargic or immobile. Thus, although confirmation awaits additional data,
Massospora levispora, like M. cicadina, may increase its chances of spreading by causing
males to be sexually attractive to males as well as females. The mechanism by which the
fungus accomplishes these changes could involve castration, which the fungus
accomplishes by destroying the abdomen and its internal organs, and which may result in
feminized behaviors.
An interesting correlate of the absence of “sing-fly” behavior in O. rimosa is the fate
of individuals infected with the fungus, Massopora levispora. O. rimosa with sporeproducing (Stage II) fungal infections exhibit no outward signs of infection and die more or
less in situ (Soper 1963; Soper, Delyzer, and Smith 1976), consistent with a fungal
adaptation to deposit spores where cicadas are likely to be found in the next generation, and
a correlate of the extreme form of philopatry associated with lack of sing-fly behavior. In
contrast, normal Magicicada males engage in sing-fly behaviors, and individuals with Stage
II Massospora cicadina infections engage in long, spore-dispersing flights consistent with a
fungal adaptation to ensure transmission in such highly mobile species (Alexander et al. in
prep.).
Sexual behaviors, both species. Males alternated long bouts of calling with
bouts of wing flicking (see Davis 1919), producing from 3-5 flicks per wing flick bout.
Male Okanagana did not respond to artificial wing flicks produced either during their call or
245
during periods of wing flicking, but sometimes two wing flicking cicadas seemed to
produce click bouts in alternation. Females of both species flew to calling males, which
they approached by walking. Except to approach a calling male, females remained still and
passive prior to mating. In some cases, males made short (< 20 sec.) calls while being
approached by females; these shortened calls could be a form of courtship, although it is
doubtful that females require males to make such calls since not all males did so. In some
cases, mating occurred after a short (ca. 10 minute) bout of wing flicking in which both
male and female were seen flicking, and in which the male never called or made any other
signal besides wing flicks. Whether or not the male made short calls or either sex wing
flicked, once the female made physical contact with the male, they mated without elaborate
courtship (5 O. canadensis, 5 O. rimosa mating sequences observed). For each species,
four complete matings were observed; O. canadensis matings lasted an average of 34
minutes, while the O. rimosa matings averaged 19 minutes (Table 8.2). Mating durations
in these two species are not significantly different, because the mating duration average in
O. canadensis is affected by one extreme outlier (Mann-Whitney; U = 9; P ≤ 0.773,
although small sample sizes make this test of dubious validity).
Males presented with an immobile female or similar object were imperturbably
persistent in investigating it and were not startled by the presence or actions of a human
observer. In 9/9 cases in which a calling male O. canadensis and in 16/20 cases in which a
calling male O. rimosa was presented with a model female, the male mounted, extruded his
genitalia, and attempted to copulate with the model (Table 8.3). When calling males were
presented with live males of their own species, they courted them as if they were females
and attempted to copulate with them. The courted male generally responded with a wing
flick or an alarm buzz, which were only moderately effective in deterring the courter.
Mating durations in Okanagana averaged 26.6 (± 21.75) minutes (Table 8.2). In contrast,
for 43 Magicicada mating pairs, copulation duration averaged 243:24 (± 121:48) minutes.
246
Discussion
The surveys of O. canadensis and O. rimosa distributions demonstrate that these
species are not strictly periodical, because adults were often present in the same locations in
succeeding years, and thus the cicadas in a given locality are not synchronized in their
development. However, these species do vary considerably in their yearly abundance, so it
may be appropriate to consider them semi-periodical. Some locations, such as the 1998
study site for O. rimosa, may, with repeated sampling, allow final determination of these
insects’ life cycles, although the lack of periodicity and year-classes may continue to make
estimation of life cycle length based on emergence records difficult.
Females O. canadensis and O. rimosa signal sexual receptivity by physically
approaching calling males, and sexual approach is thus under female control. Males
attempt to mount nearby, stationary, cicada-like objects, because the only such objects they
will ever normally encounter are sexually receptive females. Males have few opportunities
for forcing matings since unreceptive females can avoid calling males.
Wing-flicking in Okanagana may have multiple functions. One function of wing
flicking may be related to the presence of parasites. Sarcophagid flies (Colcondamyia
auditrix ) are attracted to the calling songs of O. rimosa, although infection of O.
canadensis is not known (Soper Shewell and Tyrell 1976). The flies lay eggs on the
bodies of cicadas. Singing punctuated by bouts of wing-flicking may be an effective
deterrent to parasitism because the wing flicking may disrupt the fly’s oviposition activities.
Anecdotal evidence, which suggests that touching a male will provoke a bout of wing
flicking, supports this explanation for the function of wing flicking (unpublished data).
The occasional use of wing flicks by males to deter courtship by other males supports the
view that wing flicks in O. canadensis and O. rimosa are general deterrents to close contact
or unsolicited touching, either by parasites or by conspecifics. However, parasite
247
avoidance fails to fully explain why females should wing flick, because they produce no
acoustical signals attractive to parasites. It is possible that females wing flick only when in
proximity to a calling male, and thus when in jeopardy of being found by parasites attracted
to the male’s calls.
Alternatively, male wing flicking may be a close-range, low-cost signal. In special
circumstances when males and females find themselves nearby or in the same tree without
having undergone the usual rapprochement sequence, wing flicks may allow
rapprochement without requiring the male to sing extended, easily localized songs and
reveal his presence to predators, parasites, or other males. It may benefit females to signal
back, to alert males of their proximity and decrease the likelihood that males will make
conspicuous acoustical signals. The wing flick interactions of males and females may thus
be a quiet, unobtrusive way to facilitate rapprochement while lessening the danger of
parasitism3 (see also Heller and von Helverson 1993). The acoustical component of wing
flicking, or both acoustical and visual components of this behavior, as in Magicicada, may
constitute the signal itself. Alternatively, since other homopterans are known to use
vibrational signals (Hunt and Nault 1991, Hunt et al. 1992), and since the female’s final
approach is by walking, she could potentially detect vibrations caused by the male’s wing
flicking.
Contrasts between Okanagana and Magicicada behavior.
The male genitalia of O. canadensis and O. rimosa (Fig. 8.8) lack the hooks and
claspers of Magicicada genitalia (Fig. 8.9; see also Moore and Alexander 1958). Male
Okanagana genitalia have a slender adeagus with a scoop-like structure and a long, folded,
threadlike extension. Males have been observed to fully penetrate the female genital tract
3
This hypothesis is not intended to be general for cicadas; see Popov 1981 for an account of Cicadetta
sinuatipennis, in which tymbal sounds are hypothesized to function in close-range interactions, and wing
flicks for long-range attraction.
248
within ten seconds of starting copulation. The inability of the “scoop” to penetrate the
female as deeply as the adeagus makes the function of this device as a scoop questionable.
If the scoop actually functions to remove sperm stored within the female spermatheca, then
the combination of a scoop and long tube, through which ejaculate presumably passes and
is deposited near the eggs, could be interpreted as a male adaptation to guard paternity in
the face of polyandry. The lack of male genitalic hooks and the female’s control over
sexual approach suggest that any degree of polyandry would most likely be under female
control. During observations of mated females, none remated.
Alexander et al. (1997) described the influence of the nature of male-female
rapprochement on the complexity of different stages of the sexual sequence. When males
lure females, using songs, resources, or other advertisements of quality, and when females
control sexual approach, courtship is typically complex and genitalia are simple. When
males obtain mating by force or control sexual approach, courtship is minimal and genitalia
are complex. In Okanagana, males remain stationary and use their songs to lure females,
and their genitalia appear to be simple and lack hooks or claspers (Fig. 8.8). Because
males are unlikely ever to encounter unreceptive females or to encounter competitive males
during or immediately prior to mating, structures that could be adaptations for forcing
matings, such as claspers, hooks, etc., are absent.
In contrast, Magicicada males actively search for females and use their calling songs
to assist in search. Magicicada genitalia are a virtual battery of clasping, hooking, and
ratcheting devices/appliances (Fig. 8.9). Male Magicicada have several opportunities
during the sexual sequence to employ force. First, males use their songs to cause females
to reveal their locations with wing flick signals and then approach signaling females
(Chapter 5). Searching and approaching males have opportunities to force matings on
resistant or marginally receptive females, and may use complex genitalia to do so. Second,
complex genitalia may be the result of intrasexual competition. The final stages of
249
Magicicada courtship are easily perceived and localized by other, potentially interloping
males. A copulating pair is relatively immobile and easily harassed by other males, who
attempt to insert their genitalia as well. Once a Magicicada male begins copulation, it is in
his interests to hold on as tightly as possible to avoid displacement. Note that displacement
may not necessarily be contrary to the female’s interests, since the most competitive male
may well sire the most competitive sons. Finally, copulation duration in Magicicada is
much longer than in Okanagana, and appears to be in excess of durations required for
completeness (Chapter 4). It is possible that male Magicicada use their genitalia to forcibly
prolong copulation durations contrary to female interests, perhaps as a form of mateguarding (Chapter 3).
Hypotheses concerning evolution of the unique mating system of
Magicicada.
Suppose that the non-periodical ancestor of periodical cicadas had mating behaviors
more similar to those of present-day Okanagana than to those of modern Magicicada. Such
an ancestor might have lacked mass mating aggregations, genitalic hooks and claspers,
complex courtship, and in most respects been similar to present-day non-periodical cicadas.
Females would have signaled their receptivity only by approaching stationary, calling
males.
Although the exact mechanisms by which Magicicada became periodical and year
classes (or “broods”) formed are unknown, the synergy between (1) the ability of longlived insects with prime number life cycles to avoid predators’ numerical responses and (2)
increasing population sizes (with “selfish herd” connotations) likely promoted periodicity
(See Lloyd and Dybas 1966a, b, Hoppensteadt and Keller 1976, Bulmer 1977, Martin and
Simon 1990, Heliövaara et al. 1994, Yoshimura 1997, Cox and Carlton 1998, for
discussion of the evolution of periodicity). The changes leading to periodicity also would
250
have caused changes in the sexual environment faced by members of the primordial
broods.
For cicadas with mating behaviors similar to Okanagana, the task of finding mates
is comparatively simple: Adult Okanagana males and females likely only ever encounter
receptive members of the opposite sex, because females fly to widely separated singing
males. As Magicicada emergences came to involve greater and greater numbers of cicadas,
males may have begun to benefit by searching for females instead of waiting for females to
come to them. Increasing densities present in a given year would have made it difficult for
unreceptive females to avoid the unsolicited attentions of males and at the same time made it
easier for males to locate females without waiting for females to come to them. The
intersexual conflict thus engendered may have been one factor leading to the complex
courtship of Magicicada. Whereas Okanagana male courtship behaviors seem primarily to
facilitate rapprochement, much of Magicicada singing and tactile behavior seems focused
on persuading females to copulate once contact has been made.
At some point, not far removed in time from the evolution of periodicity, or perhaps
contemporary with it, the tasks faced by female periodical cicadas became more complex as
the M. -decim, M. -cassini, and M. -decula species came into existence and stable
sympatry. The coexistence of ecologically similar, synchronized species would have
forced females to require males to produce some form of species identification, the vestiges
of which may remain in the differences in the Magicicada species’ male calling songs,
regardless of whether those differences arose incidentally in allopatry/allochrony, or as a
result of character displacement in sympatry/synchrony. Such a situation would have
promoted the equivalent of intersexual arms races, as females evolved to resist and thwart
male forcing behaviors and males evolved both more effective tools of persuasion (such as
more distinctive songs) and more effective tools for forcing, such as genitalic claspers,
hooks, and ridges. Because increasing densities would also have promoted more intense
251
male-male competition, males may have benefited from developing means of persuading or
forcing slightly immature females to copulate, and they may have benefited from
adaptations for protecting their paternity, perhaps by using prolonged copulations as a form
of mate-guarding.
Paradoxically, as Magicicada emergences became denser and came to contain
several species, it may have become more and more difficult for females to find receptive
mates; a periodical cicada chorus contains teneral, mated, same-sex, conspecific, and
heterospecific individuals, all of which could be inappropriate targets of male courtship.
Thus, females, who have little to gain by delaying once ready to mate, may, in the past,
have benefited from acting in ways to attract the attention of male conspecifics, such as by
signaling in response to their calls. The female receptivity signal of periodical cicadas
(Chapter 5) may have arisen under such a scenario. As receptive females began to signal
males, the intrasexual competition faced by males would have become more severe,
because males would have adopted the tactic of converging on signaling females. Genital
hooks, claspers, and other appliances for facilitating rapid mating, preventing usurpation,
and prolonging copulation would have been favored.
This account of the evolution of the unique mating system of periodical cicadas is
prompted by the contrasts between Okanagana and Magicicada. It is not intended to be
definitive, but rather to present some of the themes, such as male-female antagonistic
coevolution, male-male competition, and mate search strategies, that may have been
important in Magicicada mating system evolution. The next step in evaluating these themes
is to search for commonalities between Magicicada and other species in which females
signal in response to searching males (such as fireflies, Lloyd 1966, 1979, or another
cicada, Tibicen pruinosa, pers. obs).
252
Literature Cited
Alexander, R. D. 1960. Sound communication in Orthoptera and Cicadidae. In: W. E.
Lanyon, W. N .Tavolga, eds, Animal Sounds and Communication. AIBS
symposium series publication 7: 38- 92.
Alexander, R. D., and T. E. Moore. 1958. Studies on the acoustical behavior of seventeenyear cicadas (Homoptera: Cicadidae: Magicicada). Ohio J. Sci. 58:107-127.
Alexander, R. D., and T. E. Moore. 1962. The evolutionary relationships of 17-year and
13-year cicadas, and three new species. (Homoptera: Cicadidae, Magicicada).
Misc. Publ. no. 121. Univ. of Michigan Museum of Zoology, Ann Arbor, MI.
Alexander, R. D., D. C. Marshall, and J. R. Cooley, 1997. Evolutionary Perspectives on
Insect Mating. Pp. 1- 31 in B. Crespi, J. Choe, eds. The evolution of mating
systems in insects and arachnids. (Cambridge University Press).
Alexander, R. D., A. F. Richards, D. C. Marshall, and J. R. Cooley (in prep.). The
microdistributional relationships of 17- and 13- year cicadas in Illinois
(Magicicada): Broods III, 1997, and XIX, 1998.
Alexander, R. D., J. R. Cooley, and D. C. Marshall. (in prep.). A specialized fungal
parasite (Massospora cicadina) modifies the behavior of periodical cicadas
(Homoptera: Cicadidae: Magicicada).
Bulmer, M. G. 1977. Periodical Insects. Am. Nat. 111: 1099- 1117.
Cox., R. T., and C. E. Carlton. 1998. A commentary on prime numbers and life cycles of
periodical cicadas. Am. Nat. 152: 162-164.
Davis, W.T. 1919. Cicadas of the genera Okanagana, Tibicinoides, and Okanagodes, with
descriptions of several new species. Journal Of The New York Entomological
Society 27: 179- 223.
Gwynne, D. T. 1987. Sex-biased predation and the risky mate-locating behaviour of male
tick-tock cicadas (Homoptera: Cicadidae) Anim. Behav. 35: 571- 576.
Heliövaara, K., R. Väisänen, and C. Simon. 1994. Evolutionary Ecology of periodical
insects. TREE 9: 475- 480.
Heller, K. G., and von Helverson, D. 1993. Calling behavior in bushcrickets of the genus
Poecilimon with differing communication systems (Orthoptera: Tettigonioidea,
Phaneropteridae). J. Ins. Behav. 6: 361- 377.
Hoppensteadt, F. C., and J. B. Keller. 1976. Synchronization of periodical cicada
emergences. Science 194: 335- 337.
Huber, F., D. W. Wohlers, J. L. D. Williams, and T. E. Moore. 1979. Struktur und
Funktion der Hörbahn von Singzilkaden. Verh. Dtsch. Zool. Ges . 1979: 212.
253
Hunt, R. E., and L. R. Nault. 1991. Roles of interplant movement, acoustic
communication, ad phototaxix in mate-location behavior of the leafhoper
Graminella nigrigfrons. Behav. Ecol. Sociobiol 28: 315-320.
Hunt, R. E., J. P. Fox, and K. F. Haynes. 1992. Behavioral response of Graminella
nigrifrons (Homoptera: Cicadellidae) to experimentally manipulated vibrational
signals. J. Ins. Behav. 5: 1- 13.
Karban, R. 1982. Increased reproductive success at high densities and predator satiation
for periodical cicadas. Ecology 63: 321-328.
Kirby, C. S., and H. G. McPhee. 1963. Observations of the cicada, Okanagana rimosa
(Say). Canad. Dept. For Bi-Mon. Prog. Rep. 19 (5): 1
Lloyd, J. E. 1979. Sexual selection in luminescent beetles. Pp. 293- 342 in M. S. Blum
and N. A. Blum, eds. Sexual selection and reproductive competition in insects.
(Academic Press: New York).
Lloyd, J. E., 1966. Studies on the flash communication system in Photinus fireflies. .
Misc. Publ. no. 130. Univ. of Michigan Museum of Zoology, Ann Arbor, MI.
Lloyd, M. and H. S. Dybas. 1966a. The periodical cicada problem. I. Population Ecology.
Evolution 20: 133- 149.
Lloyd, M. and H. S. Dybas. 1966b. The periodical cicada problem. II. Evolution.
Evolution 20: 466- 505.
Marlatt, C. L. 1923. The periodical cicada. U.S.D.A. Bureau of Entomology Bulletin 71,
183 pp.
Marshall, D. C., J. R. Cooley, R. D. Alexander, and T. E. Moore. 1996. New records of
Michigan cicadidae (Homoptera) with notes on the use of songs to monitor range
changes. Great Lakes Ent. 29:165- 169.
Martin, A. and C. Simon. 1990. Temporal variation in insect life cycles. Bioscience 40:
359- 367.
Moore, T. E., and R. D. Alexander. 1958. The periodical cicada complex (Homoptera:
Cicadidae). Proc. X Internat. Cong. Entomol. 1: 349- 355.
Popov, A. V. 1981. Sound production and hearing in the cicada, Cicadetta sinuatipennis
Osh. (Homoptera: Cicadidae). J. Comp. Physiol. 142: 271-280.
Simon, C. M. 1979. The debut of the 17-year-old cicada. Natural History 88: 38-45.
Soper, R. S. 1963. Massospora levispora, a new species of fungus pathogenic to the
cicada Okanagana rimosa. Can. J. Bot. 41: 875- 878.
Soper, R. S., A. J. Delyzer, and L. F. R. Smith 1976. The genus Massospora
entomopathogenic for cicadas. Part II. Biology of Massospora levispora and its
host Okanagana rimosa, with notes on Massospora cicadina on the periodical
cicadas. Ann. Ent. Soc. Am. 69: 89- 95.
254
Soper, R. S., G. E. Shewell, and D. Tyrell. 1976. Colcondamyia auditrix nov. sp.
(Diptera: Sarcophagidae), a parasite which is attracted by the mating song of its
host, Okanagana rimosa (Homoptera: Cicadidae). The Canadian Entomologist 108:
61- 68.
White, J., and M. Lloyd. 1983. A pathogenic fungus, Massospora cicadina Peck
(Entomophthorales) in emerging nymphs of periodical cicadas (Homoptera:
Cicadidae). Environmental Entomology 12: 1245- 1252.
Williams, K. S., K. G. Smith, and F. M. Stephen. 1993. Emergence of 13-yr periodical
cicadas (Cicadidae: Magicicada): phenology, mortality, and predator satiation.
Ecology 74: 1143- 1152.
Williams, K. S., and C. Simon. 1995. The ecology, behavior, and evolution of periodical
cicadas. Ann. Rev. Entomol. 40: 269- 295
Yoshimura, J. 1997. The evolutionary origins of periodical cicadas during ice ages. Am.
Nat. 149:112-124.
255
Table 8.1. Lengths (minutes) of Okanagana canadensis calling
and silent bouts for 5 males.
Number of bouts
Average bout length (min:sec)
Standard deviation
Calling
4:40
15:09
9:45
0:45
3:20
1:05
0:59
1:30
1:02
2:41
0:32
2:33
1:21
0:11
0:25
1:09
1:26
0:40
0:51
1:26
2:29
0:45
Silent
1:32
4:28
4:32
0:34
8:23
1:42
0:23
2:19
0:34
0:42
4:44
0:34
0:23
At perch
4:40
17:18
12:47
6:32
43:06
22
2:29
3:30
13
2:22
2:27
5
16:53
15:30
256
Table 8.2. Mating durations in Okanagana and Magicicada
septendecim. One O. canadensis mating lasted
approximately 4 times as long as the average of the others,
so two averages, one with this long mating, and one
without, are calculated for this species.
Species
O. rimosa
Average
O canadensis
Average
Average w/o outlier
Magicicada septendecim
Average (n= 43)
Duration (min: sec)
19:00
17:00
22:09
17:04
18:48 ± 2:25
80:00
19:06
15:09
23:18
34:23 ± 30:35
19:11 ± 4:05
195
195
295
167
235
235
310
60
305
354
124
440
310
220
263
354
360
263
65
148
310
405
243:24 ± 121:48
257
315
330
442
297
235
195
500
124
111
295
331
440
64
444
295
360
250
310
526
91
60
Table 8.3. Responses of male cicadas to presentation of model cicada.
Males were scored as responding to model if they directed sexual behaviors,
such as extruding genitalia, mounting, or copulating, towards the model. In
both species, males were more likely to respond positively to the model (X2
goodness of fit; null hypothesis equal likelihood of positive and negative
response). Although, in O. rimosa, the likelihood positive and negative
responses do not differ from 50% at the significance level of P ≤ 0.05, this
is probably due to small sample size, since the response rates of the two
species do not differ (Fisher’s Exact Test P ≤ 0.280).
Species
O. canadensis
O. rimosa
Trials
9
20
Responses
Positive
Negative
9
0
16
4
258
X2 P
≤ 0.05
≤ 0.10
259
Figure 8.1. Locations of study sites, Okanagana canadensis, O. rimosa 1996- 1998.
Location of 1998 O. rimosa study
site (Figure 5)
Area included in 1996- 1998 O. rimosa
and O. canadensis distribution surveys
(Figures 2-4)
260
l
l
l l
l
St. Ignace
0
l
l
l
l l
s
l
l l
s l s
l l l
Lake Huron
l l
l
l l l
20
l l l
Cedarville
l l l
l
l
l
l
l l
l
l
l
l
l
l
l
Figure 8.2. 1996 distribution map of O. rimosa and O. canadensis. Only roads and hydrologic features relevant to
distributional study shown. s : O. rimosa; l : O. canadensis. Continuous sequences of symbols indicate continuous
distribution of cicadas in appropriate habitat.
Lake
Michigan
l
s
s
l l
l
40
s
Miles
261
l
l
l
l
l
l
l l
l
l
St. Ignace
s
s
0
l
l
l
l
l
l
l
l l
Lake Huron
l
l
l s l s
s l s l
20
l
l
l l l
Cedarville
l
l
l l l
l l l
l
l
Figure 8.3. 1997 distribution map of O. rimosa and O. canadensis. Only roads and hydrologic features relevant to
distributional study shown. s : O. rimosa; l : O. canadensis. Continuous sequences of symbols indicate continuous
distribution of cicadas in appropriate habitat.
Lake
Michigan
l
l
l
l
l
l
l
l
40
Miles
262
l
l
l
l l
l l
s
l
l
l l
l l l l l l l l l ls l l l
l l l
l l
l l
l l
l
s
l
l
l
St. Ignace
0
l
s
l
l
l
ll
l l
l
l
l l
l
l ll l l
l
l
l
l
l
l
l
l
Lake Huron
l l l l l l l l l l
l l
l l
s s
s s s
s s
l l l l
l s l l l l l l
l l
s
s
s
Cedarville
s
20
l l l
l s l l l l l l l l l l l l l l l l l l
l l
l l l l l l l
l l l
l
l
l
l l
s
s
s
l
l
l l
l
l
s l l
s
l l l l l l l l l l
l
l l l l l l
l ll l l s l s
l
s
l
s
s l
l l
l
s
l
s
l
l
40
l l l l l l l l l l l l l l
l l l l l
l
s
l
s
Figure 8.4. 1998 distribution map of O. rimosa and O. canadensis. Only roads and hydrologic features relevant to
distributional study shown. s : O. rimosa; l : O. canadensis. Continuous sequences of symbols indicate continuous
distribution of cicadas in appropriate habitat.
Lake
Michigan
l
l
l
l
l l
l l
l
l
l
l l
l s l
l l l
l
l l
l l
l l l l l
l
l
l
s s l
s
l l
s
l
l
l
l
s l
l
Miles
263
Lake Road
0
Riggsville Road
3
Douglas Lake
Figure 8.5. Location of O. rimosa study site in 1998.
3
Douglas
Open field near UMBS streams laboratory
UMBS
6 Miles
264
2
0.0
2.0
Figure 8.6. Sonogram of male Okanagana canadensis call.
kHz
4
6
8
10
Time (s)
4.0
6.0
7.0
265
0.0
2.0
4.0
Figure 8.7. Sonogram of male Okanagana rimosa call.
kHz
2
4
6
8
10
6.0
Time (s)
8.0
10.0
12.0
266
Figure 8.8. Sketches of lateral view of typical Okanagana canadensis male genitalia. Okanagana rimosa are similar.
Sketches not to scale (Otte, Anderson).
267
Magicicada septendecula
Figure 8.9. Sketches of lateral view of typical Magicicada male genitalia. Sketches not to scale.
(Otter, Anderson).
Magicicada cassini
Magicicada septendecim
APPENDICES
268
APPENDIX A
C++ code for mating system simulation. Subroutines separated by asterisks (* * * * * *).
param. h:
// Constants for program
#define PopSize 15
#define attributes 5
//Size of male population
//Number of male attributes in phenotype
#define maxthresh 100 //Maximum threshold
#define maxgen 1000
//Maximum number of generations
#define pickvalue 10 //Maximum value for random number
#define yes 1 //
#define no 0
//
(* * * * * * * *)
header.h:
// Header file
//+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
#include "param.h"
//Sets some global parameters
//function prototypes for simulation
void bubblesort(int arr[], int n);
void timeseed(void);
int picker(int n);
void wipe (void);
//sorts
//seeds
//picks
//wipes
void dump (void);
//dumps male population to screen
void
void
void
void
//gives the user a choice of functions
//gets user inputs
//writes output to disk or screen
//provides summary information
choicer(void);
inputs(void);
output(void);
summarize(void);
the array its passed
random number generator
a random # inspecified range
contents or an array
void simulate(void);
void iterate(void);
void generation(void);
//runs the simulation
//runs one iteration
//runs one generation
void fillfirstgen(void);
void fillnextgen(void);
//Fills first generation of phenotypes
//Fills next generation of phenotypes
void bestofn(void);
void threshold(void);
//Chooses best of n males
//Chooses males exceeding a threshold
struct male
{
int sum;
//male data structure
269
int mate;
int phenotype[attributes];
};
//data structure to hold user inputs
struct userparameters
{
int threshold;
int cutoff;
int mutation;
int mutvalue;
char excel;
char file;
int iterations;
int generations;
char screen;
};
//data structure to hold parameters at a given time
struct timeparameters
{
int genernum;
int crash;
int iternum;
int thdmated;
float thdvalue;
int bonmated;
float bonvalue;
};
(* * * * * * * *)
main.cp:
// Main program loop. Initializes global variables, sets contents of variables/
//arrays to zero, and calls functions of the siimulation.
#include <stdlib.h>
#include "header.h"
#include "param.h"
//includes all function prototypes
//includes global constants
//***********************Declare some global variables*******************
//initialize population of males, a global array of structures.
male thdmales[PopSize];
//male population undergoing threshold choice
male bonmales[PopSize];
//male population undergoing bn choice
//initialize a bookkeeping array
int bookeep[PopSize];
//Initialize two data structures to hold user inputs and present
// iteration/generation values.
userparameters userinputs;
timeparameters present;
//Results summaries go in these arrays
float thdsum[maxgen];
float bonsum[maxgen];
//Keep track of how many times threshold choice population crashes
270
//int tag;
int thdcrash[maxgen];
int main(void)
{
//********************Start the main program**************************
timeseed();
wipe();
//for the first time through, you don't get a choice of what to do
inputs();
//Get input parameters
simulate();
// Run the simulation
//After the first time through, you get a choice.
choicer();
//Run the loop
return 0;
//End the program
}
(* * * * * * * *)
bestofn.cpp:
//function definition for evaluating and marking best of n males
//In case of ties, all typing males contribute to next generation
#include
#include
#include
#include
extern
extern
extern
extern
<math.h>
<stdlib.h>
"header.h"
"param.h"
userparameters userinputs;
male bonmales[PopSize];
int bookeep[PopSize];
timeparameters present;
void bestofn(void)
{
present.bonmated = 0;
present.bonvalue = 0;
for (int a = 0; a < PopSize; a++)
//Test male to see whether he passes
{
bonmales[a].sum = 0;
for (int b = 0; b < attributes; b++)
{
bonmales[a].sum = (bonmales[a].sum + bonmales[a].phenotype[b]);
}
bonmales[a].mate = no;
bookeep[a] = bonmales[a].sum;
}
bubblesort(bookeep, PopSize);
for (a = 0; a < PopSize; a++)
//take the top n
{
//males and mark
if (bonmales[a].sum >= bookeep[(PopSize - userinputs.cutoff)])
271
//them as mated
{
for (int n = 0; n < userinputs.cutoff; n++)
if (bonmales[a].sum >= bookeep[((PopSize - userinputs.cutoff) +
n)])
bonmales[a].mate = (yes + n); //Males are ranked
//bonmales[a].mate = yes;
//High numbers are high quality
present.bonmated = present.bonmated + yes;
present.bonvalue = (present.bonvalue + bonmales[a].sum);
}
}
present.bonvalue = (present.bonvalue/present.bonmated);
}
(* * * * * * * *)
bubblesort.cpp:
//function definition for c version of bubblesort
//must be passed and array and the size of the array
#include
#include
#include
#include
#include
<iostream.h>
<math.h>
<stdlib.h>
"header.h"
"param.h"
void bubblesort(int arr[], int n)
{
//note that because C arrays start at zero, counter i must go down
//from arraysize-1 to 0 instead of from arraysize to 1.
for (int i = (n-1); i >= 0; i = (i-1))
for (int j = 0; j < i; j++)
if (arr[j] > arr[j+1])
{
int k = arr[j];
arr[j] = arr[j+1];
arr[j+1] = k;
}
}
(* * * * * * * *)
choicer.cpp:
// choicer-- runs the loop in which the simulation occurs. Gives the user a choice
// of entering new parameters, using already entered parameteres, or stopping.
//First simulation occurs in Main outside this loop; then Choicer is called and
// you get a choice of whether to continue or not
#include
#include
#include
#include
<iostream.h>
<stdlib.h>
"header.h"
"param.h"
// for exit()
void choicer()
{
int select;
272
extern userparameters userinputs;
cout << "\n";
cout << "\n";
cout << "Would
cout << "\n";
cout << " (1)
cout << "\n";
cout << " (2)
cout << "\n";
cout << " (3)
cout << "\n";
cin >> select;
cout << "\n";
cout << "\n";
//only here to provide "WORKING" display on screen.
you like to ";
run a simulation with new parameters? ";
run a simulation using previous parameters? ";
end the simulations? ";
if (select == 1)
//Run simulation with new input parameters
{
inputs();
//Get input parameters
if (!(userinputs.screen == 121))
{
cout << "\n";
cout << " W O R K I N G ";
cout << "\n";
}
simulate();
//Run the simulation
choicer();
//Choose what to do
}
if (select == 2)
//Run simulation with old parameters
{
if (!(userinputs.screen == 121))
{
cout << "\n";
cout << " W O R K I N G ";
cout << "\n";
}
simulate();
//Run the simulation
choicer();
//Choose what to do
}
if (select == 3)
//Stop the simulations
{
select = select;
//Basically nonsense-- just pop back out to Main.
}
}
(* * * * * * * *)
fillfirstgen.cpp:
//Algorithm for filling first generation with randomly generated
//phenotypic attributes. Best of N males and threshold males
//are filled identically.
#include <stdlib.h>
#include "header.h"
#include "param.h"
273
extern male bonmales[PopSize];
extern male thdmales[PopSize];
void fillfirstgen(void)
{
for (int a = 0; a < PopSize; a++)
//ratchet through each male
{
for (int b = 0; b < attributes; b++) //ratchet through each attribute
{
bonmales[a].phenotype[b] = picker(pickvalue); //fill bn males
thdmales[a].phenotype[b] = bonmales[a].phenotype[b];
//fill thd males the same
}
}
}
(* * * * * * * *)
fillnextgen.cpp:
//Algorithm for using mating status in present generation to determine
//phenotypic attributes of next generation
#include <stdlib.h>
#include "header.h"
#include "param.h"
extern
extern
extern
extern
male bonmales[PopSize];
male thdmales[PopSize];
timeparameters present;
userparameters userinputs;
extern int tag;
void fillnextgen(void)
{
int proportion = 0;
int counter = 0;
male holder[PopSize];
//used in setting up bonmales proportions
//Used in setting up bonmales proportions
//Temporary array for holding thd or bon males
//Clear holder array************************************************************
for (int a = 0; a < PopSize; a++)
//ratchet through each male
{
holder[a].sum = 0;
//Make sure holder is empty
holder[a].mate = 0;
for (int b = 0; b < attributes; b++) //ratchet through each attribute
{
holder[a].phenotype[b] = 0;
}
}
//Transfer Best of N male data to holder***********************************************
for (a = 0; a < PopSize; a++)
//ratchet through each male
{
holder[a].sum = bonmales[a].sum;
//transfer to storage
bonmales[a].sum = 0;
//wipe to zero
274
holder[a].mate = bonmales[a].mate;
//transfer to storage
bonmales[a].mate = no;
//wipe to zero
for (int b = 0; b < attributes; b++) //ratchet through each attribute
{
holder[a].phenotype[b] = bonmales[a].phenotype[b];
//transfer to storage
bonmales[a].phenotype[b] = 0;
//wipe to zero
}
}
//Transfer successful Best of N male data back into population array******************
if (present.bonmated > 0)
{
for (a = 0; a < PopSize; a++)
proportion = proportion + holder[a].mate;
proportion = (PopSize/proportion);
counter = 0;
for (a = 0; a < PopSize; a++)
//Fill next generation with attributes
{
if (holder[a].mate >= yes)
//of successful males
{
for (int c = (counter* proportion * holder[a].mate); c < PopSize;
c++)
{
for (int b = 0; b < attributes; b++)
//ratchet through each attribute
{
bonmales[c].phenotype[b] = holder[a].phenotype[b];
//transfer successful attributes
}
counter = (counter +1);
}
}
}
}
//Clear holder array************************************************************
for (a = 0; a < PopSize; a++)
//ratchet through each male
{
holder[a].sum = 0;
//Make sure holder is empty
holder[a].mate = 0;
for (int b = 0; b < attributes; b++) //ratchet through each attribute
{
holder[a].phenotype[b] = 0;
}
}
//Transfer threshold Choice Male data into holder**************************************
for (a = 0; a < PopSize; a++)
//ratchet through each male
{
holder[a].sum = thdmales[a].sum;
//transfer to storage
thdmales[a].sum = 0;
//wipe to zero
holder[a].mate = thdmales[a].mate;
//transfer to storage
275
thdmales[a].mate = no;
//wipe to zero
for (int b = 0; b < attributes; b++) //ratchet through each attribute
{
holder[a].phenotype[b] = thdmales[a].phenotype[b];
//transfer to storage
thdmales[a].phenotype[b] = 0;
//wipe to zero
}
}
//Transfer successful threshold choice male data back into
population*********************
if (present.thdmated > 0)
{
proportion = (PopSize/present.thdmated);
counter = 0;
for (a = 0; a < PopSize; a++)
//Fill next generation with attributes
{
if (holder[a].mate == yes)
//of successful males
{
for (int c = (counter* proportion); c < PopSize; c++)
{
for (int b = 0; b < attributes; b++)
//ratchet through each attribute
{
thdmales[c].phenotype[b] = holder[a].phenotype[b];
//transfer successful attributes
}
counter = (counter +1);
}
}
}
}
//Handle Mutation Events ************************************************************
//Note: Attributes are sampled with replacement, so the number of mutations
//will be (1 < mutations < userinputs.mutations)
for (a = 0; a < PopSize; a++)
//ratchet through each male
{
for (int b = 0; b < userinputs.mutation; b++) //number of allowable mutations
{
//For each male, pick a random attribute and mutate it
//to a random integer
bonmales[a].phenotype[picker(attributes)] = picker(userinputs.mutvalue);
if (present.thdmated > 0)
thdmales[a].phenotype[picker(attributes)] =
picker(userinputs.mutvalue);
}
}
}
(* * * * * * * *)
generation.cpp:
//One generation
#include <stdlib.h>
276
#include "header.h"
#include "param.h"
extern float thdsum[maxgen];
extern float bonsum[maxgen];
extern timeparameters present;
extern userparameters userinputs;
void generation(void)
{
bestofn();
threshold();
//commented out lines are for old way of calculating cumulative totals.
//thdsum[present.genernum] = (thdsum[present.genernum] + present.thdvalue);
//bonsum[present.genernum] = (bonsum[present.genernum] + present.bonvalue);
thdsum[present.genernum] = (thdsum[present.genernum] +
(present.thdvalue/userinputs.iterations));
bonsum[present.genernum] = (bonsum[present.genernum] +
(present.bonvalue/userinputs.iterations));
//dump();
}
(* * * * * * * *)
inputs.cpp:
//This function provides prompts for user inputs and writes summary information
//to a text file, if desired. Most inputs are checked upon entry and set
//to a default value if user input exceeds acceptable value.
#include
#include
#include
#include
#include
<iostream.h>
<fstream.h>
<stdlib.h>
"header.h"
"param.h"
// for exit()
const char * file3 = "data.txt";
const char * file4 = "data.xl";
const int Len = 40;
extern userparameters userinputs;
extern timeparameters present;
extern male bonmales[PopSize];
extern male thdmales[PopSize];
extern int bookeep[PopSize];
void inputs(void)
{
//*****************make sure all variables are empty!*************
for (int a = 0; a < PopSize; a++)
{
bookeep[a] = 0;
thdmales[a].sum = 0;
bonmales[a].sum = 0;
thdmales[a].mate = 0;
bonmales[a].mate = 0;
277
for (int b = 0; b < attributes; b++)
{
thdmales[a].phenotype[b] = 0;
bonmales[a].phenotype[b] = 0;
}
}
present.genernum = 0;
present.iternum = 0;
present.thdmated = 0;
present.thdvalue = 0;
present.bonmated = 0;
present.bonvalue = 0;
userinputs.threshold = 0;
userinputs.cutoff = 0;
userinputs.excel = 0;
userinputs.file = 0;
userinputs.iterations = 0;
userinputs.generations = 0;
userinputs.screen = 0;
//User provides information about how many Iterations
cout << "\n";
cout << "
SIMULATION PARAMETERS ";
cout << "\n";
cout << "Please enter the number of iterations";
cout << "\n";
cin >> userinputs.iterations;
cout << "\n";
cout << "\n";
cout << "Please enter the number of generations in each iteration";
cout << "\n";
cout << "(must be between 0 and ";
cout << maxgen;
cout << " )";
cout << "\n";
cin >> userinputs.generations;
if (userinputs.generations < 0)
userinputs.generations = 0;
if (userinputs.generations > maxgen)
(userinputs.generations = maxgen);
cout << "\n";
cout << "\n";
cout << "
THRESHOLD CHOICE PARAMETERS ";
cout << "\n";
cout << "Please enter a number between 0 and ";
cout << maxthresh;
cout << "\n";
cout << " to represent the threshold below which males are unacceptable:";
cout << "\n";
cin >> userinputs.threshold;
if (userinputs.threshold < 0)
userinputs.threshold = 0;
if (userinputs.threshold > maxthresh)
(userinputs.threshold = maxthresh);
278
cout << "\n";
cout << "\n";
cout << "
BEST-OF-N CHOICE PARAMETERS ";
cout << "\n";
cout << "Please enter the number of males who will be allowed to mate";
cout << "\n";
cout << "(must be between 0 and ";
cout << PopSize;
cout << " )";
cout << "\n";
cin >> userinputs.cutoff;
if (userinputs.cutoff < 0)
userinputs.cutoff = 0;
if (userinputs.cutoff > PopSize)
userinputs.cutoff = PopSize;
cout << "\n";
cout << "\n";
cout << "
MUTATION PARAMETERS ";
cout << "\n";
cout << "There are ";
cout << attributes;
cout << " inheritable attributes.";
cout << "\n";
cout << " Enter maximum number of attributes that will mutate each generation: ";
cin >> userinputs.mutation;
if (userinputs.mutation < 0)
userinputs.mutation = 0;
if (userinputs.mutation > attributes)
userinputs.mutation = attributes;
cout << "\n";
cout << "At the beginning of the simulation, phenotype attribute values ";
cout << "\n";
cout << " are randomly set to be 0 <= attribute < ";
cout << pickvalue;
cout << "\n";
cout << "Please enter a value between 0 and ";
cout << maxthresh;
cout << "\n";
cout << " so that mutations will be 0 <= mutation < value: ";
cin >> userinputs.mutvalue;
if (userinputs.mutvalue < 1)
userinputs.mutvalue = 1;
if (userinputs.mutvalue > maxthresh)
userinputs.mutvalue = maxthresh;
cout << "\n";
//Give the user a choice to write output to a file or screen.
cout << "\n";
cout << "
OUTPUT OPTIONS ";
cout << "\n";
cout << ("Would you like to write the output to a text file \n");
cout << ("called 'data.txt' (y/n) ? ");
cin >> userinputs.file;
cout << "\n";
279
cout << ("Would you like to write the output to
cout<<("an Excel compatible, tab-delimited text
cout<<("called 'data.xl'
(y/n) ? ");
cin >> userinputs.excel;
cout << "\n";
cout << ("Would you like to write the output to
cout << ("(this will slow down the simulation)
cin >> userinputs.screen;
cout << "\n";
\n");
file \n");
the screen \n");
(y/n) ? ");
//If file output is chosen,
// Write summary data to output data.txt, the text file
if (userinputs.file == 121)
{
ofstream fout3(file3, ios::out | ios::app);
if (!fout3.good())
// or if (!fin)
{
cerr << "Can't open " << file3 << " file for output.\n";
exit(1);
}
fout3 << "\n";
fout3 << "Iterations in this simulation: ";
fout3 << userinputs.iterations;
fout3 << "\n";
fout3 << "Generations in each iteration: ";
fout3 << userinputs.generations;
fout3
fout3
fout3
fout3
<<
<<
<<
<<
"\n";
"There are ";
PopSize;
" males in each population.";
fout3 << "\n";
fout3 << "In threshold choice, males below this cutoff are rejected: ";
fout3 << userinputs.threshold;
fout3 << "\n";
fout3 << "Maximum threshold is: ";
fout3 << maxthresh;
fout3
fout3
fout3
fout3
<<
<<
<<
<<
"\n";
"In Best-of-N choice, the top ";
userinputs.cutoff;
" males are allowed to mate.";
fout3 << "\n";
fout3 << "Number of attributes in each phenotype: ";
fout3 << attributes;
fout3 << "\n";
fout3 << "Number of phenotypic attributes allowed to mutate in each generation: ";
fout3 << userinputs.mutation;
fout3 << "\n";
fout3 << "Starting value for phenotypic attributes 0 <= attribute < ";
fout3 << pickvalue;
fout3 << "\n";
280
fout3 << "Mutations to phenotypic attributes 0 <= mutation < ";
fout3 << userinputs.mutvalue;
fout3 << "\n";
fout3.close();
}
}
(* * * * * * * *)
iteration.cpp:
//One iteration
#include <stdlib.h>
#include "header.h"
#include "param.h"
extern userparameters userinputs;
extern timeparameters present;
//extern int tag;
void iterate(void)
{
wipe();
fillfirstgen();
present.genernum = 0;
//tag = 0;
for (int b = 0; b < userinputs.generations; b++)
{
generation();
//output();
fillnextgen();
present.genernum = (present.genernum +1);
}
}
(* * * * * * * *)
picker.cpp:
//function definition for picker, which picks a random integer between -1 and n
//Uses modulus operator to scale the random number.
#include
#include
#include
#include
<math.h>
<stdlib.h>
"header.h"
"param.h"
int picker(int n)
{
return rand() % n;
}
(* * * * * * * *)
simulate.cpp:
//Runs the simulation
281
#include <stdlib.h>
#include "header.h"
#include "param.h"
extern userparameters userinputs;
extern timeparameters present;
extern int thdcrash[maxgen];
extern thdsum[maxgen];
extern bonsum[maxgen];
void simulate(void)
{
for (int aa = 0; aa
{
thdsum[aa] =
bonsum[aa] =
thdcrash[aa]
}
< maxgen; aa++)
//Clear out summary information
0;
0;
= 0;
present.iternum = 0;
for (int a = 0; a < userinputs.iterations; a++)
{
iterate();
present.iternum = (present.iternum +1);
}
summarize();
}
(* * * * * * * *)
summarize.cpp:
//Provide runtime screen and/or file output. Screen output slows the simulation;
//if screen output is not chosen, the procedure Simulate reports only
//the number of generations until fixation. Two file outputs are possible; one
//is a text file similar to screen output but containing input fields as well; the
//other is a tab-delimited text file that can be read by spreadsheet, graphics,
//or statistics packages. In Macintosh, since output files have no creator
//type, output files can only be read for the first time by a text editor such as BBedit.
#include
#include
#include
#include
#include
#include
<iostream.h>
<fstream.h>
<stdlib.h>
"header.h"
"param.h"
<math.h>
// for exit()
const char * afile3 = "data.txt";
const char * afile4 = "data.xl";
const int Len = 40;
extern timeparameters present;
extern int thdcrash[maxgen];
extern float thdsum[maxgen];
extern float bonsum[maxgen];
282
extern userparameters userinputs;
void summarize()
{
//SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS
if (userinputs.screen == 121)
{
cout << "\n";
cout.setf(ios::fixed, ios::floatfield);
cout << "\n";
//force fixed point display
for (int a = 0; a < userinputs.generations; a++)
{
cout << "Average value (thdcrashes thd bn) for generation ";
cout << (a + 1);
cout << ": ";
cout << thdcrash[a];
cout << " ";
if (a == 0)
cout << (thdsum[a]);
if (a > 0)
//must adjust numbers for times population has crashed
cout << ((thdsum[a]* userinputs.iterations)/
(userinputs.iterations - thdcrash[(a-1)]));
cout << " ";
cout << (bonsum[a]);
cout << "\n";
}
}
cout << "\n";
cout << "\n";
//SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS
// Write data to output afile1, the text file
if (userinputs.file == 121)
{
ofstream fout3(afile3, ios::out | ios::app);
if (!fout3.good())
// or if (!fin)
{
cerr << "Can't open " << afile3 << " file for output.\n";
exit(1);
}
//FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
//Code between these F's is written to file
fout3.setf(ios::fixed, ios::floatfield);
//force fixed point display
fout3 << "\n";
for (int b = 0;
{
fout3 <<
fout3 <<
fout3 <<
b < userinputs.generations; b++)
"Average value (thdcrashes thd bn) for generation ";
(b + 1);
": ";
283
fout3 << thdcrash[b];
fout3 << " ";
if (b == 0)
fout3 << (thdsum[b]);
if (b > 0)
//must adjust numbers for times population has crashed
fout3 << ((thdsum[b]* userinputs.iterations)/
(userinputs.iterations - thdcrash[(b-1)]));
fout3 << " ";
fout3 << bonsum[b];
fout3 << "\n";
}
fout3.close();
}
// Write data to output afile2, the Excel file
if (userinputs.excel == 121)
{
ofstream fout4(afile4, ios::out | ios::app);
if (!fout4.good())
// or if (!fin)
{
cerr << "Can't open " << afile4 << " file for output.\n";
exit(1);
}
fout4.setf(ios::fixed, ios::floatfield);
fout4 << "\n";
for (int c = 0;
{
fout4 <<
fout4 <<
fout4 <<
fout4 <<
if (c ==
//force fixed point display
c < userinputs.generations; c++)
(c + 1);
"\t";
thdcrash[c];
"\t";
0)
fout4 << (thdsum[c]);
if (c > 0)
//must adjust numbers for times population has crashed
fout4 << ((thdsum[c]* userinputs.iterations)/
(userinputs.iterations - thdcrash[(c-1)]));
fout4 << "\t";
fout4 << bonsum[c];
fout4 << "\n";
}
int cumulative = no;
//Set to "yes" to have cumulative counts of thd crashes.
if (cumulative == no)
for (int d = (maxgen - 1); d > 0; d = (d - 1))
thdcrash[d] = (thdcrash[d] - thdcrash[d-1]);
for (c = 0; c < userinputs.generations; c++)
{
fout4 << thdcrash[c];
fout4 << "\n";
}
fout4.close();
}
//FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
}
284
(* * * * * * * *)
threshold.cpp:
//function definition for evaluating whether males pass threshold and
//marking ones that do.
#include
#include
#include
#include
<math.h>
<stdlib.h>
"header.h"
"param.h"
extern userparameters userinputs;
extern male thdmales[PopSize];
extern timeparameters present;
extern int thdcrash[maxgen];
//extern int tag;
void threshold(void)
{
present.thdmated = 0; //Set mating information to zero
present.thdvalue = 0; //Set phenotypes of mated males to zero
for (int a = 0; a < PopSize; a++)
//Calculate male phenotype
{
for (int b = 0; b < attributes; b++)
{
thdmales[a].sum = (thdmales[a].sum + thdmales[a].phenotype[b]);
}
thdmales[a].mate = no;
}
for (a = 0; a < PopSize; a++)
{
if (thdmales[a].sum >= userinputs.threshold) //test each male
{
thdmales[a].mate = yes;
(present.thdmated = present.thdmated + yes);
present.thdvalue = (present.thdvalue + thdmales[a].sum);
}
}
if (present.thdmated == 0)
{
//tag = yes;
thdcrash[present.genernum] = (thdcrash[present.genernum] + 1);
}
if (present.thdmated == 0)
for (a = 0; a < PopSize; a++)
{
thdmales[a].sum = 0;
thdmales[a].mate = 0;
for (int b = 0; b < attributes; b++)
thdmales[a].phenotype[b] = 0;
}
if (present.thdmated > 0)
present.thdvalue = (present.thdvalue/present.thdmated);
285
}
(* * * * * * * *)
timeseed.cpp:
//Seeds random number generator using internal clock
#include
#include
#include
#include
<stdlib.h>
<time.h>
"header.h"
"param.h"
//includes all function prototypes
//includes global constants
void timeseed(void)
{
time_t t;
srand((unsigned) time(&t));
//Seed Random number generator
//srand(time(0));
//Alternative seed Random number generator
}
(* * * * * * * *)
wipe.cpp:
// Wipes variables clean
#include <stdlib.h>
#include "header.h"
#include "param.h"
//includes all function prototypes
//includes global constants
extern userparameters userinputs;
extern timeparameters present;
extern male bonmales[PopSize];
extern male thdmales[PopSize];
extern int bookeep[PopSize];
void wipe(void)
{
present.genernum =
present.thdmated =
present.thdvalue =
present.bonmated =
present.bonvalue =
present.crash = 0;
0;
0;
0;
0;
0;
for (int a = 0; a < PopSize; a++)
{
bookeep[a] = 0;
thdmales[a].sum = 0;
bonmales[a].sum = 0;
thdmales[a].mate = 0;
bonmales[a].mate = 0;
for (int b = 0; b < attributes; b++)
{
thdmales[a].phenotype[b] = 0;
bonmales[a].phenotype[b] = 0;
}
}
}
286
APPENDIX B
Pascal program code for comparing experimental cage densities to densities of free
cicadas observed on cage-sized branches. (Think Lightspeed Pascal for the Macintosh)
program Cage;
const
ObservationNumber = 39;
var
FarmScan: array[0..39] of integer;
greaterdiff, iterations, randomnumber, TargetDensity: integer;
Percgreaterdiff: real;
{* Seeds Random Number Generator}
procedure TIMESEED;
begin
GetDateTime(randseed);
end;
procedure Welcome;
begin
writeln('Welcome to Resampler. You can run up to 10,000
resamplings');
writeln(' of female cicada branch scan data. ');
end;
{*Picks a random number, scales it to the appropriate size}
procedure PICKRAND;
begin
randomnumber := (Trunc(((ABS(Random)) / 32768) *
ObservationNumber));
{writeln('Picked slot ', randomnumber);}
end;
procedure HardData;
begin
FarmScan[0] := 7;
FarmScan[1] := 12;
FarmScan[2] := 14;
FarmScan[3] := 7;
FarmScan[4] := 7;
FarmScan[5] := 11;
FarmScan[6] := 16;
FarmScan[7] := 22;
FarmScan[8] := 17;
FarmScan[9] := 22;
FarmScan[10] := 6;
FarmScan[11] := 9;
FarmScan[12] := 12;
FarmScan[13] := 13;
FarmScan[14] := 12;
FarmScan[15] := 27;
FarmScan[16] := 3;
FarmScan[17] := 11;
FarmScan[18] := 10;
FarmScan[19] := 13;
FarmScan[20] := 17;
287
FarmScan[21]
FarmScan[22]
FarmScan[23]
FarmScan[24]
FarmScan[25]
FarmScan[26]
FarmScan[27]
FarmScan[28]
FarmScan[29]
FarmScan[30]
FarmScan[31]
FarmScan[32]
FarmScan[33]
FarmScan[34]
FarmScan[35]
FarmScan[36]
FarmScan[37]
FarmScan[38]
FarmScan[39]
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
:=
26;
7;
15;
16;
17;
4;
11;
9;
16;
24;
15;
5;
13;
10;
8;
16;
11;
7;
12;
end;
{Gets Input parameters}
procedure INPUTS;
begin
writeln('Please provide the target density of cicadas');
read(TargetDensity);
writeln('Please provide the number of iterations');
writeln('of the simulation , maximum 10000 : ');
read(iterations);
end;
{Checks Input parameters}
procedure CHECKVAR;
begin
if (iterations > 10000) or (iterations < 1) then
begin
writeln('Sorry, but iterations must be an integer
less than 10000');
writeln('program has been terminated');
HALT;
end;
end;
procedure Resample;
var
counter: integer;
begin
pickrand;
{writeln('Picked number ', farmscan[randomnumber]);}
if (ABS((12.6 - targetdensity)) < ABS((farmscan[randomnumber] 12.6))) then
greaterdiff := (greaterdiff + 1);
end;
procedure iterate;
var
counter: integer;
begin
greaterdiff := 0;
for counter := 1 to iterations do
begin
resample;
end;
288
end;
procedure MakeDistribution;
begin
Percgreaterdiff := greaterdiff / iterations;
writeln(' ');
writeln('percent of ', iterations, ' iterations');
writeln('in which picked number was farther from mean of 12.6
than');
writeln('target value = ', targetdensity);
writeln(' ');
writeln(Percgreaterdiff);
end;
{The shell of the program}
procedure MAKEDECISION;
var
choicer: integer;
begin
writeln('
');
writeln('Would you like to:');
writeln('
(1) Resample ?');
writeln('
(2) End this session?');
writeln(' ');
writeln('Choose (1) or (2) please');
read(choicer);
if choicer = 1 then
begin
Timeseed;
Welcome;
inputs;
checkvar;
HardData;
iterate;
MakeDistribution;
Makedecision;
end;
if choicer = 2 then
HALT;
if (choicer < 1) or (choicer > 2) then
begin
writeln('Hey there are only two choices here!');
Makedecision;
end;
end;
begin
MakeDecision;
end.
289
APPENDIX C
Of the five bag cage experiments performed in 1995, in only two were females allowed to
remate. Both were male silencing experiments designed to assess the influence of male
courtship song on mating success. Unmated females were given equal numbers of
normal and silenced males with which to mate.
Silencing experiment I
8 unmated females in each of 3 cages. 11 Days
Also in each cage 8 normal males, 8 silenced males.
15/28 first matings by silenced males,
13/28 by unmanipulated males.
28 total matings, 6 rematings.
2 cages with 2 rematings each, 1 cage with 4 rematings.
1 female no matings.
Silencing experiment II
4 unmated females into each of 4 cages. 11 Days
Also in each cage 4 normal and 4 silenced males
7/14 first matings by normal males
7/14 by silenced males.
14 total matings, 3 rematings.
3 cages with 1 remating each.
2 females no matings.
Male quality, at least as affected by the silencing treatment, is irrelevant because
the treatment had no effect on male mating success. For the purposes of the resampling
analysis, these experiments can be treated as bag cages in which females had the
opportunity to mate multiply. Female remating frequencies in these cages can be
compared to remating frequencies in the large flight cage. Although the flight cage data
covers a period of 12 days, and the bag cage experiments ran for 11 days, these time
periods are comparable, since the number of days is less important than day quality (in
terms of weather, etc.). In any case, all but one mating in the flight cage occurred before
the cicada’s tenth day from tenerality, well before the end of the flight and bag cage
experiments.
290
APPENDIX D
C++ code to evaluate remating frequencies in large and small cages. Subroutines
separated by asterisks (************).
******************************
header.h
#include "param.h"
//Sets some global parameters
//function prototypes for simulation
void choicer(void);
void simulate(void);
void inputs(void);
void iterate(void);
int picker(int n);
void clearfemales(void);
void fillfemales(void);
void output(void);
//gives the user a choice of functions
//The simulation
//gets user inputs
//iterates the simulation
// pick a random integer between 0 and n
//clears female data array
//fills female data array with plug info.
//writes output to disk or screen
//data structure to hold user inputs
struct userparameters
{
char full;
char excel;
int iterations;
char file;
char screen;
};
*****************************************
param.h
// Constants for program, based on 1995 cage density experiments
#define Maxfemales 78 //Size of female population
#define Totalremate 13 //Number of females who remated
**************************************
choicer.cp
// choicer-- runs the loop in which the simulation occurs. Gives the
//user a choice of entering new parameters, using already entered //parameters, or
stopping. First simulation occurs in Main outside this
//loop; then Choicer is called and you get a choice of whether to
//continue or not
#include <iostream.h>
#include <stdlib.h>
// for exit()
291
#include "header.h"
#include "param.h"
extern
extern
extern
extern
userparameters userinputs;
int tally;
int ptile;
int iternum;
void choicer()
{
int select;
cout << "\n";
cout << "\n";
cout << "Would
cout << "\n";
cout << " (1)
cout << "\n";
cout << " (2)
cout << "\n";
cout << " (3)
cout << "\n";
cin >> select;
cout << "\n";
cout << "\n";
you like to ";
run a simulation with new parameters? ";
run a simulation using previous parameters? ";
end the simulations? ";
if (select == 1)
{
ptile = 0;
tally = 0;
iternum =1;
inputs();
iterate();
choicer();
}
if (select == 2)
{
ptile = 0;
tally = 0;
iternum =1;
iterate();
choicer();
}
//Run simulation with new input parameters
//Get input parameters
//Run the simulation
//Choose what to do
//Run simulation with old parameters
//Run the simulation
//Choose what to do
if (select == 3)
//Stop the simulations
{
select = select;
//Basically nonsense-- just pop back out
}
}
***********************
clearfemales.cp
#include <stdlib.h>
#include "header.h"
292
#include "param.h"
extern int females[Maxfemales];
void clearfemales(void)
{
for (int i = 0; i < Maxfemales; i++)
{
females[i] = 0;
}
}
**************************
fillfemales.cp
#include <stdlib.h>
#include "header.h"
#include "param.h"
extern int females[Maxfemales];
void fillfemales(void)
{
for (int i = 0; i < Totalremate; i++)
{
females[i] = 1;
}
}
**************************
inputs.cpp
//This function provides prompts for user inputs and writes summary //information to a
text file, if desired. Most inputs are checked upon
//entry and set to a default value if user input exceeds acceptable
//value.
#include
#include
#include
#include
#include
<iostream.h>
<fstream.h>
<stdlib.h>
"header.h"
"param.h"
// for exit()
const char * file1 = "output1.txt";
const char * file2 = "outXl.txt";
const int Len = 40;
extern
extern
extern
extern
int females[Maxfemales];
userparameters userinputs;
int TargetValue;
int CageSize;
void inputs(void)
{
cout
cout
cout
cout
cout
<<
<<
<<
<<
<<
"\n";
"
Cage Size Effect Resampler ";
"\n";
"\n";
"Cage size effect simulation";
293
cout << "\n";
cout << "Based on 1995 Data";
cout << "\n";
//Tell user how many total females may be in the simulation-- Max
//number is set in the param.h file.
cout << "\n";
cout << "
FEMALE POPULATION CHARACTERISTICS ";
cout << "\n";
cout<< "There are ";
cout << Maxfemales;
cout << " females in the flight cage";
cout << "\n";
cout<< "A total of ";
cout << Totalremate;
cout << " females in the flight cage remated";
cout << "\n";
cout << "\n";
//User provides information about how many Iterations
cout << "\n";
cout << "
SIMULATION PARAMETERS ";
cout << "\n";
cout << "Please enter the number of iterations";
cout << "\n";
cin >> userinputs.iterations;
if (userinputs.iterations < 1)
userinputs.iterations = 0;
cout << "\n";
//User provides information about Small Cages
cout << "\n";
cout << "
Characteristics of Small Cages ";
cout << "\n";
cout << "Please enter the number of females in a small cage";
cout << "\n";
cin >> CageSize;
if (CageSize < 1)
CageSize = 0;
if (CageSize > Maxfemales)
CageSize = Maxfemales;
cout << "\n";
cout << "Please enter the observed number of rematings in the small cage";
cout << "\n";
cin >> TargetValue;
if (TargetValue < 1)
TargetValue = 0;
if (TargetValue > CageSize)
TargetValue = CageSize;
cout << "\n";
294
//Give the user a choice to write output to a file or screen.
cout << "\n";
cout << "
OUTPUT OPTIONS ";
cout << "\n";
cout << ("Would you like to write full output? \n");
cout << ("(instead of abbreviated output) (y/n) ? ");
cin >> userinputs.full;
cout << "\n";
cout << ("Would you like to write the output to a text file \n");
cout << ("called 'output1.txt' (y/n) ? ");
cin >> userinputs.file;
cout << "\n";
if (userinputs.full == 121)
{
cout << ("Would you like to write the output to \n");
cout<<("an Excel compatible, tab-delimited text file \n");
cout<<("called 'outXl.txt'
(y/n) ? ");
cin >> userinputs.excel;
cout << "\n";
}
cout << ("Would you like to write the output to the screen \n");
cout << ("(this will slow down the simulation) (y/n) ? ");
cin >> userinputs.screen;
cout << "\n";
//If file output is chosen,
// Write summary data to output file1, the text file
if (userinputs.file == 121)
{
ofstream fout(file1, ios::out | ios::app);
if (!fout.good())
// or if (!fin)
{
cerr << "Can't open " << file1 << " file for output.\n";
exit(1);
}
fout
fout
fout
fout
<<
<<
<<
<<
"\n";
"Iterations in this simulation: ";
userinputs.iterations;
"\n";
fout << "\n";
fout << "\n";
fout<< "There are ";
fout << Maxfemales;
fout << " females in the flight cage";
fout << "\n";
fout<< "A total of ";
fout << Totalremate;
fout << " females in the flight cage remated";
fout << "\n";
fout<< "There are ";
295
fout << CageSize;
fout << " females in the small cage";
fout << "\n";
fout<< "A total of ";
fout << TargetValue;
fout << " females in the small cage remated";
fout << "\n";
fout << "\n";
fout << "\n";
fout.close();
}
}
*******************************************
iterate.cp
#include
#include
#include
#include
<stdlib.h>
<time.h>
"header.h"
"param.h"
//includes all function prototypes
//includes global constants
extern userparameters userinputs;
extern int iternum;
void iterate()
{
for (int i = 0; i < userinputs.iterations; i++)
{
simulate();
iternum = iternum +1;
}
}
*********************************************
main.cp
//Simulation to analyze cage size effect data
#include <stdlib.h>
#include <time.h>
#include "header.h"
//includes all function prototypes
#include "param.h"
//includes global constants
//initialize an integer to tally number of plugged females picked in each turn.
int tally;
//initialize an integer to generate percentile
int ptile;
//initialize an integer to hold size of small cage we're looking at
int CageSize;
296
//initialize an integer to look at target value of matings in small cage
int TargetValue;
//initialize array for females
int females[Maxfemales];
//Initialize an integer to keep track of which iteration we're on
//so that output can display the results properly
int iternum =1;
//Declare two characters to control file output.
char write1;
char write2;
//Initialize two data structures to hold user inputs and present
// iteration/generation values.
userparameters userinputs;
int main(void)
{
//********************Start the main program**************************
//Clear out userparameters
clearfemales();
fillfemales();
tally = 0;
userinputs.iterations = 0;
userinputs.excel = 0;
userinputs.screen = 0;
userinputs. file = 0;
ptile= 0;
srand(time(0));
//Seed Random number generator
//for the first time through, you don't get a choice of what to do
inputs();
iterate();
//Get input parameters
//Run the simulation
//After the first time through, you get a choice.
choicer();
return 0;
//Run the loop
//End the program
}
***************************************************
output.cp
//Provide runtime screen and/or file output. Screen output slows the //simulation;if
screen output is not chosen, the procedure Simulate
//reports only the number of generations until fixation. Two file
//outputs are possible; one is a text file similar to screen output but //containing
input fields as well; the other is a tab-delimited text
//file that can be read by spreadsheet, graphics,or statistics packages.
//In Macintosh, since output files have no creator type, output files
//can only be read for the first time by a text editor such as BBedit.
297
#include
#include
#include
#include
#include
#include
<iostream.h>
<fstream.h>
<stdlib.h>
"header.h"
"param.h"
<math.h>
// for exit()
const char * afile1 = "output1.txt";
const char * afile2 = "outXl.txt";
const int Len = 40;
extern
extern
extern
extern
extern
int females[Maxfemales];
userparameters userinputs;
int tally;
int ptile;
int TargetValue;
extern int iternum;
void output()
{
//SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS
float percentile;
cout.setf(ios::fixed, ios::floatfield);
cout.setf(ios::showpoint);
cout.precision(3);
if (userinputs.screen == 121)
{
if (userinputs.full == 121)
{
cout << "\n";
cout << "\n";
cout << "ITERATION ";
cout << (iternum);
cout << "\n";
cout << "\n";
cout << "Number of rematings is
cout << tally;
cout << "\n";
}
}
";
if (iternum == userinputs.iterations)
{
cout << "\n";
cout << "\n";
cout << "Percentage of iterations in which
cout << TargetValue;
cout << "\n";
cout << "or more rematings occurred: ";
";
percentile = ptile;
//Tricky way this has to be
//calculated to
percentile = percentile/userinputs.iterations;
//avoid implicit conversion
cout << percentile;
298
cout << "\n";
}
//SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS
// Write data to output afile1, the text file
if (userinputs.file == 121)
{
ofstream fout(afile1, ios::out | ios::app);
if (!fout.good())
// or if (!fin)
{
cerr << "Can't open " << afile1 << " file for output.\n";
exit(1);
}
fout.setf(ios::fixed, ios::floatfield);
//force fixed point display
//FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
//Code between these F's is written to file
if (userinputs.full == 121)
{
fout << "\n";
fout << "\n";
fout << "ITERATION ";
fout << (iternum);
fout << "\n";
fout << "\n";
fout << "Number of rematings is
fout << tally;
fout << "\n";
";
}
if (iternum
{
fout
fout
fout
fout
fout
fout
fout
fout
}
== userinputs.iterations)
<<
<<
<<
<<
<<
<<
<<
<<
"\n";
"\n";
"Percentage of iterations in which
TargetValue;
"\n";
"or more rematings occurred: ";
percentile;
"\n";
";
fout.close();
}
if (userinputs.full == 121)
{
// Write data to output afile2, the Excel file
if (userinputs.excel == 121)
{
ofstream fout2(afile2, ios::out | ios::app);
if (!fout2.good())
// or if (!fin)
{
299
cerr << "Can't open " << afile2 << " file for output.\n";
exit(1);
}
fout2 << "\n";
fout2 << (tally);
fout2.close();
}
//FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
}
}
******************************************
picker.cp
//function definition for picker, which picks a random integer between
//-1 and n
//Uses modulus operator to scale the random number.
#include
#include
#include
#include
<math.h>
<stdlib.h>
"header.h"
"param.h"
int picker(int n)
{
return rand() % n;
}
*********************************************
simulate.cp
#include
#include
#include
#include
<stdlib.h>
<time.h>
"header.h"
"param.h"
//includes all function prototypes
//includes global constants
extern userparameters userinputs;
extern int females[Maxfemales];
extern
extern
extern
extern
int
int
int
int
tally;
ptile;
TargetValue;
CageSize;
void simulate()
{
for (int i = 0; i < CageSize; i++)
300
{
int pick = 0;
pick = picker(Maxfemales);
//Pick a random female
tally = (tally + females[pick]);
//Tally up number of rematings in this iteration
}
if (tally >= TargetValue)
//test number of rematings in this iteration
ptile = (ptile +1);
//prepare to set up percentiles
output();
tally = 0;
}
301
APPENDIX E
C++ code for seminal plug resampling algorithm. Subroutines separated by
asterisks (* * * * * * * *).
param.h:
// Constants for program, based on 1996 Farmville data,
//mating interruption experiment,
//females interrupted after 1 or 2 hours, assayed for plugs,
//and allowed to remate.
#define
#define
#define
#define
Maxfemales 36 //Size of female population
Pluggedfemales 11 //Number of plugged females
Totalremate 20 //Number of females who remated
Pluggedremate 4 //Number of plugged females who remated
(* * * * * * * *)
header.h:
#include "param.h"
//Sets some global parameters
//function prototypes for simulation
void choicer(void);
void simulate(void);
void inputs(void);
void iterate(void);
int picker(int n);
void clearfemales(void);
void fillfemales(void);
void output(void);
//gives the user a choice of functions
//The simulation
//gets user inputs
//iterates the simulation
// pick a random integer between 0 and n
//clears female data array
//fills female data array with plug info.
//writes output to disk or screen
//data structure to hold user inputs
struct userparameters
{
char full;
char excel;
int iterations;
char file;
char screen;
};
(* * * * * * * *)
main.cp:
//Simulation to analyze 1996 plugged female remating data.
//male polygyny simulation.
#include <stdlib.h>
#include <time.h>
#include "header.h"
Based on
//includes all function prototypes
302
#include "param.h"
//includes global constants
//initialize an integer to tally number of plugged females picked in each turn.
int tally;
//initialize an integer to generate percentile
int ptile;
//initialize array for females
int females[Maxfemales];
//Initialize an integer to keep track of which iteration we're on
//so that output can display the results properly
int iternum =1;
//Declare two characters to control file output.
char write1;
char write2;
//Initialize two data structures to hold user inputs and present
// iteration/generation values.
userparameters userinputs;
int main(void)
{
//********************Start the main program**************************
//Clear out userparameters
clearfemales();
fillfemales();
tally = 0;
userinputs.iterations = 0;
userinputs.excel = 0;
userinputs.screen = 0;
userinputs. file = 0;
ptile= 0;
srand(time(0));
//Seed Random number generator
//for the first time through, you don't get a choice of what to do
inputs();
iterate();
//Get input parameters
//Run the simulation
//After the first time through, you get a choice.
choicer();
return 0;
//Run the loop
//End the program
}
(* * * * * * * *)
inputs.cp:
//This function provides prompts for user inputs and writes summary information
//to a text file, if desired. Most inputs are checked upon entry and set
303
//to a default value if user input exceeds acceptable value.
#include
#include
#include
#include
#include
<iostream.h>
<fstream.h>
<stdlib.h>
"header.h"
"param.h"
// for exit()
const char * file1 = "output1.txt";
const char * file2 = "outXl.txt";
const int Len = 40;
extern int females[Maxfemales];
extern userparameters userinputs;
void inputs(void)
{
cout
cout
cout
cout
cout
cout
cout
cout
cout
cout
cout
cout
<<
<<
<<
<<
<<
<<
<<
<<
<<
<<
<<
<<
"\n";
"
PLUGGED FEMALE REMATING SIMULATION ";
"\n";
"\n";
"Plugged female remating simulation in which females";
"\n";
"whose first mating is interrupted are assayed for a";
"\n";
"sperm plug and allowed to remate.";
"\n";
"Based on 1996 Farmville Data";
"\n";
//Tell user how many total females may be in the simulation-- Max
//number is set in the param.h file.
cout << "\n";
cout << "
FEMALE POPULATION CHARACTERISTICS ";
cout << "\n";
cout<< "There are ";
cout << Maxfemales;
cout << " females in the population";
cout << "\n";
cout<< "There are ";
cout << Pluggedfemales;
cout << " plugged females in the population";
cout << "\n";
cout<< "A total of ";
cout << Totalremate;
cout << " females in the population will remate";
cout << "\n";
cout << "\n";
304
//User provides information about how many Iterations
cout << "\n";
cout << "
SIMULATION PARAMETERS ";
cout << "\n";
cout << "Please enter the number of iterations";
cout << "\n";
cin >> userinputs.iterations;
if (userinputs.iterations < 1)
userinputs.iterations = 0;
cout << "\n";
//Give the user a choice to write output to a file or screen.
cout << "\n";
cout << "
OUTPUT OPTIONS ";
cout << "\n";
cout << ("Would you like to write full output? \n");
cout << ("(instead of abbreviated output) (y/n) ? ");
cin >> userinputs.full;
cout << "\n";
cout << ("Would you like to write the output to a text file \n");
cout << ("called 'output1.txt' (y/n) ? ");
cin >> userinputs.file;
cout << "\n";
if (userinputs.full == 121)
{
cout << ("Would you like to write the output to \n");
cout<<("an Excel compatible, tab-delimited text file \n");
cout<<("called 'outXl.txt'
(y/n) ? ");
cin >> userinputs.excel;
cout << "\n";
}
cout << ("Would you like to write the output to the screen \n");
cout << ("(this will slow down the simulation) (y/n) ? ");
cin >> userinputs.screen;
cout << "\n";
//If file output is chosen,
// Write summary data to output file1, the text file
if (userinputs.file == 121)
{
ofstream fout(file1, ios::out | ios::app);
if (!fout.good())
// or if (!fin)
{
cerr << "Can't open " << file1 << " file for output.\n";
exit(1);
}
fout
fout
fout
fout
<<
<<
<<
<<
"\n";
"Iterations in this simulation: ";
userinputs.iterations;
"\n";
fout << "\n";
fout << "\n";
305
fout<< "There are ";
fout << Maxfemales;
fout << " females in the population";
fout << "\n";
fout<< "There are ";
fout << Pluggedfemales;
fout << " plugged females in the population";
fout << "\n";
fout<< "A total of ";
fout << Totalremate;
fout << " females in the population will remate";
fout << "\n";
fout<< "A total of ";
fout << Pluggedremate;
fout << " plugged females in the 1996 population remated";
fout << "\n";
fout << "\n";
fout << "\n";
fout.close();
}
}
(* * * * * * * *)
iterate.cp
#include <stdlib.h>
#include <time.h>
#include "header.h"
#include "param.h"
//includes all function prototypes
//includes global constants
extern userparameters userinputs;
extern int iternum;
void iterate()
{
for (int i = 0; i < userinputs.iterations; i++)
{
simulate();
iternum = iternum +1;
}
}
(* * * * * * * *)
simulate.cp:
#include <stdlib.h>
#include <time.h>
#include "header.h"
//includes all function prototypes
306
#include "param.h"
//includes global constants
extern userparameters userinputs;
extern int females[Maxfemales];
extern int tally;
extern int ptile;
void simulate()
{
for (int i = 0; i < Totalremate; i++)
{
int pick = 0;
pick = picker(Maxfemales);
//Pick a random female
tally = (tally + females[pick]);
//Tally up number of matings in this iteration
}
if (tally <= Pluggedremate)
//test number of plugged matings in this iteration
ptile = (ptile +1);
//prepare to set up percentiles
output();
tally = 0;
}
(* * * * * * * *)
fillfemales.cp:
#include <stdlib.h>
#include "header.h"
#include "param.h"
extern int females[Maxfemales];
void fillfemales(void)
{
for (int i = 0; i < Pluggedfemales; i++)
{
females[i] = 1;
}
}
(* * * * * * * *)
clearfemales.cp:
#include <stdlib.h>
#include "header.h"
#include "param.h"
extern int females[Maxfemales];
void clearfemales(void)
{
307
for (int i = 0; i < Maxfemales; i++)
{
females[i] = 0;
}
}
(* * * * * * * *)
picker.cp:
//function definition for picker, which picks a random integer between -1 and n
//Uses modulus operator to scale the random number.
#include
#include
#include
#include
<math.h>
<stdlib.h>
"header.h"
"param.h"
int picker(int n)
{
return rand() % n;
}
(* * * * * * * *)
choicer.cp:
// choicer-- runs the loop in which the simulation occurs.
//Gives the user a choice
// of entering new parameters, using already entered parameters, or stopping.
//First simulation occurs in Main outside this loop; then Choicer is called and
// you get a choice of whether to continue or not
#include
#include
#include
#include
extern
extern
extern
extern
<iostream.h>
<stdlib.h>
"header.h"
"param.h"
// for exit()
userparameters userinputs;
int tally;
int ptile;
int iternum;
void choicer()
{
int select;
cout
cout
cout
cout
cout
cout
cout
cout
cout
cout
<<
<<
<<
<<
<<
<<
<<
<<
<<
<<
"\n";
"\n";
"Would
"\n";
" (1)
"\n";
" (2)
"\n";
" (3)
"\n";
you like to ";
run a simulation with new parameters? ";
run a simulation using previous parameters? ";
end the simulations? ";
308
cin >> select;
cout << "\n";
cout << "\n";
if (select == 1)
{
ptile = 0;
tally = 0;
iternum =1;
inputs();
iterate();
choicer();
}
//Run simulation with new input parameters
if (select == 2)
{
ptile = 0;
tally = 0;
iternum =1;
iterate();
choicer();
}
//Run simulation with old parameters
//Get input parameters
//Run the simulation
//Choose what to do
//Run the simulation
//Choose what to do
if (select == 3)
//Stop the simulations
{
select = select;
//Basically nonsense-- just pop back out to Main.
}
}
(* * * * * * * *)
output.cp:
//Provide runtime screen and/or file output. Screen output slows the simulation;
//if screen output is not chosen, the procedure Simulate reports only
//the number of generations until fixation. Two file outputs are possible; one
//is a text file similar to screen output but containing input fields as well; the
//other is a tab-delimited text file that can be read by spreadsheet, graphics,
//or statistics packages. In Macintosh, since output files have no creator
//type, output files can only be read for the first time by a text editor such as
//BBedit.
#include
#include
#include
#include
#include
#include
<iostream.h>
<fstream.h>
<stdlib.h>
"header.h"
"param.h"
<math.h>
// for exit()
const char * afile1 = "output1.txt";
const char * afile2 = "outXl.txt";
const int Len = 40;
extern
extern
extern
extern
int females[Maxfemales];
userparameters userinputs;
int tally;
int ptile;
extern int iternum;
309
void output()
{
//SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS
float percentile;
cout.setf(ios::fixed, ios::floatfield);
cout.setf(ios::showpoint);
cout.precision(3);
if (userinputs.screen == 121)
{
if (userinputs.full == 121)
{
cout << "\n";
cout << "\n";
cout << "ITERATION ";
cout << (iternum);
cout << "\n";
cout << "\n";
cout << "Number of matings by plugged females is
cout << tally;
cout << "\n";
}
}
if (iternum == userinputs.iterations)
{
cout << "\n";
cout << "\n";
cout << "Percentage of iterations in which
cout << Pluggedremate;
cout << "\n";
cout << "or fewer plugged females remated:
";
";
";
percentile = ptile;
//Tricky way this has to be
calculated to
percentile = percentile/userinputs.iterations;
//avoid implicit conversion
cout << percentile;
cout << "\n";
}
//SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS
// Write data to output afile1, the text file
if (userinputs.file == 121)
{
ofstream fout(afile1, ios::out | ios::app);
if (!fout.good())
// or if (!fin)
{
cerr << "Can't open " << afile1 << " file for output.\n";
exit(1);
}
fout.setf(ios::fixed, ios::floatfield);
//force fixed point display
//FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
310
//Code between these F's is written to file
if (userinputs.full == 121)
{
fout << "\n";
fout << "\n";
fout << "ITERATION ";
fout << (iternum);
fout << "\n";
fout << "\n";
fout << "Number of matings by plugged females is
fout << tally;
fout << "\n";
";
}
if (iternum == userinputs.iterations)
{
fout << "\n";
fout << "\n";
fout << "Percentage of iterations in which
fout << Pluggedremate;
fout << "\n";
fout << "or fewer plugged females remated:
fout << percentile;
fout << "\n";
}
";
";
fout.close();
}
if (userinputs.full == 121)
{
// Write data to output afile2, the Excel file
if (userinputs.excel == 121)
{
ofstream fout2(afile2, ios::out | ios::app);
if (!fout2.good())
// or if (!fin)
{
cerr << "Can't open " << afile2 << " file for output.\n";
exit(1);
}
fout2 << "\n";
//fout2 << (iternum);
//fout2 << "\t";
//fout2 << (Maxfemales);
//fout2 << "\t";
//fout2 << (Totalremate);
//fout2 << "\t";
//fout2 << (Pluggedfemales);
//fout2 << "\t";
fout2 << (tally);
fout2.close();
}
311
//FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
}
}
312
Appendix F. Male frequency collection, 1998 Brood XIX.
Series
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98
DJ98
DJ98V
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98V
DJ98
DJ98
DJ98V
DJ98
DJ98
DJ98
DJ98
DJ98
Tape Side
Counter Cicada (FC) Main
Frequency
kHz
Abdomen
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1B
B
1
2A
2A
2A
2A
B
1B
B
B
2A
2A
B
B
B
B
B
B
1B
2A
B
B
2A
1
0:13:02
1:33:52
0:01:22
0:10:46
0:48:02
0:56:54
1:03:42
0:55:02
0:29:52
1:19:52
0:36:23
1:05:58
0:58:11
0:17:37
0:22:57
0:06:00
96
254
0:39:24
199
89
183
275
111
41
211
243
288
281
222
157
197
231
180
170
62
217
140
146
243
0:31:56
B
126
0:25:50
295
185
200
174
267
4
1
2
2
2
4
3
2
3
4
2
2
4
2
2
2
2
2
2
2
2
2
2
4
2
2
2
2
1
2
4
3
1
2
2
2
2
2
3
2
2
2
2
1
1
2
2
2
2
1
2
2
2
2
2
A
B
B
B
A
313
1
2
3
4
5
6
8
9
10
11
13
14
15
17
18
19
20
21
22
23
24
25
27
28
30
31
32
32
33
34
37
38
39
40
42
43
46
49
51
52
53
54
55
59
60
62
63
64
65
1.09
1.55
1.76
1.78
1.70
1.04
1.07
1.70
1.70
1.19
1.70
1.61
1.73
1.71
1.76
1.64
1.75
1.70
1.77
1.69
1.75
1.64
1.66
1.70
1.68
1.69
1.69
1.74
1.89
1.78
1.10
1.89
1.72
1.72
1.72
1.64
1.70
1.70
1.88
1.68
1.73
1.73
1.65
1.71
1.78
1.70
1.69
1.65
1.55
DJ98
DJ98
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98V
DJ98
DJ98
DJ98V
DJ98V
DJ98
DJ98V
DJ98
DJ98
DJ98V
DJ98
DJ98
DJ98V
DJ98V
DJ98
DJ98
DJ98
DJ98
DJ98
2
2
1
1
1
1
1
1
1
1
1
1
3
2
2
2
2
2
B
B
2
2
2
2
2
2
2
2
2
2
2
1
2
2
1
1
1
1
2
2
1
1
2
1
2
2
1
2
2
1
1
2
2
2
2
2
A
B
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
B
B
B
B
72
108
1:17:34
1:10:40
0:49:20
1:16:20
1:00:37
0:52:45
1:12:35
0:46:55
1:35:22
1:56:50
195
125
209
44
77
14
110
226
65
249
54
190
236
229
100
22
34
1:47:28
0:32:52
0:37:51
1:59:44
1:21:39
1:39:40
1:31:09
66
208
1:48:33
1:38:39
168
1:57:48
158
232
1:55:32
115
84
1:28:47
1:44:12
222
100
196
191
151
314
66
67
68
69
70
71
72
73
74
75
76
78
81
83
84
85
86
87
88
89
90
92
93
94
95
96
97
98
99
100
104
108
111
112
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
1.73
1.75
1.71
1.10
1.16
1.11
1.11
1.69
1.81
1.66
1.15
1.69
2
2
2
4
3
4
3
2
2
2
4
2
1.72
1.76
1.08
1.71
1.60
2
2
4
2
2
4
2
2
2
2
3
2
2
2
2
2
2
2
3
3
2
4
4
2
2
1
2
2
2
2
1
2
2
2
2
2
2
3
2
2
2
2
1.70
1.65
1.81
1.68
1.43
1.79
1.58
1.69
1.72
1.71
1.70
1.62
1.63
1.00
1.50
1.06
1.04
1.71
1.72
1.75
1.73
1.69
1.76
1.72
1.70
1.72
1.70
1.13
1.77
1.78
1.61
1.67
1.71
1.73
1.76
1.65
DJ98V
DJ98V
DJ98V
DJ98V
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98V
DJ98V
DJ98
DJ98
DJ98
DJ98
DJ98V
DJ98V
DJ98V
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98V
DJ98V
DJ98
DJ98V
DJ98V
DJ98V
DJ98V
DJ98
DJ98
DJ98
DJ98V
DJ98
DJ98
DJ98
DJ98V
DJ98
DJ98V
DJ98V
DJ98
DJ98V
DJ98V
DJ98
DJ98
DJ98V
DJ98
DJ98
1
1
1
1
2
2
3
3
3
3
3
2
2
3
3
3
3
2
2
3
3
3
3
3
2
2
3
2
2
2
2
3
3
3
2
3
3
3
2
3
2
2
3
2
2
3
3
2
3
3
B
B
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
1:52:09
1:49:51
1:54:14
2:00:27
90
125
197
364
235
275
265
1:15:35
0:51.22
270
347
285
241
1:12:30
0:48:02
206
371
281
360
259
1:05:19
1:07:00
220
0:43:37
1:16:02
0:41:27
1:14:16
321
253
343
1:02:27
305
375
212
0:45:37
340
0:49:26
1:17:38
231
0:40:12
0:51:55
247
328
1:10:12
332
356
315
135
136
137
138
139
140
200
201
202
203
204
205
206
207
209
210
212
213
214
215
217
218
219
220
222
223
225
227
228
229
230
231
232
233
234
235
236
237
238
239
240
240
241
242
243
244
245
246
247
248
250
1.08
1.67
1.11
1.70
1.10
1.83
1.66
1.69
1.77
1.69
1.65
1.72
1.62
1.61
1.03
1.80
1.63
1.69
1.77
1.63
1.61
1.64
1.60
1.66
1.64
1.74
1.10
1.72
1.70
1.14
1.71
1.78
1.70
1.69
1.84
1.70
1.83
1.13
1.06
1.10
1.12
1.69
1.75
1.79
1.15
1.72
1.63
1.64
1.84
4
2
3
1
4
2
3
2
2
2
2
2
2
2
4
1
2
2
2
2
2
2
2
2
2
2
1
2
4
2
3
4
2
2
2
2
3
2
1
4
4
4
3
2
2
2
4
2
2
2
2
Appendix G. Playback Experiments to females, 1998 Brood XIX.
Teneral Date
Cicada 1
05/13/98
n 1
05/15/98
m 4
05/15/98
o 0
05/15/98
p 0
05/16/98
1 0
05/16/98
2 0
05/16/98
3 4
05/16/98
4 0
05/16/98
5 0
05/16/98
6 0
05/16/98
7 1
05/16/98
8 0
05/16/98
9 0
05/16/98
10 0
05/16/98
11 0
05/16/98
12 0
05/16/98
13 0
05/16/98
14 1
05/16/98
15 0
05/16/98
16 3
05/18/98
3a 0
05/18/98
5a 0
05/18/98
8a 4
05/18/98
a 4
05/18/98
b 0
05/18/98
c 0
05/18/98
d 0
05/18/98
e 5
05/18/98
f 1
05/18/98
g 8
05/18/98
m 1
05/18/98
o 0
05/18/98
p 0
05/19/98
0 7
05/19/98
1 1
05/19/98
2 0
05/19/98
3 0
05/19/98
4 0
05/19/98
5 2
05/19/98
6 0
05/19/98
7 0
05/19/98
9 0
05/19/98
10 2
05/19/98
11 5
05/19/98
13 0
05/19/98
14 0
05/19/98
15 1
05/19/98
16 0
05/19/98
17 0
05/19/98
18 0
05/19/98
21 0
05/19/98
23 0
05/19/98
24 0
05/19/98
4 0
05/19/98
4 0
05/19/98
4 0
05/21/98
12 0
05/21/98
3 0
05/23/98
4 0
05/23/98
8 0
05/21/98
3 0
05/23/98
4 0
05/23/98
4 0
05/23/98
8 0
05/23/98
4 0
05/31/98
A 2
05/31/98
C 0
05/31/98
DD 1
05/31/98
F 0
05/31/98
FF 0
05/31/98
JJ 0
05/31/98
M 2
05/31/98
MM 1
05/31/98
NN 0
05/31/98
RR 0
05/31/98
W 0
05/31/98
X 0
1.1
2
3
0
0
0
0
4
0
0
0
4
0
0
0
0
0
1
0
0
4
0
0
4
6
0
1
1
6
0
8
2
0
0
8
1
0
0
0
2
0
0
0
3
6
1
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
3
1
1
0
1
0
1.2
0
4
0
0
0
0
5
1
0
0
5
0
0
0
0
0
1
0
0
4
0
0
3
4
0
1
4
4
0
6
4
0
1
7
1
0
0
0
1
0
0
0
5
3
0
0
2
0
0
0
1
0
0
0
0
0
0
1
1
0
0
0
0
0
0
4
1
2
1
2
0
3
1
0
0
2
1
1.3
0
4
0
2
0
0
3
1
0
1
7
3
0
0
0
1
1
0
0
4
0
0
3
2
0
1
6
3
0
7
4
0
3
7
2
1
0
1
0
0
2
0
4
3
0
0
5
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
2
0
2
1
2
1
4
1
2
0
1
0
1.4
0
4
0
4
0
0
2
1
1
2
8
4
0
1
1
3
1
0
0
2
0
0
1
0
0
3
8
3
0
7
3
0
3
5
2
1
0
1
0
2
0
0
1
0
0
0
7
0
1
0
4
0
1
0
0
0
1
0
0
0
1
0
0
0
0
2
0
2
0
1
1
4
2
0
1
2
0
1.5
0
1
0
2
0
2
1
1
1
2
8
3
2
2
0
3
1
1
1
1
1
0
0
0
1
5
7
0
0
5
0
0
4
2
4
1
1
0
0
3
4
1
0
0
1
0
7
1
0
0
4
0
2
0
1
0
0
0
0
0
0
0
0
0
0
1
1
1
0
0
0
3
1
0
0
0
0
1.6
0
1
0
3
0
1
0
1
0
2
7
7
1
3
0
4
0
3
0
0
2
0
0
0
0
6
8
0
0
2
0
0
4
0
4
1
3
1
0
4
5
3
0
0
0
3
6
1
0
0
2
3
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
1
0
0
0
0
1.7
0
0
0
3
1
3
0
2
1
2
5
3
3
3
0
6
1
3
0
0
1
0
0
0
1
6
8
0
0
2
0
0
4
0
6
2
4
1
0
3
8
2
0
0
3
3
7
1
1
1
1
4
1
1
1
1
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
3
2
0
0
0
0
316
1.8
0
0
0
4
2
3
0
0
0
4
5
6
7
2
0
6
0
3
0
0
2
1
0
0
1
5
7
0
0
1
0
0
4
0
6
0
5
2
0
2
8
3
0
0
3
5
7
4
1
2
0
6
2
1
1
1
0
0
0
1
0
0
0
1
1
0
0
0
0
0
0
3
1
0
0
0
0
1.9
0
0
0
4
1
3
0
1
0
2
4
4
7
1
0
5
1
3
0
0
1
1
0
0
0
4
6
0
0
2
0
0
4
0
6
0
4
3
0
0
7
2
0
0
3
3
6
5
1
1
0
6
2
0
1
0
0
0
0
1
0
1
1
0
0
0
0
0
0
0
0
3
1
0
0
0
0
2
0
0
1
2
0
2
0
0
1
1
2
4
4
1
0
4
1
2
0
0
1
2
0
0
0
5
6
0
0
1
0
0
4
0
4
0
3
2
0
0
6
2
0
0
2
0
4
2
0
1
0
4
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
2.1
0
0
0
2
0
1
0
0
0
0
1
2
4
0
0
4
1
0
0
0
1
3
0
0
0
3
5
0
0
1
0
1
4
0
0
0
0
4
0
1
5
0
0
0
3
1
1
4
0
0
0
2
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2.2
0
0
0
1
0
0
0
0
0
1
1
0
0
0
0
4
1
0
0
0
0
3
0
0
0
2
4
0
0
0
0
0
2
0
0
0
0
0
0
1
3
0
0
0
2
0
2
1
0
0
0
2
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2.3 Weighted Average
0
1.067
0
1.238
0
2.000
1
1.736
0
1.800
0
1.793
0
1.189
0
1.538
0
1.650
0
1.700
0
1.522
0
1.692
0
1.857
0
1.677
0
1.400
3
1.837
0
1.640
0
1.719
0
1.500
0
1.206
0
1.778
1
2.082
0
1.153
0
1.125
0
1.667
1
1.737
0
1.673
0
1.167
0
1.000
0
1.334
0
1.243
0
2.100
0
1.732
0
1.203
2
1.695
0
1.533
0
1.785
0
1.847
0
1.080
0
1.669
4
1.850
0
1.762
0
1.193
0
1.124
2
1.905
0
1.780
2
1.644
0
1.900
0
1.700
1
1.917
0
1.462
0
1.867
0
1.670
0
1.750
0
1.725
0
1.750
0
1.400
0
1.250
0
1.650
0
1.967
0
1.400
0
1.950
0
1.950
0
1.750
0
1.750
0
1.245
0
1.350
0
1.267
0
1.250
0
1.250
0
1.350
0
1.475
0
1.467
0
1.233
0
1.400
0
1.267
0
1.200
Min Max Abdomen
1 1.1
3
1 1.1
2.0 1.6
1.3 2.3
1.7 1.9
1
1.5 2.1
3
1.0 1.5
3
1.2 1.9
1
1.4 2.0
3
1.3 2.2
2
1.0 2.2
2
1.3 2.1
4
1.5 2.1
3
1.4 2.0
2
1.4 1.4
3
1.3 2.3
2
1.1 2.2
2
1.0 2.0
3
1.5 1.5
1
1.0 1.5
4
1.5 2.1
1
1.8 2.3
3
1.0 1.4
4
1.0 1.3
2
1.5 1.8
3
1.1 2.3
2
1.1 2.2
2
1.0 1.4
2
1.0 1.0
1.0 2.1
3
1.0 1.4
4
2.1 2.1
2
1.2 2.2
2
1.0 1.5
4
1.0 2.3
2
1.3 1.7
3
1.5 2.0
2
1.3 2.1
1
1.3 1.2
1.4 2.2
2
1.3 2.3
2
1.5 2.0
3
1.0 1.4
4
1.0 1.3
4
1.1 2.3
2
1.6 2.1
2
1.0 2.3
3
3
1.4 1.9
2
1.7 2.3
2
1.2 1.7
2
1.6 2.2
2
1.4 1.9
2
1.0
1.2
1.0
1.2
1.1
1.3
1.0
1.0
1.1
1.4
1.1
1.2
1.5
1.5
1.5
1.3
1.4
1.4
2.0
1.9
1.3
1.4
1.4
1.2
317
1
3
3
3
3
2
2
2
2
2
2
2
4
4
4
4
1
1
1
1
1
1
1
1
1
4
4
4
4
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Tape
1
1
1
1
1
1
1
1
Side Counter Location
B
447 S. of Montgomery, Rt 24 1.5 m E. of US31 at river
B
S. of Montgomery, Rt 24 1.8 mi E of Montgomery Co border at Federal Rd
B
354 #6 at 19.5m N. of Double Springs on Rt 33, Black Warrior WMA at Lawrence Co Rt 90 jnc.
B
354 #5 at 19.5m N. of Double Springs on Rt 33, Black Warrior WMA at Lawrence Co Rt 90 jnc.
B
354 #4 at 19.5m N. of Double Springs on Rt 33, Black Warrior WMA at Lawrence Co Rt 90 jnc.
B
354 #3 at 19.5m N. of Double Springs on Rt 33, Black Warrior WMA at Lawrence Co Rt 90 jnc.
B
354 #2 at 19.5m N. of Double Springs on Rt 33, Black Warrior WMA at Lawrence Co Rt 90 jnc.
B
354 #1 at 19.5m N. of Double Springs on Rt 33, Black Warrior WMA at Lawrence Co Rt 90 jnc.
S. of Marvin on Rt. 51 2.6 m South of US80
S. of Marvin on Rt. 51 3.8 m S of US80
B
336 Hardy-Weaver
A
8 Black Rock
A
45 N of Hardy on 63
A
70 Mammoth Springs
A
133 1/3 way south from JNC 342 x 63 to Hardy on 63
B
327 Strawberry
B
345 58 x 115
B
359 115 x Strawberry
B
378 Shirey
B
397 310 x 309
B
463 E. of Lawrence Co. 303
B
487 Lake Charles B
A
135 27 x 14
A
290 27 x 65
A
338 235 x 418
B
451 14 X 281
B
343 #6 at Daisy State Park
B
342 #5 at Daisy State Park
B
316 #4 at Daisy State Park
B
240 #3 at Daisy State Park
B
240 #2 at Daisy State Park
B
240 #1 at Daisy State Park
B
200 Ca. 8m W. of Arkadelphia on Rt. 8
A
28 Harold Alexander WMA Taillight
B
352 Harold Alexander WMA, Power Line
A
400 Prairie De Rocher
A
465 900N
B
64 Pere Marquette
B
214 Sugar Creek
B
225 Allerton
A
38 Sam Parr SP nr. Newton
A
18 Fox Ridge SP, SE of Matoon
A
451 Dixon Spr. SP on Rt 146 #1
A
451 Dixon Spr SP on Rt 146 #2
A
451 Dixon Spr SP on Rt 146 #3
A
423 Golconda Marina #1
A
384 Ohio River Rec Area (Golconda) #2
A
384 Ohio River Rec Area (Golconda) #1
A
355 Rt. 5 6.8 mi. S. of Rt. 145, NW of Golconda Sample 1
A
355 Rt. 5 6.8 mi. S. of Rt. 145, NW of Golconda Sample 2
A
310 Rt. 145 2.5 mi. S. of Forest Rd. Rt 402 ("Rt 145 Collection Site"), 2.5 mi. S. of Delwood
A
275 Rt. 402 0.9m W of Rt. 145 (Delwood), "Site 3"
A
250 Rt. 402 0.5m W of Rt. 145 (Delwood), "Site 2" Sample 1
A
250 Rt. 402 0.5m W of Rt. 145 (Delwood), "Site 2" Sample 2
A
Rt. 402 0.4m W of Rt. 145 (Delwood), "Site 1" Sample 1
Appendix H. Locality and chorus pitch data, 1998 Brood XIX.
Series
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DJ98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
DD98
State
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
AR
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
Sharp
Lawrence
Fulton
Fulton
Fulton
Sharp
Sharp
Lawrence
Lawrence
Lawrence
Lawrence
Lawrence
Searcy
Searcy
Searcy
Boone
Pike
Pike
Pike
Pike
Pike
Pike
Clark
Sharp
Sharp
Randolph
Monroe
Jersey
Sangamon
Piatt
Jasper
Coles
Pope
Pope
Pope
Pope
Pope
Pope
Pope
Pope
Pope
Pope
Pope
Pope
Pope
County
Montgomery
Montgomery
Lawrence
Lawrence
Lawrence
Lawrence
Lawrence
Lawrence
40.008
39.016
39.403
37.386
37.386
37.386
37.386
37.377
37.377
37.424
37.424
37.546
37.58
37.58
37.58
37.58
Lat
32.176
32.177
34.393
34.393
34.393
34.393
34.393
34.393
32.404
32.38
36.315
36.122
36.337
36.474
36.442
36.111
35.999
36.028
36.022
36.019
36.061
36.081
35.923
35.95
36.106
36.449
34.233
34.233
34.233
34.233
34.233
34.233
34.19
36.208
36.256
38.088
38.156
38.972
Lon
Date
Hi kHz
Hi XJ/Hz X Lo kHz Lo XJ/Hz
X
86.345
06/03/98
1.12
86.386
06/03/98
1.09
1.36 u
87.315
06/03/98
1.12
6.32 u
87.315
06/03/98
1.21
4.42 u
87.315
06/03/98
1.15
2.7 u
87.315
06/03/98
1.11
15.1 u
87.315
06/03/98
1.14
6.1 u
87.315
06/03/98
1.11
2.05 u
85.363
1.09
1.39 u
85.379
1.08
896 n
91.477
05/19/98
1.69
1.93 n
91.151
05/22/98
1.15
91.3 n
91.492
05/22/98
1.74
617 n
1.13
117 n
91.52
05/22/98
1.65
4740 n
1.05
457 n
91.493
05/22/98
1.74
59.6 n
1.11
146 n
91.452
05/22/98
1.76
99.8 n
1.17
1190 n
91.487
05/22/98
1.76
14.9 n
1.15
572 n
91.329
05/22/98
1.69
0.041 n
1.13
6.68 n
91.235
05/22/98
1.68
4.88 n
1.15
118 n
91.192
05/22/98
1.15
987 n
91.145
05/22/98
1.8
18.1 n
1.17
159 n
91.127
05/22/98
1.74
6.66 n
1.16
776 n
92.617
05/28/98
1.11
67.6 n
92.727
05/28/98
1.14
34.7 n
92.837
05/28/98
1.64
5.05 n
93.074
06/02/98
1.64
201 p
93.742
06/02/98
1.17
1.52 u
93.742
06/02/98
1.15
910 n
93.742
06/02/98
1.12
894 n
93.742
06/02/98
1.12
1160 n
93.742
06/02/98
1.14
981 n
93.742
06/02/98
1.11
1.15 u
93.222
06/02/98
1.16
766 n
91.447 5/12-24/1998
1.73
32
1.14
133
91.462 5/19-25/1998
1.702167325
442.85
1.12
114.5625
90.09
05/29/98
1.7
16.7 n
1.1
624 n
90.052
05/29/98
1.63
100 n
1.07
500 n
90.543
05/29/98
1.48
27.6 n
05/30/98
1.41
2.236 a
88.644
05/30/98
1.4
346 n
88.125
05/31/98
1.41
80 n
88.143
05/31/98
1.37
1.3 u
88.671
06/01/98
1.83
0.515 u
1.12
11.8 u
88.671
06/01/98
1.12
3.89 u
88.671
06/01/98
1.64
0.2 u
1.09
4.9 u
88.48
06/01/98
1.67
0.36 u
1.1
9.22 u
88.485
06/01/98
1.09
4.19 u
88.485
06/01/98
1.09
4.19 u
88.514
06/01/98
1.69
0.386 u
1.1
4.2 u
88.514
06/01/98
1.65
2 u
1.12
15.1 u
88.582
06/01/98
1.76
0.903 u
1.12
3.55 u
88.588
06/01/98
1.78
0.972 u
1.1
8.3 u
88.581
06/01/98
1.78
1.41 u
1.12
4.27 u
88.581
06/01/98
1.76
1.5 u
1.12
6.24 u
88.578
06/01/98
1.78
3.15 u
1.12
6.98 u
318
DD98
DD98
DJ98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
DD98
DD98
DD98
DD98
DD98
ZV98
ZV98
ZV98
ZV98
DJ98
DJ98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
JRC98
DD98
DD98
DD98
DD98
4
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
1
1
1
1
1
B
B
A
A
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
B
B
B
B
A
A
A
A
A
A
A
B
B
B
B
B
B
B
A
A
A
B
B
B
B
B
1 A
1 A
Rt. 402 0.4m W of Rt. 145 (Delwood), "Site 1" Sample 2
163 0.1m off Rt. 145 on Battle Ford Rd. (near Mitchellsville) Sample 3
Nole Alexander Farm
238 IL 37 X I57
275 Forbes 1
329 IL 161 X I 57
420 Muddy Creek
451 Rayse Creek
8 1425 X 400
52 500E
159 2000N
183 Lake Jaycee B
207 Sam Dale A
325 Rend Lake
464 Lonely Oak
42 100E X Berryville
105 Middle Fork
178 1100N x 2300E
184 Lake Barkley Collection Site Sample #1
184 Lake Barkley Collection Site Sample #2
50 Lake Barkley near powerlines
50 Lake Barkley near powerlines
20 Rt. 62 2m. E. of Princeton
Jct Rt 235/Diamond Lane (S. of Dameron)
Bishop House end of Demko Rd
End of Three-second lane, N. of Dameron
End of Kessler Way
373 I44 X US 50
406 I44 ex 2301 Rol Hil
4 Rte 112 MTNF
54 Rte. 112 pullout
72 Roaring River
82 FR 1200
102 Hwy 173
111 Eag Cr 1 M N of Lake
164 AA914 x Grace Rd.
199 Bennet Nature
452 14 x AP
475 14 x Co 5130
8 1-2960 x WW
29 1.1- 2960 x WW
60 60-571
248 Possum Creek
285 Hilltop
311 1.0C X 65
333 1.5C X 65
350 0.5 140 X 399
365 0.7 140 X 399
383 1.5 dn 391
399 Mt. Olivet
424 Fristo X Wasson
62 439 X US24
159 Karma 1
0.8 mi S from Rt 49 on Tuckertown Rd, N. of Tuckertown
Vista Pt SRA E. of Pittsboro
Rt. 87 N. of Pittsboro, 1.0 m S of SR1551
Nashville, Percy Warner SP
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
KY
KY
KY
KY
KY
MD
MD
MD
MD
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
NC
NC
NC
TN
Pope
Saline
Piatt
Effingham
Marion
Marion
Jefferson
Jefferson
Jefferson
Jefferson
Jefferson
Jefferson
Wayne
Jefferson
Washington
Lawrence
Champaign
Livingston
Trigg
Trigg
Trigg
Trigg
Caldwell
St. Mary's
St. Mary's
St. Mary's
St. Mary's
St. Louis
Franklin
Barry
Barry
Barry
Barry
Stone
Stone
La Clede
La Clede
Howell
Howell
Howell
Howell
Shannon
Wayne
Wayne
Dallas
Dallas
Hickory
Hickory
Benton
Benton
Pettis
Chariton
Randolph
38.536
38.279
36.543
36.574
36.586
36.602
36.73
36.735
37.7
37.726
36.811
36.8
36.961
36.962
36.999
36.968
37.042
37.683
37.682
37.962
37.962
38.17
38.451
38.786
39.426
39.608
35.506
35.703
35.818
36.069
37.58
37.633
40.071
38.993
38.73
38.516
38.31
38.328
38.332
38.402
38.417
38.37
38.55
38.216
38.281
38.6
40.385
40.787
36.849
36.849
36.849
36.849
37.107
38.145
38.169
38.16
88.578
88.532
88.515
88.59
88.777
88.959
88.988
89.107
89.071
89.052
88.9
88.887
88.586
88.986
89.357
87.891
87.959
88.489
87.914
87.914
87.914
87.914
87.941
76.365
76.371
76.369
0
90.503
91.124
93.859
93.836
93.837
93.802
93.511
93.489
92.709
92.856
92.054
91.991
91.714
91.709
91.31
90.297
90.296
93.124
93.126
93.197
93.197
93.138
93.275
93.2
92.754
92.455
80.181
79.05
79.254
86.886
06/01/98
06/01/98
06/09/98
06/10/98
06/10/98
06/10/98
06/11/98
06/11/98
06/11/98
06/11/98
06/11/98
06/11/98
06/12/98
06/12/98
06/12/98
06/13/98
06/14/98
06/14/98
06/01/98
06/01/98
06/01/98
06/01/98
06/01/98
05/25/98
05/27/98
05/28/98
06/01/98
06/01/98
06/01/98
06/03/98
06/03/98
06/03/98
06/03/98
06/03/98
06/03/98
06/03/98
06/03/98
06/06/98
06/06/98
06/06/98
06/06/98
06/06/98
06/06/98
06/06/98
06/07/98
06/07/98
06/07/98
06/07/98
06/07/98
06/07/98
06/07/98
06/07/98
06/07/98
15.6
736
194
1.43
463
40.05
87.15
7.45
48.5
4.59
98.6
69
64.2
49.3
140
351
3.41
13.8
2.3
12.6
7.56
7.37
272
151
110
1.48
1.46
1.67
1.59
1.59
1.59
1.59
1.55
1.48
1.48
1.6
1.48
1.62
1.6
1.67
1.77
1.48
1.49
1.49
1.5
1.5
1.46
1.45
1.48
1.5
4.16 n
286 p
1.37
1.36
p
n
n
n
n
n
n
a
p
p
n
p
p
n
p
a
a
n
n
n
a
p
p
p
a
n
n
n
p
n
p
n
a
17.1
2.25
14.6
833
16
1
130
1.52
1.65
1.66
1.66
1.62
1.65
1.65
1.11
1.72
u
u
u
n
a
a
2.35
0.25
100
6.58
41.3
992
1.78
1.77
1.37
1.46
1.58
1.57
2.48 n
1.13
1.08
1.15
1.1
1.06
1.19
5.05 u
2.14 u
935
687 p
605 n
50 n
1 u
16.6 n
939 p
1.09
1.1
1.1
1.09
1.12
1.15
1.08
1.05
1.1
1.12
30.7
17.5
12.2
9.9
1
n
n
n
n
p
3.97 u
10 u
1.09
1.13
1.1
1.08
1.09
1.09
1.11
APPENDIX I
Holding jig for symmetry measurements.
An X-acto “X-tra Hands” tool (Hunt Mfg. Co., Speedball Road, Statesville, NC 28677)
was modified in the following manner to make a holding jig.
1. Alligator clips were removed and discarded.
2. Copper alligator clips were filed so that the gripping surfaces were flat when gripping
two microscope slides and two sheets of brass.
3. Small rectangles of brass were soldered into the filed jaws of the copper alligator
clips.
4. The brass rectangles were glued to two pieces of microscope glass using “liquid nails”
adhesive.
5. All exposed areas of adhesive were coated with red GLPT insulating varnish. This
step is critical since the insulating varnish forms a barrier that protects the adhesive from
exposure to alcohol or other preserving fluids.
6. The clip/slide assembly was installed in the X-tra hands device.
319
320
F14
Cicada Symmetry
1 0.00991736
2 0.01324503
3
0
4 -0.02040816
5 -0.0033389
6 0.01333333
7
0
8
0
16 -0.00317965
17 0.00305344
18 -0.01769912
19 -0.0356564
20 0.01550388
21 -0.02843602
22 0.05405405
23 -0.02346041
24 0.02337229
25 -0.0030349
26 -0.01298701
27
0
28 -0.00634921
29 -0.04334365
30 0.01948052
31 0.01360544
39
0
40 0.01215805
49 0.00916031
50 -0.0033389
51 -0.02033898
52 0.01269841
53
0
54 -0.00933126
55
-0.0125
56 -0.00350263
57 0.00630915
58 -0.02643172
59 0.00649351
60 0.01960784
61 -0.02368866
62 0.02150538
63
0
64 0.0030349
65 0.00351494
66 -0.00321027
67 0.00343053
68 0.0060241
69 0.00628931
70 -0.02492212
71 0.02298851
72 -0.00331675
9 -0.00327332
10 -0.01584786
F 14
Size
11.4233842
11.4045025
11.2912128
11.1023965
11.3100944
11.328976
11.8198983
13.5947712
11.8765432
12.3674655
12.8017429
11.6499637
12.1786492
11.9520697
12.5751634
12.8772694
11.3100944
12.442992
11.6310821
12.4618736
11.8954248
12.1975309
11.6310821
11.1023965
11.328976
12.4241104
12.3674655
11.3100944
11.1401598
11.8954248
11.8954248
12.140886
12.0842411
10.7814089
11.9709513
12.8583878
11.6310821
11.5555556
11.1590414
12.291939
11.328976
12.442992
10.7436456
11.7632534
12.3674655
12.8017429
11.6499637
12.1220044
11.4989107
11.3856209
11.5366739
11.9143065
F 15
F 15
Symmetry Size
-0.0096 11.8010167
0
0.00310078 12.1786492
-0.02764977 12.291939
0
0.00304414 12.4052288
-0.00930233 12.1786492
0.00283688 13.3115468
-0.01791045 12.6506899
0.00860832 13.1604938
0.00291971 12.9339143
0 12.4241104
-0.00902256 12.5562818
0.02067947 12.7828613
0
-0.03156385 13.1604938
-0.01246106 12.1220044
0 12.4618736
-0.00593472 12.7262164
0.01503759 12.5562818
-0.00291121 12.9716776
0.03003003 12.5751634
-0.01230769 12.2730574
0.00638978 11.8198983
-0.00913242 12.4052288
-0.00615385 12.2730574
0
-0.00607903 12.4241104
0
-0.01183432 12.7639797
-0.02143951 12.3297023
-1.4E-16 12.6884532
-0.01176471 12.8395062
-0.0096 11.8010167
-0.00308166 12.2541757
-0.02043796 12.9339143
0.00584795 12.9150327
-0.00944882 11.989833
0.00316957 11.9143065
0.02756508 12.3297023
-0.00581395 12.9905592
0.02002861 13.1982571
0.00328407 11.4989107
0.00609756 12.3863471
-0.00617284 13.1604938
0.00865801 12.9339143
0.0140647 12.4241104
-0.01173021 12.8772694
0.00621118 12.1597676
-0.00309119 12.2164125
0.01490313 12.6695715
0
F 27
Symmetry
0.00291121
0.00578035
-0.0173913
-0.0149925
0
-0.01801802
0.01146132
0.0137931
-0.0115942
0.01142857
0.01671309
-0.00302572
0
-0.0057971
0.00581395
-0.04255319
0.00569801
0.00302572
0
0.01694915
-0.00554017
-0.02253521
0.02105263
-0.00623053
0
0.00297177
0.00272851
0
-0.01526718
-0.00297177
-0.01204819
0.00561798
-0.01430615
-0.00323102
0
0.01426534
-0.0027894
0.00578035
0.00283688
0.02008608
-0.00285307
0.00266312
0.00625
0.00301659
0
0.00557103
-0.01117318
-0.01149425
-0.00297177
-0.00607903
0.00873362
0.00289436
F 27
Size
12.9716776
13.0660857
13.0283224
12.594045
12.4618736
12.5751634
13.1793755
13.6891794
13.0283224
13.2171387
13.557008
12.4807553
13.0283224
13.0283224
12.9905592
13.3115468
13.254902
12.4807553
13.254902
13.3681917
13.6325345
13.405955
12.5562818
12.1220044
12.4996369
12.7073348
13.8402324
13.0283224
12.3674655
12.7073348
12.5374001
13.4437182
13.1982571
11.6877269
13.5947712
13.2360203
13.5381264
13.0660857
13.3115468
13.1604938
13.2360203
14.1801017
12.0842411
12.5185185
13.2171387
13.557008
12.4807553
13.1416122
12.7073348
12.4241104
12.9716776
13.0472041
-0.00638978
0.00680272
-0.01615509
0
-0.00626959
-0.01769912
-0.02547771
-0.0034904
-0.01846154
0.00623053
0
0
0
-0.00953895
0.00956938
0.01886792
0.022187
0.00329489
0
0
-0.01526718
-0.03231598
0.02782071
0.00636943
0.01282051
0.00956938
0.01960784
0.00636943
0.0294599
3.0446E-16
-0.03647416
0.01875
0.00327332
0.01727116
0
0.00953895
H 40
Symmetry
0.0221169
0
-0.02955665
-0.01011804
0.01658375
0.01360544
-0.0033389
0
0
0.01536098
0.03389831
-0.01011804
0.00638978
Appendix J. Symmetry and suze measurements for mated and unmated cicadas.
Mate
Status
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
Yes
H 40
Size
11.9520697
11.328976
11.4989107
11.1968046
11.3856209
11.1023965
11.3100944
12.8395062
11.6310821
12.291939
12.2541757
11.1968046
11.8198983
0
11.7821351
11.83878
11.5555556
11.8576616
11.5366739
11.6688453
12.4241104
12.0842411
11.5366739
10.9324619
10.7625272
11.8765432
0
12.0087146
11.9143065
11.4611474
12.2730574
11.1401598
12.3674655
10.5170661
12.2164125
11.8576616
0
11.8198983
11.1023965
11.6877269
11.8954248
12.0464779
10.6681191
11.8576616
12.291939
12.2541757
11.1968046
12.1597676
11.7066086
11.6310821
11.8765432
11.83878
H 46
Size
12.7073348
13.4437182
12.1220044
12.5751634
13.0472041
12.6318083
12.5562818
0
0
0
0
13.0283224
13.1227306
0
0
13.6136529
0
13.5947712
13.0472041
13.2737836
13.8024691
13.5947712
12.5374001
12.6318083
12.7828613
0
0
13.254902
11.9331881
13.254902
13.2360203
0
13.8402324
11.8576616
13.2926652
13.557008
0
13.0660857
12.2541757
13.3304285
13.5758896
13.8213508
11.9143065
13.2926652
0
0
-0.00834492 13.5758896
0
0.02014388 13.1227306
0
-0.00550964 13.708061
0.02890173
0.02157165
0
-0.00278164
0
0.00316957
0
0.00272851
-0.00636943
0
-0.00557103
-2.6803E-16
0.01898734
0.02279202
0.00285307
0
-0.00289436
-0.00284495
-0.01915185
0.01111111
0.01204819
0.01494768
-0.00886263
-0.00277393
0
0.01438849
H 46
Symmetry
0.00297177
0.00561798
0
0.01201201
0.01447178
0.00298954
0.00300752
Fore Femur
Symmetry
0.029752066
-0.013029316
-0.00625
0.00619195
0.006557377
0.006493506
0.016638935
0.00317965
-0.006493506
0
0.012269939
0.019480519
0.009508716
-0.026578073
0.009538951
0.003025719
0.009756098
0.00317965
0.006644518
0.003305785
0
0.01610306
-0.010050251
0.003327787
0.022764228
0
0.009630819
-0.013157895
0.009569378
-0.01610306
0
-0.016051364
0.012195122
-0.006688963
0.012779553
0.012779553
0
0.003327787
-0.01320132
0.02907916
-0.009360374
0.00312989
0.017035775
0.029268293
0.02027027
-0.003149606
0.00921659
0.009884679
0.0224
0.009884679
0.012383901
0.003159558
Fore Femur
Size
4.459616546
4.525957949
4.717610892
4.761838494
4.496472881
4.540700483
4.430131478
4.636526954
4.540700483
4.599670619
4.806066096
4.540700483
4.651269488
4.437502745
4.636526954
4.872407499
4.533329216
4.636526954
4.437502745
4.459616546
4.614413153
4.577556818
4.40064641
4.430131478
4.533329216
4.422760211
4.592299352
4.481730347
4.62178442
4.577556818
4.422760211
4.592299352
4.835551164
4.408017677
4.614413153
4.614413153
4.673383289
4.430131478
4.466987813
4.562814284
4.724982159
4.710239625
4.32693374
4.533329216
4.599670619
4.806066096
4.540700483
4.47435908
4.607041886
4.47435908
4.761838494
4.666012022
0.014184397
-0.011385199
-0.003710575
-0.010544815
0
0
-0.003629764
0.003514938
-0.010657194
-0.003565062
0.021505376
-0.020408163
-0.03364486
0.003514938
-0.007092199
0
-0.003577818
-0.007434944
0.010810811
0.003384095
0
0.003898635
0
-0.011173184
0
0.003350084
0.007092199
-0.003514938
-0.021428571
0.013745704
0.017953321
Hind Femur
Symmetry
0.003724395
-0.014234875
0
-0.017761989
0.010810811
-0.003853565
-0.019267823
-0.031088083
0.010810811
0
0.007017544
-0.007326007
-0.007042254
-0.003565062
0
-0.016977929
0.003656307
0
0.003696858
0.010810811
Hind Femur Tymbal Rib
Size
Symmetry
4.91776455
5.146710758
5.256604938
5.155868607
5.08260582
4.75292328
4.75292328
5.30239418
5.08260582
5.494708995
5.219973545
5.000185185
5.201657848
5.13755291
5.091763668
5.393972663
5.009343034
5.219973545
4.954395944
5.08260582
0
5.165026455
4.826186067
4.936080247
5.210815697
5.018500882
5.165026455
5.045974427
5.210815697
5.155868607
5.13755291
5.110079365
5.384814815
4.899448854
5.210815697
5.165026455
5.201657848
5.119237213
4.926922399
5.08260582
5.41228836
5.494708995
4.69797619
5.036816578
5.494708995
5.219973545
5.000185185
5.165026455
5.210815697
5.128395062
5.329867725
5.100921517
Cumulative
Cumulative
(FA Char.)
(All)
0 0.039352066 0.058876793
0
0 0.009350775 0.02674208
0 0.03384172 0.099016388
0
0 0.009537647 0.047732099
0 0.025941261 0.05967792
0 0.00601653
0 0.024403954
1 0.008608321
0 0.015189647
0 0.019480519 0.065488647
0 0.018531273 0.055465891
0 0.047257541
0
0 0.034589564 0.12035502
0 0.022217157 0.054943757
1 0.00317965 0.00924027
0 0.012579236 0.032157463
0 0.018343379 0.048948293
0 0.002911208
0 0.04613309 0.137307462
0 0.022357944 0.086324487
0 0.009717563 0.048211793
0 0.031896648 0.051304092
0 0.006153846
0
0 0.019236922 0.026205584
1
0 0.027937379 0.077056776
0 0.02143951 0.039905832
0 0.016051364
0 0.023959828 0.073902656
1 0.016288963 0.063036895
0 0.015861217 0.025685304
0 0.033217509 0.086577791
0 0.005847953 0.01513086
0 0.012776606 0.070644348
0 0.016370892 0.071903028
0 0.056644244 0.109046515
0 0.015174328 0.024193131
1 0.023158503 0.02885652
0 0.020319847 0.037152993
0 0.035365854 0.041592718
0 0.02644311
0 0.011807615
0 0.023281287 0.044093864
0 0.021614884 0.073468377
0 0.02861118
0 0.012975869 0.063944102
1 0.027287031
0 0.003159558 0.045364737
321
Mate
Status
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Cicada
11
12
13
14
15
32
33
34
35
36
37
38
41
42
43
44
45
46
47
48
73
74
75
76
77
F14
Symmetry
-0.02994012
0.02446483
0.03389831
0
0.00318979
0.00657895
-0.04778157
0.0126183
0.00941915
-0.00317965
0.04636785
0.00322061
-0.00668896
0.00913242
0.01043478
0
0.01540832
-0.01476015
0.01574803
0.00607903
0.00630915
-0.01647446
0.00332779
0.02711864
0.01733102
F 14
Size
12.6129267
12.3485839
11.1401598
11.4045025
11.83878
11.480029
11.0646333
11.9709513
12.0275962
11.8765432
12.2164125
11.7254902
11.2912128
12.4052288
10.8569354
12.6506899
12.2541757
10.2338417
11.989833
12.4241104
11.9709513
11.4611474
11.3478577
11.1401598
10.8946986
-2.9584E-16
0
0.01197605
0.00316957
-0.00281294
0.0089955
-0.00664452
0.00588235
0
0
0.01230769
-0.00606061
0.00641026
-0.02461538
-0.00636943
F 15
Symmetry
-0.02086438
0.01173021
0.01941748
0.02439024
F 15
Size
12.6695715
12.8772694
11.6688453
12.3863471
0
11.7821351
12.2730574
11.8576616
0
12.5374001
12.2730574
12.4618736
0
12.4618736
0
13.4248366
12.594045
11.3667393
12.8395062
0
0
12.0087146
12.9150327
12.6129267
11.9143065
F 27
Symmetry
-0.01101928
-0.00557103
0.00571429
0.00292826
0.00578035
0.0058651
0
0.01574803
0
0.00289436
0.01485884
0.00878477
-0.01569859
0.00846262
0.0143472
0
0.002886
0.00662252
0
-0.00863309
-0.01117318
0
-0.00598802
0.00881057
0.002886
F 27
Size
13.708061
13.557008
13.2171387
12.8961511
13.0660857
12.8772694
12.4618736
11.989833
13.4437182
13.0472041
12.7073348
12.8961511
12.0275962
13.3870733
13.1604938
13.5947712
13.0849673
11.4045025
13.0283224
13.1227306
13.5192447
13.5947712
12.6129267
12.8583878
13.0849673
0.02276423
0
-0.03870968
-0.03355705
-0.01642036
0.00699301
0.00966184
-0.01550388
0.02261712
0.01269841
0.04388715
0.01324503
0.00664452
0.02020202
0.02531646
-0.02702703
0.00638978
0.02329451
0.00981997
0.02870813
0.00315956
0.02143951
0.03389831
H 40
Symmetry
0
0.01680672 13.4814815
0
0
-0.00825309 13.7269426
0.00584795 12.9150327
0.01173021 12.8772694
0.00302572 12.4807553
0.0056338 13.405955
0 13.0283224
0.0058309 12.9527959
0.00557103 13.557008
-0.00611621 12.3485839
0.01958042 13.5003631
0.01204819 12.5374001
0
0
0
0.01117318 13.5192447
0 13.0283224
0.00273598 13.8024691
0
0
0.01204819 12.5374001
0
0
12.1786492
11.6877269
11.8954248
12.0464779
11.4045025
11.3667393
11.2156863
11.9331881
11.177923
11.8198983
11.3478577
11.5366739
11.83878
11.9520697
12.3297023
12.2541757
0
11.6122004
12.3863471
11.7066086
11.2534495
11.4989107
10.8002905
11.7254902
H 46
Size
H 46
Symmetry
H 40
Size
Fore Femur
Symmetry
-0.012820513
-0.003189793
0.003327787
-0.003231018
-0.013245033
0.013157895
0.003327787
0.003395586
0.003210273
0.012944984
0
0.006451613
0.013071895
-0.003159558
0.009917355
0
0.009360374
0
0.003025719
0.022544283
0.003072197
-0.00317965
0
0.006802721
-0.006644518
Fore Femur
Size
4.599670619
4.62178442
4.430131478
4.562814284
4.452245279
4.481730347
4.430131478
4.341676274
4.592299352
4.555443017
4.422760211
4.570185551
4.511215415
4.666012022
4.459616546
4.865036232
4.724982159
3.950999122
4.872407499
4.577556818
4.798694829
4.636526954
4.658640755
4.334305007
4.437502745
0.003539823
0.018416206
0.003490401
0.02166065
0
0.006968641
0.007936508
0.017152659
0.014234875
0.013605442
-0.010544815
0
-0.041745731
0.003642987
Hind Femur
Symmetry
-0.025316456
-0.003577818
0.007092199
0.017761989
-0.011378556
0.003552398
-0.007434944
0
-0.007067138
-0.01814882
Hind Femur Tymbal Rib
Size
Symmetry
5.064290123
5.119237213
5.165026455
5.155868607
5.231420855
5.155868607
4.926922399
4.945238095
5.183342152
5.045974427
0
5.174184303
4.97271164
5.24744709
5.073447972
5.219973545
5.256604938
4.615555556
5.339025573
5.146710758
5.384814815
5.210815697
5.201657848
4.826186067
5.02765873
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0.041846823
0.041412552
0.094889891
0.041157059
0.065340401
Cumulative
(All)
0.00317965
0
0.018778769 0.108501909
0.00981409
0.00281294
0.018355877
0.006644518
0.008908072 0.052981946
0.003159558 0.043825422
0.012944984 0.03716781
0.012307692
0.012512219 0.033628458
Cumulative
(FA Char.)
0.033684894
0.014919998
0.022745263
0.027621262
0.013245033
0.019568151
0.027943172
0.009765012
APPENDIX K
Code for resampling algorithm to evaluate whether males with asymmetrical wing
venation were over- or under-represented among males that did not mate in mating
experiments. (Think Lightspeed Pascal for the Macintosh)
program Extra;
const
TargetMatings = 2;
{1= not mated; 2= mated}
malenumber = 16;
var
Tabarray, NewData, HardData: array[0..77] of integer;
Distribution: array[0..10000] of integer;
NumWTargMatings, iterations, randomnumber, FemaleNumber: integer;
{* Seeds Random Number Generator}
procedure TIMESEED;
begin
GetDateTime(randseed);
end;
procedure Welcome;
begin
writeln('Welcome to Resampler. You can run up to 10,000 resamplings');
writeln(' of abnormal wing veined male mating data. ');
end;
{*Picks a random number, scales it to the appropriate size}
procedure PICKRAND;
begin
randomnumber := (Trunc(((ABS(Random)) / 32768) * 77));
{writeln('Picked number ', randomnumber);}
end;
procedure FillHardSlot;
begin
pickrand;
{writeln(randomnumber);}
if HardData[randomnumber] = 2 then
begin
FillHardSlot;
{writeln('repeat picked!');}
end;
HardData[randomnumber] := 2;
end;
322
procedure HardDataMaker;
var
counter: integer;
begin
for counter := 1 to 77 do
HardData[counter] := 1;
for counter := 1 to 27 do
begin
FillHardSlot;
end;
for counter := 1 to 77 do
{Writeln(HardData[counter]);}
end;
{Gets Input parameters}
procedure INPUTS;
begin
writeln('Please provide the number of iterations');
writeln('of the simulation , maximum 10000 : ');
read(iterations);
end;
{Checks Input parameters}
procedure CHECKVAR;
begin
if (iterations > 10000) or (iterations < 1) then
begin
writeln('Sorry, but iterations must be an integer less than
10000');
writeln('program has been terminated');
HALT;
end;
end;
procedure Resample;
var
counter: integer;
begin
for counter := 1 to malenumber do
begin
pickrand;
if randomnumber = 0 then
pickrand;
NewData[counter] := HardData[randomnumber];
{Writeln(NewData[counter]);}
end;
end;
{clears this array}
procedure ClearDistribution;
var
counter: integer;
begin
for counter := 0 to 10000 do
Distribution[counter] := 0;
323
end;
{creates a variable containing number of females that got TargetMatings matings}
procedure TABULATEmATINGS;
var
counter: integer;
begin
NumWTargMatings := 0;
for counter := 1 to malenumber do
begin
if (NewData[counter] = TargetMatings) then
NumWTargMatings := NumWTargMatings + 1;
end;
end;
procedure iterate;
var
counter: integer;
begin
for counter := 1 to iterations do
begin
resample;
TabulateMatings;
Distribution[counter] := NumWTargMatings;
{Writeln(counter, 'Distribution at counter', distribution[counter]);}
{Writeln(NumWTargMatings, 'NumWTargMatings');}
{Writeln(' ');}
end;
end;
{Sorts distribution array to prepare for generating percentiles}
procedure bubblesort;
var
i, j, k: integer;
begin
for i := iterations downto 2 do
for j := 1 to i - 1 do
if (distribution[j] > distribution[j + 1]) then
begin
k := distribution[j];
distribution[j] := distribution[j + 1];
distribution[j + 1] := k;
end;
end;
procedure MakeDistribution;
var
DistributionTab, count, cumucount, countera, counterb, counterc: integer;
counter: integer;
begin
bubblesort;
{for counterc := 0 to iterations do}
{begin}
{writeln('post-sort');}
{writeln(distribution[counterc]);}
{end;}
cumucount := 0;
{ratchets thru loking for each value; set some kind of a tab ratchet on it each}
324
{time it gets a "hit"}
for DistributionTab := 0 to malenumber do
begin
count := 0;
for counterb := 1 to iterations do
if (Distribution[counterb] = DistributionTab) or
(Distribution[counterb] < DistributionTab) then
count := count + 1;
{cumucount := cumucount + count;}
Writeln('Frequency of trials in which', DistributionTab, ' or fewer');
Writeln('deformed males achieved matings');
{Writeln(cumucount / iterations);}
Writeln(count / iterations);
{Writeln(count);}
Writeln('
');
end;
writeln('
');
{for counter := 1 to iterations do}
{Writeln(distribution[counter]);}
{writeln('
');}
end;
{The shell of the program}
procedure MAKEDECISION;
var
choicer: integer;
begin
writeln('
');
writeln('Would you like to:');
writeln('
(1) Resample ?');
writeln('
(2) End this session?');
writeln(' ');
writeln('Choose (1) or (2) please');
read(choicer);
if choicer = 1 then
begin
Timeseed;
Welcome;
inputs;
checkvar;
ClearDistribution;
HardDataMaker;
iterate;
MakeDistribution;
Makedecision;
end;
if choicer = 2 then
HALT;
if (choicer < 1) or (choicer > 2) then
begin
writeln('Hey there are only two choices here!');
Makedecision;
end;
end;
begin
MakeDecision;
end.
325
APPENDIX L
Miscellaneous field notes on Okanagana
1. The calling behaviors and morphology of O. canadensis and O. rimosa are
superficially similar to those of O. bella found in the western United States. The mating
system of this species may be similar to that of the eastern cicadas.
2. Nymphs and eclosing nymphs were found only during the dense 1998
emergence of O. rimosa. Nymphal skins of O. rimosa and O. bella are similar, with
prominent dark, dorsal abdominal bands. Emerging O. rimosa nymphs are a pale green,
and can be found between approximately 10 AM and 2 PM, not in the evening as with
Tibicen or the evening and morning as with Magicicada. Nymphs darken within an hour
of emergence, and, like newly emerged Tibicen nymphs, are capable of weak flight
within this period.
3. Okanagana oviposit in live twigs. Females excavate shallow, unbifurcated
eggnests.
326
Errata
Sexual behavior in North American cicadas of the genera Magicicada and Okanagana
by
John Richard Cooley
Page 78: First line of results should read: ". . .matings lasted an average of 270.4 minutes. . ."
Table 4.1 (Page 90)
Average for 43 cases should read "270.42 ± 122.07"
Page 103: Line 4 should read: ". . . we made six daily collections. . ."
Page 118: Two lines omitted at end of page.
Last paragraph should read: ". . .and that the alterations in behavior are different for the two fungal life history
stages in ways that are appropriate for the stage-specific modes of transmission. Our observations suggest that the
first stage of the fungus has apparently evolved. . ."
Table 5.9 (Page 137)
P- value for M. cassini "click vs. control, first call bout: should read "P< 0.980."
Table 5.13 (Page 141)
Replicate "D" "no buzz" treatment: 1.93 ± 0.58 females responded to each call.
Figure 5.1 (Page 142)
Caption line 1 should read ". . . male begins court I (CI) calls. . ."
Caption line 4 should read " . . .male begins court III (CIII) calling. . ."
Figure 5.5 (Page 146)
Caption line 4 should read ". . . pairwise comparisons of treatments placed below. . . "
Table line 1 should read "Foreleg vibrate."
Figure 5.6 (Page 147)
Y axis label should read "Number of males"
"Less than or equal to" signs missing from caption
Caption line 7 should read ". . . treatments did not differ in either species."
Figure 5.9 (Page 150)
X axis should read "Background intensity"
Page 247
First line, second paragraph should read: "Female O. canadensis and O. rimosa signal. . . "