Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00075
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 1994
Copyright Date: 1917
Subject: Florida Entomological Society
Entomology -- Periodicals
Insects -- Florida
Insects -- Florida -- Periodicals
Insects -- Periodicals
General Note: Eigenfactor: Florida Entomologist: http://www.bioone.org/doi/full/10.1653/024.092.0401
 Record Information
Bibliographic ID: UF00098813
Volume ID: VID00075
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text

(ISSN 0015-4040)


(An International Journal for the Americas)

Volume 71, No. 4 December, 1988


Announcement 72nd Annual Meeting .......................................................... i

Preface ......................................................... ................................. ii
AGEE, H. R.-How Do Acoustic Inputs to the Central Nervous System of the
Bollworm Moth Control Its Behavior? ........................................... 393
BURK, T.-Acoustic Signals, Arms Races and the Costs of Honest Signaling ... 400
CALKINS, C. 0., AND J. C. WEBB-Temporal and Seasonal Differences in Move-
ment of the Caribbean Fruit Fly Larvae in Grapefruit and the Relationship
to Detection by Acoustics ................................................................. 409
FORREST, T. G.-Using Insect Sounds to Estimate and Monitor TheirPopulations
..... ........... ..... .... .... .......... ........ ..... ........ .................. 416
HAACK, R. A., R. W. BLANK, F. T. FINK, AND W. J. MATrSON-Ultrasonic
Acoustical Emissions from Sapwood of Eastern White Pine, Northern Red
Oak, Red Maple and Paper Birch: Implications for Bark- and Wood-Feeding
Insects ......................................................................................... 427
HAGSTRUM, D. W., J. C. WEBB, AND K. W. VICK-Acoustical Detection and
Estimation of Rhyzopertha dominica Larval Populations in Stored Wheat 441
RYKER, L, C.-Acoustic Studies of Dendroctonus Bark Beetles .................... 447
SIVINSKI, J.-What Do Fruit Fly Songs Mean? ....... ............................... 462
SPANGLER, H. G.-Sound and the Moths That Infest Beehives .................... 467
LITZKOW-A Sound-Insulated Room Suitable for Use With an Acoustic
Insect Detection System and Design Parametersfor a Grain Sample Holding
Container ...................................................................................... 478
WALKER, T. J.-Acoustic Traps for Agriculturally Important Insects ............ 484
WEBB, J. C., D. C. SLAUGHTER, AND C. A. LrrzKOw-Acoustical System to
Detect Larvae in Infested Commodities ............................................... 492

Preface .................................................................................................. 506
ORB, D. B.-Scelionid Wasps as Biological Control Agents: A Review ............ 506
JOHNSON, D. W.-Euchartidae (Hymenoptera: Chalcidoidea): Biology and Poten-
tial for Biological Control ............................ .................................. 528
HOLCK, A. R.-Current Status of the Use of Predators, Pathogens and Parasites
for the Control of Mosquitoes .......................................................... 537
COOK, C. A., AND C. M. SMITH-Resistant Plants as an Alternative to Chemical
Control of Insects: Pitfalls to Progress ............................................... 546
SHOWLER, A. T., R. M. KNAUS, AND T. E. REAGAN-The Versatility of Radio-
tracer Methods for Studying Insect Ethology and Ecology ..................... 554

Continued on Back Cover

Published by The Florida Entomological Society


President ................................................ ................ R. S. Patterson
President-Elect .................................. ................. J. E. Eger
Vice-President ........................................... ................... J. F. Price
Secretary ...................... ...................... .. ............... ... J. A. Coffelt
Treasurer ....... ....................................................... A C. Knapp

J. L. Taylor
C. O. Calkins
F. Bennett
Other Members of the Executive Committee .................. E. nPe
J. E. Pefia
N. Hinkle
M. F. Antolin
J. R. McLaughlin


Editor ....................................... ......... ..................... J. R. McLaughlin

Associate Editors
Arshad Ali Carl S. Barfield Ronald H. Cherry
John B. Heppner Michael D. Hubbard Lance S. Osborne
John Sivinski Omelio Sosa, Jr. Howard V. Weems, Jr.
William W. Wirth

Business Manager ....................................... .................... A. C. Knapp

FLORIDA ENTOMOLOGIST is issued quarterly--March, June, September, and De-
cember. Subscription price to non-members is $30 per year in advance, $7.50 per copy.
Membership in the Florida Entomological Society, including subscription to Florida
Entomologist, is $25 per year for regular membership and $10 per year for students.
Inquires regarding membership, subscriptions, and page charges should be addres-
sed to the Business Manager, P. O. Box 7326, Winter Haven, FL 33883-7326.
Florida Entomologist is entered as second class matter at the Post Office in DeLeon
Springs and Winter Haven, FL.
Manuscripts from all areas of the discipline of entomology are accepted for consider-
ation. At least one author must be a member of the Florida Entomological Society.
Please consult "Instructions to Authors" on the inside back cover. Submit the original
manuscript, original figures and tables, and 3 copies of the entire paper. Include an
abstract in Spanish, if possible. Upon receipt, a manuscript is acknowledged by the
Editor and assigned to an Associate Editor who sends it out for review by at least 3
knowledgeable peers. Reviewers are sought with regard only for their expertise; Soci-
ety membership plays no role in their selection. Page charges are assessed for printed
Manuscripts and other editorial matter should be sent to the Editor, JOHN R.
MCLAUGHLIN, 4628 NW 40th Street, Gainesville, FL 32606.

This issue mailed December 30, 1988


The 72nd annual meeting of the Florida Entomological Society will be held August
7-10, 1989 at the Daytona Beach Hilton, 2637 So. Atlantic Ave., Daytona Beach, FL
32018; telephone (904)-767-7350. Registration forms and information will be mailed to
members and will appear in the Newsletter and the March, 1989 Florida Entomologist.

Notice of Change of Deadline for Submission of Papers

The deadline for submission of papers and posters for the 72nd annual meeting
of the Florida Entomological Society will be May 1, 1989. The meeting format will
be much the same as in the past with eight minutes allotted for presentation of oral
papers (with 2 minutes for discussion) and separate sessions for members who elect to
present a Project (or Poster) Exhibit. The three oral student papers and the three
student Project Exhibits judged to be the best on content and delivery will be awarded
monetary prizes during the meeting. Student participants in the judged sessions must
be Florida Entomological Society Members and must be registered for the meeting.

James R. Price, Chairman
Program Committee, FES
University of Florida, IFAS
Gulf Coast Research & Education Center
5007 60th Street East
Bradenton, FL 34203

National Center for Physical Acoustics


As part of its broad research effort in agroacoustics, The National Center for Phys-
ical Acoustics (NCPA) was pleased to sponsor the first National Agroacoustics Sym-
posium in Jackson, Mississippi on April 26-27, 1988. Florida Entomologist has provided
a forum to the participants of that symposium by the dedication of this issue to ag-
roacoustics. We gratefully acknowledge the assistance of J. C. Webb and Carrol O.
Calkins with reviewing and editing the papers.
Agroacoustics is a developing discipline which blends the detailed and systematic
observational techniques of the biological scientist with the ever increasing technical
sophistication of physical science and engineering. As this field grows, and more biolog-
ical scientists utilize acoustics instrumentation to acquire additional information about
the behavior of an individual organism, an increased need will develop for specialized
devices that are adapted to a specific insect or data acquisition modality. The NCPA
wishes to help scientists meet their individual and group research needs. We can partici-
pate in joint research efforts, provide technical advice on available equipment and, in
certain instances, develop specialized instrumentation.
The recent development of the discipline of agroacoustics is evidenced by the follow-
ing milestones: The first acoustics session of the Entomological Society of America
meeting was held in 1987; a National Agroacoustic Symposium was organized in 1988;
this agroacoustics issue of Florida Entomologist in 1988 continues the exciting progress.
The National Center for Physical Acoustics is pleased to see this progress and hopes
that it can assist existing programs and participate in the development of additional
efforts that would result in benefit to this discipline and to the country at large.
It is our sincere hope that others find the meeting and its proceedings as profitable
as we have.

Dr. Ralph R. Goodman Dr. Robert T. Walden
Laboratory Director Chairman of Symposium
Oxford, Mississippi Oxford, Mississippi

Agee: Symposium on Agroacoustics 393


Insect Attractants, Behavior, and Basic Biology Research Laboratory,
Agricultural Research Service, U.S. Department of Agriculture,
Gainesville, Florida 32604.


The nervous system of the bollworm moth, Heliothis zea (Boddie), a noctuid moth
that is a major pest of cotton, corn, and tomatoes, is served by two pairs of acoustic
sense cells. The moths use the acoustic receptors to detect the ultrasonic cries of predat-
ory bats that feed on these moths. Bats use pulsed high frequency sounds to echolocate
and capture moths for food. The moths have developed an avoidance behavioral reaction
that protects them from predatory bat capture when they detect the echolocating cries
of the bats.
A pair of acoustic receptors are located in each tympanic organ located on the lateral
wall of the metathorax on each side of the moth. Al receptor, the most sensitive unit,
can detect 20 kilohertz freqencies at sound pressure levels of 35 dB (0 dB re 20 kPa).
The A2 receptor is about 20 dB less sensitive and is also tuned to be most sensitive to
20 kHz sounds. Pulse rates of 10/sec and pulse durations of 10 msec were most effective
for eliciting evasive reactions in the bollworm moth.
In field and laboratory behavior tests, we have determined that the moths can detect
85 dB pulses of ultrasound (20 kHz) at a distance of 50-80 feet from the moth and after
detection the moths make evasive reactions.
My recent research has focused on identification of the neural circuits from the
acoustic receptors to and through the central nervous system (meso- metathoracic gang-
lia and prothoracic ganglion and brain) to the motor nerves responsible for executing
the evasive reactions. The structure of the various parts of the circuits responsible for
the behavioral reactions have been identified using histochemical techniques (cobalt
chloride and lucifer yellow) that mark only the axons carrying the acoustic information
(action potentials) and the motor nerve commands from these nerves to the muscles
responsible for directed flight.
Electrophysiological techniques were used to monitor the information flow in the
acoustic axons that feed the moth coded information on the high frequency sounds in
its environment. If the information is from the Al receptor, it is processed in the brain
to produce behavior commands that are transmitted by the motor nerves to generate
a behavioral reactions that produce turn reactions. The information from the A2 recep-
tor is transmitted to neurons in the mesothoracic ganglion directly and produce rapid
unpredictable evasive reactions (spirals, dives, and cessation of flight) and do not re-
quire "brain" processing. The anatomical circuits, behavioral reaction times, and elec-
trophysiological monitoring of neural activities confirm these findings. These and other
studies have demonstrated that the behavior of the moth is influenced or controlled by
sensory inputs that can have positive and negative effects on the moth behavior. When
the flying moth is attracted to an ultraviolet light and a sound source at the light source
generates a pulses of high frequency sound, the moth will make an evasive reaction to
the sound stimuli in preference to the continued attraction to the visual stimuli. In other
instances, another nocturnal moth species that is attracted to a sex pheromone (an
olfactory attractant) can be terminated (behavior turned off) if the trap containing the
pheromone is constructed of specific colors that cause an avoidance reaction to the visual
stimuli that are dominant over the attractive odor.
These model acoustic studies are establishing the boundaries and conditions that
must be met in the neural circuits of the central nervous system of the moth for specific
sensory stimuli to be functionally effective. Normal or usual behaviors can be turned

Florida Entomologist 71(4)

on and off when the proper sensory stimuli are presented according to specific "criter-
ion" conditions. To obtain maximum benefits from the use of non-insecticidal
technologies to control insects, a full understanding of the levels of neural processing
of sensory stimuli is needed, as is an understanding of the spheres and levels of domi-
nance that specific sensory stimuli exert in the control of the behavior of the insect pest.


El sistema nervioso de Heliothis zea (Boddie), que es una alevilla noctuida y una
plaga mayor del algod6n, maiz y tomatoes, es asistido por un par de c6lulas del sentido
acustico. Las alevillas usan los receptores ac6sticos para detectar el grito ultras6nico
de los murcielagos predatores de estas alevillas. Los murci6lagos usan pulsaciones de
sonido de alta frecuencia para localizar por el eco y capturar las alevillas como comida.
Las alevillas han desarrollado una reacci6n en su comportamiento donde evitan la cap-
tura cuando detectan el grito localizador de ecos de los murcialagos.
Un par de receptores acuisticos estAn localizados en cada 6rgano timpAnico que se
encuentran en la pared lateral del metat6rax a cada lado de la alevilla. El receptor Al
que es la unidad mas sensitiva, puede detectar frecuencias de 20 kilohertz a niveles de
presiof de sonido de 35 dB (0 dB re 20 xPa). El receptor A2 es como 20 dB menos
sensitive y tambi6n es el mas afinado y sensitive a los sonidos de 20 kHz. Pulsaciones a
raz6n de 10/segundo y pulsaciones durando 10 megasegundos fueron los mas efectivos
en educir reacciones evasivas en las alevillas.
Hemos determinado en pruebas de comportamiento en el campo y en el laboratorio,
que las alevillas pueden detectar pulsaciones de 85 dB de ultrasonidos (20 kHz) a una
distancia de 50-80 pies, y que despu6s de detectados, las alevillas reaccionan
Mis investigaciones reciente se han enfocado en la identificaci6n de los circuitos
neurales de los receptores ac6sticos hacia y a travis de del sistema nervioso central
(ganglio meso-metatoracico y ganglio protoracico y el cerebro).

The nervous system of the bollworm moth, Heliothis zea (Boddie), a noctuid moth
that is a major pest of cotton, corn, and tomatoes, is served by two pair of acoustic
sense cells (Agee 1967, Roeder & Treat 1957). The moths use the acoustic receptors to
detect the ultrasonic cries of predatory bats that feed on these moths (Agee 1969a). The
bats use pulsed high frequency sounds to echolocate and capture moths for food. The
moths have developed an avoidance behavioral reaction which protects them from pred-
atory bat capture when they detect the echolocating cries of the bats.
A pair of acoustic receptors is located in each tympanic organ located on the lateral
wall of the metathorax on each side of the moth. The Al receptor, the most sensitive
unit, can dfetect 20 kilohertz frequencies at sound pressure levels of 35 dB (Agee 1967).
The A2 receptor is about 20 dB less sensitive than the Al receptor and is most sensitive
at 20 kHz. Pulse rates of 10/sec and pulse durations of 10 msec were most effective for
eliciting evasive reactions in the bollworm moth (Agee 1969a, 1969b, Agee & Webb
1969). In response to bat cries or electronically generated pulses of ultrasound, the
acoustic receptors generate action potentials that are transmitted to the central nervous
system. In the flying moths, this information causes evasive reactions that include
directed turns, unpredictable dives, and falls to the ground. The behavior of non-flying
moths is less affected by pulses of ultrasound (Agee 1969b).
In field and laboratory behavior tests, we have determined that the moths can detect
ultrasounds at an SPL of 85 dB at a distance of 50-80 feet and make evasive reactions
(Agee 1969a,b) after detecting this sound.
Recent research has focused on identification of the neural circuits from the acoustic
receptors to and through the central nervous system (meso- metathoracic ganglia and


December, 1988

Agee: Symposium on Agroacoustics 395











AUDIO OSCILLOSCOPE *----------------

SIGNAL AVERAGER .------------------------




Fig. 1. Flow diagram of data acquisition and analysis system for neurobiological

prothoracic ganglion and brain) to the motor nerves responsible of executing the evasive
A special electronic data acquisition and analysis system was developed to selectively
record and analyze the electronic events occurring in the central nervous system in
response to pulsed ultrasound (Agee 1985a) (Fig. 1).
The action potential caused by the stimulation of ultrasound can be tracked from the
tympanic nerve through the central nervous system using special electrodes. Figure 2
compares the action potentials from the tympanic nerve, the axon in the ganglion, the
coded information on a pulse of sound identified as an action potential from a pulse
marker neuron and the action potential produced by the non-acoustic B cell in the
central nervous system as shown in Figure 1.
The structure of the various parts of the sensory input and motor nerve output
circuits responsible for behavioral reactions have been identified using histochemical

Florida Entomologist 71(4)

Fig. 2. Examples of acoustic responses recorded at (A) the tympanic nerve, (B)
repeater neuron in the mesothorax, (C) pulse marker neuron in the prothoracic ganglion,
and (D) B cell recorded from position 3 in the prothoracic ganglion. Time scale for A,
B, and C indicated on C.

techniques (cobalt chloride and lucifer yellow) that mark only the axons carrying the
acoustic information (action potentials) and the motor nerve commands from these
nerves to the muscles responsible for directed flight (Fig. 3) (Paul 1973, Orona & and
Agee 1987a,b, Tyrer & Altman 1974.
Electrophysiological techniques were used to monitor information flow in the acous-
tic axons that feed the moth coded information on the high frequency sounds in its
environment (Agee 1985a). If the information is from the Al receptor, it is processed
in the brain to produce behavior commands that are transmitted by the motor nerves
to generate behavioral reactions that produce coordinated turning. The information
from the A2 receptor is transmitted to neurons in the mesothoracic ganglion directly
and produce rapid unpredictable evasive reactions (spirals, dives, and cessation of flight)
that do not require "brain" processing. The anatomical circuits, behavioral reaction
times, and electrophysiological monitoring of neural activities confirm these findings.
These and other studies have demonstrated that the behavior of the moth is influenced
or controlled by sensory inputs that can have positive and negative effects on the moth

December, 1988


0- k

I T.)

Agee: Symposium on Agroacoustics



100 jJ

Fig. 3. Representative camera lucida reconstructions of the cells labeled following
cobalt infiltration of the tympanic nerves. The Al and B cells have synaptic terminals
scattered throughout the thoracic ganglia. The axonal terminations of the A2 cell are
confined to the meso-metathoracic ganglia.

behavior. For example, when the flying moth is attracted to an ultraviolet light and
a sound source at the light source generates pulses of high frequency sound, the moth
will make an evasive reaction to the sound stimuli in preference to the continued attrac-
tion to the visual stimuli (Agee & Webb 1969). In other instances, a nocturnal moth
species that is attracted to a sex pheromone (an olfactory attractant) can be terminated
(behavior turned off) if the trap containing the pheromone is constructed of specific
colors that cause an avoidance reaction. Visual stimuli are dominant over the attractive
odor (Mitchell et al., unpublished data). If an olfactory attractant, i.e., sex pheromone,
is presented to the moths without a repellent visual or acoustic stimulus the olfactory
attractant is effective. Figure 4 illustrates these events graphically.
Figure 5 shows what we have learned to date regarding the flow of information in
the acoustic network from the receptors through the tympanic nerve, meso-
metathoracic ganglia, prothoracic ganglion and brain to the motor nerves that control
flight of the moth. As mentioned before, there are two networks that function in the
moth to avoid bat predators. In the rapid reaction network the acoustic inputs from the
A2 receptor cause the motor nerves to operate directly and provides for the quickest
reaction possible to protect the moth. In the slow reaction network, acoustic information
that arrives in the central nervous system is transmitted to the prothoracic ganglion
and brain and receives some coding, such as the pulse marker neuron; a longer and
slower route for the information to travel before a behavior can executed. This produces



Florida Entomologist 71(4)

Effects of Sensory Stimuli on Moth Behavior










Fig. 4. Schematic diagram of the effects of sensory stimuli on the behavior of the
bollworm moth. The width of the arrows indicate the relative influence of sensory input
to the central nervous system of a flying moth. The bars indicate blockage of behaviors
elicited by specific stimuli. In event #1, all three inputs affect behavior. Event #2, the
attractant behavior induced by ultraviolet light or sex pheromones are blocked if a
repellent acoustic stimuli is received. In event #3, the olfactory stimulus of a sex
pheromone is blocked when it is being release from a color trap that is repellent. In
event #4, an olfactory stimulus (a sex pheromone) is attractive when no repellent
acoustic or visual stimuli are present.

a precise turn behavior that puts the maximum distance between the moth and the
sound source.
These "model" acoustic studies are establishing the boundaries and conditions that
must be met in the neural circuits of the central nervous system of the moth for specific
sensory stimuli to be functionally effective in controlling of their behavior. Normal or
usual behaviors can be turned on and off when the proper sensory stimuli are presented
according to specific conditions. To obtain maximum benefits from the use of non-insec-
ticidal technologies to control insects, a full understanding of the levels of neural proces-
sing of sensory stimuli is needed. We also need to know the spheres and levels of
dominance that specific sensory stimuli exert in controlling the behavior of pest insects.

December, 1988


Agee: Symposium on Agroacoustics

Antennal Lobe

Optic Lobe -







B Cell Axon--- ,
Al Axon--' SA2 Axon

Fig. 5. Schematic diagram of the tympanic neurones and associated circuitry to the
motoneurones involved in evasive flight behavior. Two behaviorally-relevant systems
appear to be present. The A2 acoustic cell has direct monosynaptic connections to the
motoneurones within the meso-metathoracic ganglia, forming a rapid reaction network.
On the other hand, the A and B cells appear to be additionally linked to interneurones
and the brain reflecting their involvement in a slower and directed reaction network.


AGEE, H. R. 1967. Response of acoustic sense cell of the bollworm and tobacco bud-
worm to ultrasound. J. Econ. Ent. 60: 366-369.
AGEE, H. R. 1969a. Response of flying bollworm moths and other tympanate moths
to pulsed ultrasound. Ann. Ent. Soc. America 62: 801-807.
AGEE, H. R. 1969b. Response of Heliothis spp. (Lepidoptera: Noctuidae) to ul-
trasound when resting, feeding, courting, mating, or ovipositing. Ann. Ent. Soc.
America 62: 1122-1128.
AGEE, H. R. 1985a. Neurobiology of the bollworm moth: Information flow in the
central nervous system. J. Agric. Ent. 2: 277-284.



Florida Entomologist 71(4)

December, 1988

AGEE, H. R. 1985b. Neurobiology of the bollworm moth: Neural elements controlling
behavioral responses to pulsed ultrasound. J. Agric. Ent. 2: 345-350.
AGEE, H. R., AND E. ORONA. 1988. Neural basis of evasive flight behavior in re-
sponse to acoustic stimulation in Heliothis zea (Lepidoptera: Noctuidae): I. Or-
ganization of the tympanic nerves. Ann. Ent. Soc. America (In Press).
AGEE, H. R., AND J. C. WEBB. 1969. Effects of ultrasound on capture of Heliothis
zea and Ostrinia nubilalis moths in traps equipped with ultraviolet lamps. Ann.
Entomol. Soc. America 62: 1248-1252.
ORONA, E., AND H. R. AGEE. 1987a. Thoracic mechanoreceptors in the wing bases
of Heliothis zea (Lepidoptera : Noctuidae) and their central projections. J. Insect
Physiol. 33: 713-721.
ORONA, E., AND H. R. AGEE. 1987b. An insect model system for the analysis of
sensory coding: auditory processing in the noctuid moth, Heliothis zea. Proceed-
ings of the Society for the Advancement of Chicanos and Native Americans in
Science (SACNAS) September 1986, Pasadena, CA.
ORONA, E., AND H. R. AGEE. 1988. Neural basis of evasive flight behavior in re-
sponse to acoustic stimulation in Heliothis zea (Lepidoptera: Noctuidae):
motoneuronal innervation of flight muscles. Ann. Ent. Soc. America (In press).
PAUL, D. H. 1973. Central projections of the tympanic fibres in noctuid moths. J.
Insect Physiol. 19: 1785-1792.
ROEDER, K. D., AND A. E. TREAT. 1957. Ultrasonic reception by the tympanic organ
of noctuid moths. J. Exp. Zool. 134: 127-158.
ROEDER, K. D. 1966. Interneurons of the thoracic nerve cord activated by tympanic
nerve fibres in noctuid moths. J. Insect Physiol. 12: 1227-1244.
TYRER, N. M., AND J. S. ALTMAN. 1974. Motor and sensory flight neurones in a
locust demonstrated using cobalt chloride. J. Comp. Neurol. 157: 1177-138.


Biology Department, Creighton University,
Omaha, Nebraska 68178-0103, U.S.A.


Animal signals evolve as adaptations to social as well as physical environments.
Where the interests of signallers and responders differ, a coevolutionary "arms race"
cycle of signal adoption, exaggeration, and devaluation may result. Stable, reliable
signals evolve when costs of signalling constrain the evolution of bluff and exaggeration.
Acoustic signals are especially costly, compared to other signal types, and are therefore
especially likely to evolve as reliable signals in such "social competition" situations.
Costs of acoustic signals include physiological costs of growth and signal production,
probes by conspecific rivals and discriminating members of the other sex, and attacks
by natural enemies such as predators and parasites.


Sefiales dadas por los animals se han desarrollado como adaptaciones al medio
ambiental fisico asi como al social. Donde el interns del que sefiala y del que responded
difieren, pudiera entonces resultar en un ciclo coevolucionario de "carrera hacia las
armas" en la sefial de adaptaci6n, exageraci6n, y devaluaci6n. Senales estables y confi-

Burk: Symposium on Agroacoustics

ables se desarrollan cuando los costs de sefiala constrifien la evoluci6n de la decepci6n
y la exageraci6n. Las sefiales acisticas son especificamente costosas cuando se comparan
con otros tipos de sefiales, y es muy possible que se desarrollen como sefiales confiables
en situaciones de "competencia social". Los costs de sefiales achsticas incluyen el costo
fisiol6gico de crecimiento y la producci6n de la serial, exploraci6n por rivals con-es-
pecificos y miembros discriminantes del otro sexo, y ataques por enemigos naturales
como predatores y parAsitos.


Are the communication channels used by different animal species arbitrary? Or,
does the variation among species in communication channels represent different adap-
tive solutions to varying ecological and social situations? To answer these questions,
behavioral ecologists have recently begun to explore the mesh between communication
channels and the environment (see discussion in Burk 1988). The approach has often
been to take each type of signal in turn, and ask what its costs and benefits are, with
the hope that knowing these will allow one to predict the kinds of environments in which
it would be adaptive. For example, Alcock (1984) provides a table in which chemical,
auditory, visual, and tactile channels are compared for ability to reach receivers, infor-
mation available, and cost to sender. Acoustic (auditory) signals-my focus in this
paper-are characterized as long range, with fast transmission rates, flowing around
barriers, of use day or night, having fast fadeout time, being fairly easy to localize,
having a medium risk of exploitation (visual signals considered to be of higher risk),
and being of high broadcast expense (all other signals being characterized as of low or
low-moderate expense).
Such assessments are of great value, but a slightly different perspective can greatly
add to our understanding of the reasons for the frequent evolution of acoustic signalling
in animals. This involves an appreciation of the dynamics of communication evolution.
Communication necessarily involves two individuals, a signaller and a receiver, both
subject to natural selection. Krebs & Dawkins (1984, see also Dawkins & Krebs 1978)
have characterized these two parties as a "manipulator" and a "mind-reader", respec-
tively. Manipulator-signallers are selected to produce signals that effectively elicit re-
sponses in receivers that are beneficial to the signaller. Mind-reader-receivers, on the
other hand, are under selection only to respond in ways that are beneficial to them-
selves. Sometimes, the interests of signallers and receivers will lead to an "agreed"
response. Under such circumstances, as in communication between close relatives or
between mates involved in brood care, communication may evolve to be accurate &
efficient. An excellent example is the waggle-dance foraging communication system of
honeybees (Winston 1987). On the other hand, where the interests of the signaller and
receiver do not necessarily coincide, as in intrasexual competition or intersexual mate
choice situations, a complex co-evolutionary process is expected to be set in train. By
analogy to human processes, Dawkins & Krebs (1979) called such dynamic coevolutio-
nary events "arms races."
How will signalling systems evolve in such "social competition" arms races (West-
Eberhard 1979)? A number of authors, for a number of contexts, have suggested that
complex signal systems with multiple signals, each a rather weak predictor of sub-
sequent behavior, will result (Barnard & Burk 1979: dominance hierarchies; Burk 1981:
sexual signalling; Andersson 1980: threat signals generally). Andersson's (1980) paper,
"Why are there so many threat displays?", presents the expected scenario clearly: In
agonistic encounters, some incidental movement, such as baring the teeth, may be well
correlated with likelihood of attack. Natural selection may favor a retreat response by
opponents when such a movement occurs. However, once a response spreads, natural

Florida Entomologist 71(4)

December, 1988

selection favors "bluff" teeth-baring by individuals with lower likelihood of attack.
Now natural selection on responders will lower their likelihood of retreat; the original
threat display, while not meaningless, has nevertheless become devalued. At this point,
responders may be selected to use a second attack-associated movement as a cue along
with teeth-baring. This second movement will then go through a similar cycle of rituali-
zation and devaluation. The cycle may occur many times, leading to agonistic encounters
consisting of a string of threat displays. A similar arms race scenario may occur in
courtship: females look for indicators of male quality, these indicators become ritualized
and elaborated, and ultimately females devalue them and evolve responses to additional
indicator traits (West-Eberhard 1979, Burk 1981).


If such elaboration-devaluation arms races are common, why aren't agonistic or
courtship encounters infinitely long? Does this simply reflect the possibility that we
commonly see communication processes that are somewhere near the beginning of a
coevolutionary spiral? Or is it rather due to the existence of some factor that can halt
the arms race spiral and stabilize a communication system at a point where displays
contain only slight elements of bluff and where there will be little selection for response
devaluation? That the answer is likely to be the latter was suggested by Zahavi in an
important paper in 1977 (Zahavi 1977). Zahavi argued that stable "honest" or "reliable"
communication systems will evolve when responders only consider ". .. signals that are
not easily open to cheating . a signal is reliable when the difficulty of its performance
is related to its meaning .. ." Zahavi argued that ". . cost is a necessary component
of the signal; the more significant the signal, the higher the cost to the performer ...
characters important in determining quality should be affected adversely by the signal"
(Zahavi 1977).
To state Zahavi's thesis more explicitly, some signals are costlier to produce than
others. Only vigorous, healthy signallers will be able to develop the signal producing
mechanisms these require, and/or to withstand the physiological burden of their produc-
tion. When, in the course of communication coevolution such a signal is "chosen" by
natural selection on responders, exaggeration and bluffing are constrained. Little de-
valuation of response evolves either; the communication system is stabilized when
costly, and therefore "reliable", signals have evolved (see discussion in Wiley 1983).


Increase in
Metabolism Over
Animal Resting Rate Reference

Euconocephalus nasutus katydids 14.2X Stephens & Josephson 1977
Neoconocephalus robustus katydids 15.3X Stephens & Josephson 1977
Teleogryllus commodus crickets 3.9X Kavanagh 1987
Anurogryllus arboreus crickets 10.0-15.8X Prestwich & Walker 1981
Oecanthus celerinictus tree crickets 6.2-12.0X Prestwich & Walker 1981
0. quadripunctatus tree crickets 6.5-8.0X Prestwich & Walker 1981
Gryllotalpa australis mole crickets 13.4X Kavanagh 1987
Cystosoma saundersii cicadas 18.4X MacNally & Young 1981
Physalaemus pustulosus frogs 2.1-4.3X Ryan 1985
Hyla versicolor tree frogs 5-22X (X = 12.4X) Taigen & Wells 1985
H. crucifer tree frogs 14.0X Taigen et al. 1985


Burk: Symposium on Agroacoustics


The thesis of this paper is that acoustic signals are particularly likely to be the stable
end points of communication coevolutionary events. Paradoxically, acoustic signals will
commonly act as honest or reliable cues because they are especially costly to produce,
compared to other types of signals. I will argue that three types of costs maintain the
reliability and thus stability of acoustic communication systems: (1) Acoustic signals are
physiologically very expensive to produce; (2) Acoustic signals, because they are con-
spicuous, elicit probes of the signaller's vigor by conspecific social competitors (bluffs
may be called!); (3) Acoustic signals, because they are conspicuous, attract natural
enemies such as predators, parasitoids, and parasites.
In the remainder of this paper, I attempt to document the costliness of acoustic
signals and the way in which they become reliable cues. To properly test my hypothesis,
it would be preferable to compare the costs of acoustic signals with those of other
signals, such as visual or pheromonal ones. However, a cursory glance at the literature
suggests that few attempts to quantify the costs of signals other than acoustic ones have
been made. I will therefore review only the information available on acoustic signals,
in the hope that this may stimulate others to collect comparative data for other signals.



Some male frogs and insects are so highly motivated to produce acoustic signals that
they will do so normally even when enclosed in a respirometer (Kavanagh 1987). Using
this device, one can measure metabolic rates of calling animals and can compare them
with metabolic rates at rest or when performing other behaviors such as walking or
eating. A range of values from the literature is given in Table 1. On average, there is
an increase in metabolism by calling animals of about an order of magnitude. Acoustic
calling is usually one of the most energetically expensive things an animal does: the
mass-specific rate of oxygen consumption by calling Hyla versicolor frogs is the highest
measured for any ectothermic vertebrate (Given 1988), while only the demands of flight
exceed those of calling in some insects (Prestwich & Walker 1981).
Knowing the energy cost of sound production, and measuring the sound energy
levels in the acoustic field around a signaller, allows one to calculate the energetic
efficiency of sound production. Values from the literature are given in Table 2. They
are all very low, in no case more than a few percent. Earlier higher estimates, such as
those of Bennet-Clark (1971) for mole crickets (35%) and Counter (1977) for katydids
(26%), have been re-evaluated recently and lowered by an order of magnitude


Animal % Efficiency Reference

Cystosoma saundersii cicadas 0.82 MacNally & Young 1981
Neoconocephalas robustus katydids 2.1 MacNally & Young 1981
Gryllotalpa vinae mole crickets 3.41 Kavanagh 1987
G. gryllotalpa mole crickets 0.5 Kavanagh 1987
G. australis mole crickets 1.05 Kavanagh 1987
Teleogryllus commodus crickets 0.05 Kavanagh 1987
Anurogryllus arboreus crickets 0.23 Kavanagh 1987
Physalaemus pustulosus frogs 0.05-1.2 Ryan 1985
Gallus domesticus cockerels 2.0 Brackenbury 1980

Florida Entomologist 71(4)

(Kavanagh 1987, MacNally & Young 1981). It is interesting that the one value for a
bird is in the same range as those of anurans and insects.
In the absence of information on the metabolic cost and efficiency of other signal
types, it is impossible to draw firm conclusions. It seems likely, however, that acoustic
signals will turn out to be highest in cost and lowest in efficiency, because of their
transitory nature (requiring their repetition), omnidirectionality, and the extensive
muscle movements involved in their production-often a production mechanism involv-
ing substantial frictional forces.


The above information, while striking, fails to give an adequate impression of signal-
ling costs. The presence of "social competitors" (West-Eberhard 1979) such as rival
males or discriminating females imposes certain physical signal forms and high signal
rates on acoustic callers, so that total costs of signalling are multiplied, as the following
discussion will show.
Sexually advertising males, for example, often call at high rates for hours a day,
and for weeks or months during prolonged breeding seasons. Rates of acoustic display
by a variety of animals are taken from the literature and presented in Table 3. In a
variety of species, mate-choosing females have been shown to respond disproportion-
ately often to males who call at the highest rates or for the longest durations (fruit flies:
Sivinski et al. 1984; crickets: Hedrick 1986; frogs: Halliday 1987; sage grouse: Gibson &
Bradbury 1986; red deer: McComb 1987). In many species, males greatly increase their
calling rates when females are present (Sivinski & Webb 1986, Taigen & Wells 1985),
or switch to a more energetically demanding call type (Ryan 1985).
The physical form of acoustic signal favored by choosing females is often demanding
of male vigor. Females prefer loud calls (Forrest 1980, Halliday 1987), and measured
intensity levels of male calls are often astounding. Hyla versicolor frog calls averaged
109 dB at 50 cm from the caller (Wells & Taigen 1986), Scapteriscus acletus mole cricket
calls 91 dB at 15 cm (Forrest 1980), Neoconocephalus robustus katydid calls 116 dB at
1 cm (Counter 1977), and Cystosoma saundersii cicada calls 91 dB at 20 cm (MacNally
& Young 1981). Females also prefer individual calls of long duration (Halliday 1987),
calls with additional notes (Ryan 1985) and-very commonly-calls of low frequency
that can be produced only by the largest males (Halliday 1987, Morton 1977, Webb et
al. 1984, Latimer & Sippel 1987). [Anuran calls are longer in duration and lower in
frequency at lower temperatures; it is possible that a small frog or toad could fake larger


Animal Display Rate Reference

Neoconocephalus Wing closures during stridulation Stephens & Josephson
robustus katydids (250C): 673,000/hr. 1977
Cystosoma saundersii tymbal muscle "twitch" rate MacNally & Young
cicadas (21.5C): 1981
72,000/30 min. calling period
Hyla crucifer call rate (160C): Taigen et al. 1985
tree frogs 4500/hr.
Centrocercus urophasianus "strut" vocalizations: Krebs & Harvey 1988
sage grouse 720-1080/2-3 hour display period
Cervus elephus roaring rate during rut (31 days): Clutton-Brock et al.
red deer 2.7/min. 1982


December, 1988

Burk: Symposium on Agroacoustics

size by calling from colder sites in the water at a pond's edge. In Fowler's toad (Bufo
woodhousei fowleri), Fairchild (1981) has shown that such cheating is prevented by
male-male competition: large males occupy these pond-edge sites & displace small males
onto the wormer pond bank. Thus males that already have long duration, low frequency
calls exaggerate these traits even more. "To those that have shall be given . ."]
Comparative studies support the hypothesis that female choice leads to the evolution
of especially low-pitched calls: female Physalaemus pustulosus frogs and Tettigonia
cantans katydids have been shown to mate preferentially with males producing low-fre-
quency calls, and each species has a much lower calling song fundamental frequency
than would be expected for a species of that size in that taxonomic group (Ryan 1985,
Latimer & Schatral 1986).
The costs of such strenuous & extended signalling can be seen in the weight loss
experienced by breeding males. Some frogs lose over 1% per day (Halliday 1987), some
orthopterans 2-3% in 1-2 h (Dodson et al. 1983), and red deer stags 20% in a month
(Clutton-Brock et al. 1982). The males that were in the best physical condition at the
beginning of the sexual display season may be able to sustain costly signalling in the
face of such losses longer than less vigorous males. In a number of frogs, the best
correlate of mating success is length of time spent at a mating site (Halliday 1987); this
is probably true for many other animals with extended sexual advertisement seasons.
Effects of costly signalling extend beyond the immediate breeding season. With
males putting as much as 86% of their energy assimilation into calling (Ranidella frogs:
MacNally 1981), it is not surprising that there is often a negative relationship between
reproductive activity and growth (frogs: Given 1988) or survival ungulatess: Clutton-
Brock et al. 1982). Just to develop the sound producing structures may impose consid-
erable costs on signallers, with effects on overall growth and mortality rates. In frogs,
the trunk muscles used in vocalizations amount to as much as 15% of a male's body
mass, compared to 3% of a female's (Taigen et al. 1985).
Production of conspicuous acoustic signals attracts another class of "social com-
petitors", rival males. The songs that are favored by females have also been shown to
attract rival males in animals as diverse as crickets (Cade 1979, Burk 1983), frogs (Given
1987), and cowbirds (West et al. 1981). The aggressive encounters that subsequently
occur are frequently settled on the basis of qualities of acoustic signals that are unbluff-
able, such as low frequency-correlated with size (frogs & toads: Arak 1983, Given 1987;
katydids: Latimer & Sippel 1987; birds & mammals: Morton 1977)--or high acoustic
display rate-correlated with physical condition (red deer: Clutton-Brock et al. 1982).
Small, subordinate, or out-of-condition males may not be able to bear the costs of such
aggressive challenges and may opt for non-calling alternative strategies (Cade 1980),
leaving acoustic signals as relatively reliable indications of the vigor, large size, or
dominance of their producers (Burk 1983).


As we have seen, an acoustically-signalling animal may be heard by individuals other
than its intended targets: a calling male attracts rivals as well as potential mates.
Perhaps even more costly to signallers is another class of "eavesdroppers" (Alcock
1984): predators, parasitoids, and parasites may orient acoustically to calling hosts. The
first demonstration of such acoustic orientation by natural enemies was by Walker
(1964); the number of examples has steadily increased, as seen in Table 4. Mortality
rates of callers from such eavesdroppers can be very high: 1% or more per hour calling
in Physalaemus pustulosus frogs (Ryan 1985), over 90% per season for crickets &
katydids (Cade 1979, Burk 1982). Noteworthy is the fact that the calls which are most
attractive to females, such as loud, complex calls, are also typically more attractive to


Florida Entomologist 71(4)

December, 1988


Prey Predator/Parasite Reference

Crickets Cat Walker 1964
Crickets Ormiine Tachinid Flies Cade 1975
Crickets Herons Bell 1979
Crickets Geckos Sakaluk & Belwood 1984
Cicadas Sarcophagid Flies Soper et al. 1976
Katydids Bats Tuttle et al. 1985
Katydids Ormine Tachinid Flies Burk 1982
Frogs Mosquitos McKeever 1977
Frogs Bats Ryan 1985

natural enemies (Ryan 1985). The costs of conspicuous calling may be bearable for
vigorous males that may be more able to escape predators or to compete successfully
within the large choruses that evolve as anti-predation defenses (Ryan 1985), but these
costs may lead to a loss of calling by less vigorous or less dominant individuals. Again,
the cost of acoustic signalling tends to restrict the subset of individuals that sings,
keeping the communication system more "honest" for conspecific responders.


Without making exaggerated claims, it seems to me that the view of the importance
of acoustic signals presented above is revelant to applied entomologists for three
reasons. First, and most generally, in order to design the most appropriate integrated
pest management programs, one needs to know as much as possible about the ecology
and behavior of the target species. Since development rates, relative development of
different parts of the body, extent and type of sexual dimorphism, dispersion and disper-
sal characteristics, and many other features of a species' biology may be shaped by
sexual selection involving acoustic display, one certainly needs to appreciate the impor-
tance of such signals. Second, for pest species producing conspicuous acoustic signals,
the strong possibility exists that there are acoustically-orienting natural enemies. A
systematic search for such species, especially dipteran parasites of pest orthopterans,
would be justified (see Walker, this symposium, for one example). Third, I have argued
that acoustic signalling is one of the most demanding activities animals engage in, one
that only possessors of vigorous phenotypes can sustain for long. If so, then acoustic
signals should be one of the first traits to change as laboratory populations undergo
domestication. In mass rearing programs, as in applications of the sterile insect
technique, quality control managers could use decrease in call rate or degradation of
acoustic signal form as a "canary-in-a-coal-mine" early indicator of quality declines.


To understand the evolution of animal communication systems, one has to adopt a
dynamic, coevolutionary perspective. Signals will evolve that reflect not only a fit to
the physical environment, but also to the social situation involved. Stable communication
systems, ones buffered against arms-race cycles of exaggeration and devaluation, will
be ones in which signals remain reliable or honest because of their high costs to signal-
lers. Because acoustic signals are particularly costly, they may frequently act as such
reliable cues. As with other coevolutionary processes, the resultant communication
systems may not be "optimal", but they will be evolutionarilyy stable" (Dawkins 1980).

Burk: Symposium on Agroacoustics 407

Note added in proof: Similar ideas about the evolution of acoustic signals have recently
appeared in an article by M. J. Ryan (American Zool. 28: 885-898).


I would like to thank Drs. R. T. Walden and J. C. Webb for their roles in my
invitation to participate in the National Agroacoustics Symposium, Drs. J. Sivinski, T.
J. Walker, W. J. Bailey, and M. D. Greenfield for commenting on the manuscript, Mrs.
Pat Smith for preparing the manuscript, and Dean M. Proterra, S.J., of the College of
Arts & Sciences of Creighton University for travel support.


ALCOCK, J. 1984. Animal Behavior: an Evolutionary Approach. 3rd ed. Sinauer: Sun-
derland, Ma. 596 p.
ANDERSSON, M. 1980. Why are there so many threat displays? J. Theor. Biol. 86:
ARAK, A. 1983. Sexual selection by male-male competition in natterjack toad
choruses. Nature 306: 261-2.
BARNARD, C. J., AND T. BURK. 1979. Dominance hierarchies and the evolution of
"individual recognition". J. Theor. Biol. 81: 65-73.
BELL, P. D. 1979. Acoustic attraction of herons by crickets. J. New York Entomol.
Soc. 87: 126-7.
BENNET-CLARK, H. C. 1971. Acoustics of insect song. Nature 234: 255-9.
BRACKENBURY, J. 1980. Respiration and production of sounds by birds. Biol. Rev.
55: 363-78.
BURK, T. 1981. Signaling and sex in acalyptrate flies. Florida Entomol. 64: 30-43.
.1982. Evolutionary significance of predation on sexually signalling males.
Florida Entomol. 65: 90-104.
S1983. Male aggression and female choice in a field cricket (Teleogryllus
oceanicus): the importance of courtship song. Pp. 97-119 In D. T., Gwynne and
G. K. Morris, eds. Orthopteran Mating Systems: Sexual Competition in a Diverse
Group of Insects. Westview: Boulder.
1988. Insect behavioral ecology: some future paths. Ann. Rev. Entomol. 33:
CADE, W. H. 1975. Acoustically orienting parasitoids: fly phonotaxis to cricket song.
Science 190: 1312-3.
S1979. The evolution of alternative male reproductive strategies in field cric-
kets. Pages 343-79 In M. S. and N. A. Blum, eds. Sexual Selection and Reproduc-
tive Competition in Insects. Academic: New York.
1980. Alternative male reproductive behaviors. Florida Entomol. 63: 30-45.
Behavior and Ecology of Two Sexes. Univ. of Chicago: Chicago. 278 p.
COUNTER, S. A., JR. 1977. Bioacoustics and neurobiology of communication in the
tettigoniid Neoconocephalus robustus. J. Insect Physiol. 23: 993-1008.
DAWKINS, R. 1980. Good strategy or evolutionarily stable strategy? Pages 331-67 In
G. W. Barlow and J. Silverberg, eds. Sociobiology: Beyond Nature/Nurture.
Westview: Boulder.
-- AND J. R. KREBS. 1978. Animal signals: information or manipulation? Pages
282-309 In J. R. Krebs and N. B. Davies, eds. Behavioural Ecology: an Evolutio-
nary Approach. Blackwell Sci. Publ.: Oxford.
-- AND 1979. Arms races between and within species. Proc. R. Soc.
Lond. B205: 489-511.
DODSON, G. N., G. K. MORRIS, AND D. T. GWYNNE. 1983. Mating behavior in the
primitive orthopteran genus Cyphoderris (Haglidae). Pages 305-18 In D. T.
Gwynne and G. K. Morris, eds. Orthopteran Mating Systems: Sexual Competi-
tion in a Diverse Group of Insects. Westview: Boulder.

Florida Entomologist 71(4)

FAIRCHILD, L. 1981. Mate selection and behavioral thermoregulation in Fowler's
toads. Science 212: 950-1.
FORREST, T. G. 1980. Phonotaxis in mole crickets: its reproductive significance.
Florida Entomol. 63: 45-53.
GIBSON, R. M., AND J. W. BRADBURY. 1986. Male and female mating strategies on
sage grouse leks. Pages 379-398 In D. I. Rubenstein and R. W. Wrangham, eds.
Ecological Aspects of Social Evolution: Birds and Mammals. Princeton Univ.:
GIVEN, M. F. 1987. Vocalization and acoustic interactions of the carpenter frog, Rana
virgatipes. Herpetologica 43: 467-81.
1988. Growth rate and cost of calling activity in male carpenter frogs, Rana
virgatipes. Behav. Ecol. Sociobiol. 22: 153-60.
HALLIDAY, T. R. 1987. Physiological constraints on sexual selection. Pages 247-264
In J. W. Bradbury and M. B. Andersson, eds. Sexual Selection: Testing the
Alternatives. John Wiley: Chichester.
HEDRICK, A. V. 1986. Female preferences for male calling bout duration in a field
cricket. Behav. Ecol. Sociobiol. 19: 73-7.
KAVANAGH, M. W. 1987. The efficiency of sound production in two cricket species,
Gryllotalpa australis and Teleogryllus commodus (Orthoptera: Grylloidea). J.
Exp. Biol. 130: 107-19.
KREBS, J. R., AND R. DAWKINS. 1984. Animal signals: mind-reading and manipula-
tion. Pages 380-402 In J. R. Krebs and N. B. Davies, eds. Behavioural Ecology:
an Evolutionary Approach, 2nd ed. Sinauer: Sunderland, Ma.
---, AND P. H. HARVEY. 1988. Lekking in Florence. Nature 333: 12-3.
LATIMER, W., AND A. SCHATRAL. 1986. Information cues used in male competition
by Tettigonia cantans. Anim. Behav. 34: 162-168.
- AND M. SIPPEL. 1987. Acoustic cues for female choice and male competition
in Tettigonia cantans. Anim. Behav. 35: 887-900.
MACNALLY, R. C. 1981. An analysis of factors affecting metabolic rates of two species
of Ranidella (Anura). Comp. Biochem. Physiol. 69A: 731-7.
-- AND D. YOUNG. 1981. Song energetic of the bladder cicada, Cystosoma
saundersii. J. Exp. Biol. 90: 185-96.
MCCOMB, K. 1987. Roaring by red deer stags advances the date of oestrus in hinds.
Nature 330: 648-9.
MCKEEVER, S. 1977. Observations of Corethrella feeding on tree frogs (Hyla). Mos-
quito News 37: 522-3.
MORTON, E. S. 1977. On the occurence and significance of motivation-structural rules
in some bird and mammal sounds. Am. Nat. 111: 855-69.
PRESTWICH, K. N., AND T. J. WALKER. 1981. Energetics of singing in crickets:
effect of temperature in three trilling species (Orthoptera: Gryllidae). J. Comp.
Physiol. 143: 199-212.
RYAN, M. J. 1985. The Tungara Frog: a Study in Sexual Selection and Communica-
tion. Univ. of Chicago: Chicago. 230 p.
SAKALUK, S. K., AND J. J. BELWOOD. 1984. Gecko phonotaxis to cricket calling song:
a case of satellite predation. Anim. Behav. 32: 659-62.
SIVINSKI, J., T. BURK, AND J. C. WEBB. 1984. Acoustic courtship signals in the
Caribbean fruit fly, Anastrepha suspense (Loew). Anim. Behav. 32: 1011-6.
- AND J. C. WEBB. 1986. Changes in a Caribbean fruit fly acoustic signal with
social situation (Diptera: Tephritidae). Ann. Entomol. Soc. America 79: 146-149.
SOPER, R. G., 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). Canadian Entomol. 108:
STEPHENS, E. D., AND R. K. JOSEPHSON. 1977. Metabolic rate and body tempera-
ture in singing katydids. Physiol. Zool. 50: 31-42.
TAIGEN, T. L., AND K. D. WELLS. 1985. Energetics of vocalization by an anuran
amphibian (Hyla versicolor). J. Comp. Physiol. B 155: 163-70.


December, 1988

Calkins & Webb: Symposium on Agroacoustics


- AND R. L. MARSH. 1985. The enzymatic basis of high metabolic rates
in calling frogs. Physiol. Zool. 58: 719-26.
TUTTLE, M. D., M. J. RYAN, AND J. J. BELWOOD. 1985. Acoustical resource parti-
tioning by two species of phyllostomid bats (Trachops cirrhosus and Tonatia
sylvicola). Anim. Behav. 33: 1369-71.
WALKER, T. J. 1964. Experimental demonstration of a cat locating orthopteran prey
by the prey's calling song. Florida Entomol. 47: 163-5.
WEBB, J. C., J. SIVINSKI, AND C. LITZKOW. 1984. Acoustical behavior and sexual
success in the Caribbean fruit fly, Anastrepha suspense (Loew) (Diptera: Tep-
hritidae). Environ. Entomol. 13: 650-656.
WELLS, K. D., AND T. L. TAIGEN. 1986. The effect of social interactions on calling
energetic in the gray treefrog (Hyla versicolor). Behav. Ecol. Sociobiol. 19:
WEST, M. J., A. P. KING, AND D. H. EASTZER. 1981. Validating the female bioassay
of cowbird song: relating differences in song potency to mating success. Anim.
Behav. 29: 490-501.
WEST-EBERHARD, M. J. 1979. Sexual selection, social competition, and evolution.
Proc. American Phil. Soc. 123: 222-34.
WILEY, R. H. 1983. The evolution of communication: information and manipulation.
Pages 156-189 In T. R. Halliday and P. J. B. Slater, eds. Animal Behaviour-2:
Communication. W. H. Freeman: N.Y.
WINSTON, M. L. 1987. The Biology of the Honey Bee. Harvard Univ: Cambridge,
Ma. 281 p.
ZAHAVI, A. 1977. Reliability in communication systems and the evolution of altruism.
Pages 253-259 In B. Stonehouse and C. Perrins, eds. Evolutionary Ecology.
MacMillan: London.


Insect Attractants, Behavior, and Basic Biology Research Laboratory
Agriculture Research Service, U. S Department of Agriculture
Gainesville, Florida 32604


Larvae of the Caribbean fruit fly Anastrepha suspense (Loew) can be detected by
the sounds they make while feeding using an acoustical detection system. Efforts to
improve and determine the efficiency of the system led to detailed studies of the feeding
behavior of larvae in grapefruit. Movement, feeding and growth are related to the
maturity of the fruit. As the fruit matured, it became more sweet and larvae fed more
consistently and voraciously, moved into the pulp portion earlier, and developed more
rapidly. Larvae were detected within hours after they hatched from eggs, when they
are often too small to be seen in the fruit with the unaided eye. The efficiency of the
system in detecting infested fruit has been demonstrated to be more efficient than the
accepted method of cutting and visual examination.


Larvas de la mosca de frutas del Caribe, Anastrepha suspense (Loew), se pueden
detectar por el sonido que hacen cuando comen usando un sistema de detecci6n acustica.

410 Florida Entomologist 71(4) December, 1988

Esfuerzos para mejorar y determinar la eficiencia del sistema nos dirigi6 hacia studios
detallados del comportamiento de las larvas cuando comen toronjas. El movimiento,
alimentaci6n y crecimiento estdn relacionados con la madurez de la fruta. A media que
la fruta madura, se pone mas dulce, y las larvas se alimentaron mis consistent y
vorazmente, se movieron hacia las porciones de pulpa mas temprano, y se desarrollaron
mas rapidamente. Las larvas se detectaron a pocas horas de salir de los huevos cuando
a menudo todavia eran muy pequefas para verse a simple vista en la fruta. Se ha
demostrado la eficiencia del sistema en detectar frutas infestadas y es mIs eficiente que
el aceptado m6todo de cortar y examiner visualmente.

The Caribbean fruit fly, Anastrepha suspense (Loew) (caribfly), has inhabited
Florida since 1966. Although most of its hosts are wild and dooryard fruits, it does
occasionally infest citrus, particularly grapefruit and oranges (Swanson & Baranowski
1972). Although it has never been found to cause extensive damage in citrus groves,
the threat of the presence of its larvae in citrus fruit being shipped to other tropical
and subtropical areas has caused it to become subject to quarantine regulations. Florida
citrus is now quarantined by Arizona, California, Hawaii, Texas and Japan.
Prior to 1983, fumigation by ethylene dibromide (EDB) successfully controlled eggs
and larvae in fruit. However, after the withdrawal of this compound, citrus exporters
were only able to use methyl bromide, cold treatment and shipment of fruit from fly-free
areas as means of overcoming the quarantine restrictions. In all cases, because these
methods are more complicated and less effective than EDB, a large sample of fruit must
be examined by cutting to determine the presence or absence of caribfly larvae.
When it was discovered that sounds of feeding by larvae of the caribfly could be
detected by use of an accelerometer (Webb & Landolt 1984), a new technology for the
detection of fruit fly infested fruit was developed. Subsequent improvements in the
system eliminated the accelerometer and the frequency of the feeding sounds produced
were altered at the detector due to the physics of the system (Webb et al. 1988). To
improve on the efficiency of the system and to help explain the variability in the sounds
detected, it was necessary to determine the feeding behavior of the larvae in the target
fruit. It also became necessary to determine patterns of movement in relation to age of
larvae producing the audible signal and the maturity of the fruit so that standard com-
parisons could be made between different equipment modifications.


Grapefruit in Central Florida blooms in March. The fruit picking starts in late Oc-
tober and extends to May of the following year. Grapefruit (var. Marsh White) used in
this study were picked every four weeks from November, 1984 to April, 1985 from a
grove maintained on Merritt Island, Florida. The fruits were transported to the Insect
Attractants, Behavior and Basic Biology Research Laboratory in Gainesville where
they were washed and infested with eggs from gravid female caribflies. Fruits were
placed into cages containing large numbers of sexually mature flies and the females
were observed for probing and oviposition activity. The fruits were exposed to the flies
for 6 hours, then were removed from the cage and placed on trays and incubated for 3
days at 25 C. From each collection, 45 to 60 fruits were infested, 15 on each of 3 days.
After 3 days, the fruits were placed individually on the sound detector (Webb et al.
1988) to see if eggs had hatched and the larvae had begun feeding. Under ideal condi-
tions in the laboratory at 25' C., eggs hatch in 3 days. If feeding sounds were detected,
the date of detection was written on the fruit and the fruit was returned to the incuba-
tion tray. If no sound was heard, the fruits were examined acoustically each day there-
after until larval sounds were apparent.

Calkins & Webb: Symposium on Agroacoustics 411





Fig. 1. The three major regions of a citrus fruit. Flavedo is the colored outside layer
containing the oil glands. The albedo is the white fibrous inner layer. The pulp is the
edible portion from which the juice is extracted.

The three main layers of fruit are illustrated in Figure 1. The flavedo is the yellow
portion of the grapefruit peel. It contains most of the oils in the peel. The albedo is the
white layer between the flavedo and the pulp and is made up mostly of pectin, but
several chemicals including narangin and limonin also occur there (Greany et al. 1983,
Kefford & Chandler 1970). The pulp is the yellow edible center portion of the fruit.
Each of these portions of the fruit were dissected to locate larvae.
Three fruits were dissected to locate larvae at 2-day intervals after the first larval
sounds were heard during the months of November through February. During March,
dissections were made daily for 5 days because larval development and movement were
so rapid. During dissections, the fruits were first cut in half and then into quarters.
Then a cut was made to separate the albedo from the pulp. After each cut, the fruit
portion was examined carefully with a 2X magnifier/lamp or with a 10X dissecting
microscope for feeding trails or larvae. The flavedo and the albedo were carefully teased
apart with forceps and probes. Neonate larvae are very small and identical in color to
the albedo layer in grapefruit which makes them difficult to see. As they become larger,
they and their feeding trails were more easily found. An example of a feeding trail is
shown in Figure 2.
Data were recorded on a diagram of the fruit indicating where each larva was lo-
cated. The distance that the larva moved from the outer perimeter of the fruit was
measured with a direct line from the outer edge to the center. No attempt was made
to determine lateral distances the larvae may have moved while feeding because of the
great variations found and the difficulty in following the complete feeding route.


The average location of larvae and the number of days after acoustic detection for
each month are shown in Figures 3 and 4 for each month from November through
March. The lines on either side of the dot is the range of depths that larvae had pene-

Florida Entomologist 71(4)

Fig. 2. An example of a feeding trail in the albedo of a grapefruit made by a Carib-
bean fruit fly larva.

treated by that day. Because growth and movement of larvae in November and De-
cember were slow, the figure only indicated locations at 4-day intervals.
Fruits picked in November were just beginning to turn yellow. The peel was still
firm to the touch. In all cases when eggs were discovered, always in the flavedo, they
were found to be laid singly rather than in aggregates. During the first 4 days after
larval chewing was detected, the larvae were only found in the flavedo. By day 8, larvae
were found in the albedo and in the outer regions of the pulp. On day 12, most of the
larvae found were in the pulp with a few still in the albedo. On day 14, one mature larva
left the fruit to pupate. By day 16, all of the larvae found were in the pulp. This was
the last observation made because all of the infested fruit from the November picking
date had been cut for examination. The number of larvae found on the days of examina-
tion ranged from 11 to 33, more larvae were found in the later examination periods as
might be expected because they were larger in size and easier to see.
In December, larvae were recorded only in the flavedo on day zero. On day 4, they
were found in both the flavedo and the albedo. On day 8, they were found in all 3 regions
and by day 12 and 16 were found exclusively in the pulp (Figure 3). The first larva
emerged from the fruit on day 13.
From January through March, fruits were collected every 2 weeks so that 60 fruits
could be infested each week. They were examined every two days after larval sounds
were detected. Five fruits per day were cut and examined. During the first 2 days after
egg hatch, larvae were found exclusively in the flavedo. On day 4, most of the larvae
were found in the albedo with a few still in the flavedo and 2 larvae were found at the
border between the pulp and the albedo. By day 6, although most of the larvae were
still in the albedo a few larvae had penetrated the pulp, one as deep as 26 mm. On day
8, most of the larvae were found in the pulp with a few still found at the border between
the pulp and the albedo. The number of larvae found per fruit ranged from 14 to 44.


December, 1988

Calkins & Webb: Symposium on Agroacoustics


Day 0: first day chewing sound


Day 0: first day chewing sound

Fig. 3. The mean location of Caribbean fruit fly larvae in grapefruit picked during
November and December at specified days after larval eclosion.

From day 10 on, almost all of the larvae were found in the pulp (Figure 4). The first
larvae to emerge during the first week's infestation was on day 13. Later in the month,
the first emergence of mature larvae occurred on days 12, 10 & 10 for the next 3 weeks,
respectively. The number of larvae found ranged from 13 to 55/fruit.
In February, larvae were already found in the albedo on day 2 and one larvae was
found in the pulp at the interface between the pulp and the albedo. By day 4, a few


Florida Entomologist 71(4)

larvae were moving into the pulp, one as deep as 5 mm while most were at the interface.
By day 6, most of the larvae were in the pulp as deep as 26 mm (Figure 4). Most of the
larvae had completed development and had exited the fruit before the 14th day. Unfor-
tunately, the date of first emergence was not recorded.
During March, fruits were examined every day for 5 days. On day zero, the first
day larval sounds were detected, all of the newly hatched larvae were still in the
flavedo. On day 1, several had already migrated into the albedo. By day 2, most were
found in the albedo. By day 3, although most were in the albedo, a few had already
entered the pulp as deep as 2 mm. On days 4 and 5, the larvae were found both in the
albedo and the pulp. The first larva began exiting the fruit on day 6 (Figure 4). Feeding


Day 0: first day chewing sound detected


Day 0: first day chewing sound detected

December, 1988

Calkins & Webb: Symposium on Agroacoustics


and growth were very rapid in fruit picked in March. The number of larvae found
ranged from 16 to 81/fruit. The feeding passages were difficult to follow because there
was such a large number of larvae in each fruit and the feeding was so extensive. The
fruit also became infested with fungi and mold and broke down very quickly, partly as
a consequence of the extensive larval feeding.


Results of this study indicate that the maturity of the fruit has a great influence on
the development rate and movement of larvae within the fruit. The rate of egg hatch
in November took from 6 to 12 days to hatch. In February, the egg hatch occurred
within 4.6 days. An interesting side observation of this study was that about 50% of
the eggs were found in the oil glands and about 50% between the oil glands in fruit from
all picking dates. Greany et al. (1983) observed a high mortality of eggs and newly-
hatched larvae that were inside of oil glands. The increase in development rate and
movement of larvae into the pulp occurred earlier in consecutive months from November
through March. The greatest increase occurred in March when complete larval develop-
ment occurred as early as 5 days after egg hatch. During this month the fruit has
reached a stage of maturity when grapefruit is truly susceptible to attack from the
Caribfly. Feeding sounds of larvae are most easily detected by acoustical techniques
when fruit is most mature because larvae feed almost continually at this stage (Webb,
unpublished data). There are also several changes in the chemistry of the peel with
maturity and, these appear to affect feeding and development rates early in the season.
(Shaw & Calkins, unpublished).


Day 0: first day chewing sound detected

Fig. 4. The mean location of Caribbean fruit fly larvae in grapefruit picked during
January, February and March at specified days after larval eclosion.

Florida Entomologist 71(4)

December, 1988


We thank the efforts of Ronald Thalman, Lloyd Davis and Shuichi Masuda for mak-
ing this study possible, and Richard Guy for the photography.


1983. Biochemical resistance of citrus to fruit flies. Demonstration and elucida-
tion of resistance to the Caribbean fruit fly, Anastrepha suspense. Entomol.
Exp. & Appl. 34: 40-50.
KEFFORD, J. F., AND B. V. CHANDLER. 1970. The chemical constituents of citrus
fruits. Academic Press, New York. 246 pp.
SWANSON, R. W., AND R. M. BARANOWSKI. 1972. Host range and infestation by
the Caribbean fruit fly, Anastrepha suspense (Diptera: Tephritidae), in South
Florida. Proc. Florida State Hort. Soc. 85: 271-274.
WEBB, J. C., AND P. J. LANDOLT. 1984. Detecting insect larvae in fruit by vibrations
produced. J. Environ. Sci. Health A19: 367-375.
WEBB, J. C., C. A. LITZKOW, AND D. L. SLAUGHTER. 1988. A computerized acous-
tical larval detection system. Appl. Eng. in Agric. 4: 268-274.

-^- -C- -*-- *-L --*--- **--- --^>


National Center for Physical Acoustics
P.O. Box 847
University, MS 38677


Accurate estimates of population size are needed to understand the population
dynamics of any species. They are also needed to determine when to implement a
specific control tactic, and to measure whether that control tactic has been effective.
This paper discusses the use of acoustic signals produced by insects and the feasibility
of using these signals to census populations.
Insect sounds are either incidental (produced as a by-product of some activity) or
non-incidental (produced to cause a response in some other animal). Incidental sounds
differ from non-incidental sounds with respect to several features that are important to
using sound to census populations. These features include species specificity, frequency
content, ease of localization, distance traveled, and the duration and timing of sound
Studies of crickets show that information about which individuals in a population are
producing sound, when the individuals produce sound (seasonally and daily), and the
probability that individuals produce sound during census periods must be known to
accurately estimate the size of a population.


Se necesitan estimados precisos del tamafo de la poblaci6n para entender el di-
namismo de la poblaci6n de cualquier especie. Tambi6n se necesitan para determinar


Forrest: Symposium on Agroacoustics 417

cuindo es necesario implementar una tactica especifica de control, y para medir la
efectividad de dicha tActica. Se discute el uso de sefiales acisticas producidas por insec-
tos y la posibilidad de usarlas en censos de poblaciones.
Los sonidos producidos por insects son incidentales (producidos como product sec-
undario de otra actividad) o no incidentales (producidos para inducir una respuesta en
otro animal) Los sonidos incidentales difieren de los no incidentales con respect a
algunas caracteristicas que son importantes en el uso del sonido en el censo de pob-
laciones. Estas caracteristicas incluyen especificidad de las species, el contenido de la
frecuencia, la facilidad en localizarla, distancia cubierta, y la duraci6n y lo oportuno del
sonido producido.
Estudios hechos con grills muestran oue se debe de tener informaci6n sobre aquellos
individuos que produce sonidos en la poblaci6n, cuando produce los sonidos
(diariamente y estacionalmente), y la probabilidad que los individuos produzcan sonidos
durante el censo para poder estimar con exactitud el tamafio de la poblaci6n.

An often asked question of animal populations is 'Do individuals occur in a certain
area?'. While knowing if they occur at a locality is necessary to study them, an even
more important question is 'How many animals are there?' or 'What is the number of
individuals in a certain area?'. One of the most important properties of any population
is its size. Population size or density is information that is needed to understand the
dynamics of a population, and it has import in all areas concerned with ecological mod-
eling of populations.
Knowing the dynamics of pest populations is crucial in determining effective control
tactics, when to initiate the tactics, and if the tactics, once implemented, are successful.
Knowing how a population changes through time must be considered paramount in the
management of a pest, and therefore, accurate and precise estimates of pest population
density are needed.
For more than 20 years researchers have used sounds produced by insects to detect
their presence (Adams et al. 1953, Wojcik 1968; see also Webb, Calkins, Wolfenbarger
et al., Toba, Vick et al. this symposium). In contrast, few attempts have been made to
use insect generated sound to estimate population size and density.
This paper will discuss the use of sounds produced by insects as a means of estimat-
ing their population size and in monitoring the population size over time.


In this paper the term "sound" is used in a broad sense: a vibration in some medium.
By using this definition, all insects produce sounds during all stages of their lives. These
sounds can be classified into two broad categories (Table 1). The first category contains
sounds that are by-products of some activity of the animal. They are termed incidental
sounds. In the second category are sounds that function to produce a particular response
in another animal. These non-incidental sounds are termed communication. In most
instances communication occurs between members of the same species, but interspecific
communication also occurs, for instance in warning and anti-predator signals. Communi-
cation, as it is used here, includes deceptive or false signals (see Burk this symposium).
An insect's activity can be monitored by listening for either incidental or non-incidental
sounds, and therefore the sounds may be used to census insect populations.
Examples of the two categories of sounds produced by insects are shown in Table
1. The two categories differ in one fundamental aspect. Whereas, sounds used in com-
munication are produced to be heard by another animal, incidental sounds are not.

418 Florida Entomologist 71(4) December, 1988


walking mating signals
flying (adults) courtship signals
chewing territorial displays
swimming social & subsocial signals
breathing anti-predator signals
heartbeats warning signals


Whenever an insect produces sound, that sound is a potential cue that predators and
parasites might use in locating prey or hosts (Burk 1982 and refs.). Selection will favor
the production of sounds that avoid the attention of predators and parasites. Because
communication signals and incidental sounds differ in the function for which they are
produced, it might be expected that differences in certain characteristics will be re-
flected between the categories. Understanding these differences will be important if we
are to use insect sounds to monitor and census individuals within a population (Table 2).

Species Specificity: One of the major differences to be expected between the two
categories is whether the sound is specific to a particular species. Sounds produced as
by-products of certain activities should not be as species specific as those used in com-
munication, especially when the communication mediates sexual pair formation.
Specificity is important in whether sounds are suitable to monitor a particular species.
When more than one species in an area produce the same or similar sounds it will be
difficult to estimate population size for the species of interest. Likewise, it will be
difficult to accurately estimate a population size if different members of same population
produce different sounds.

Frequency Content: Generally incidental sounds have a broader frequency range than
sounds used in communication, and a spectral 'signature' of the sound produced by an
insect may be important in its detection and identification. However, the spectrum
(relative power at different frequencies) of a sound changes with distance from the
source, and is influenced by the particular habitat over which the sound propagates
(Marten & Marler 1977, Wiley & Richards 1978). Thus, sounds with a narrow frequency
range (non-incidental or communication signals) will be more convenient to use to locate
and count individuals in a population.


Species Specificity low high
Frequency Range broad narrow
Localization difficult easy
Distance Traveled short long
Duration short long
Timing unpredictable predictable

Forrest: Symposium on Agroacoustics 419

Ease of Localization, Distance Traveled and Duration Produced: Because sound may
attract the attention of predators and parasites, selection should favor incidental sounds
that are not easy to localize, do not travel far, and are produced for short durations
and/or at unpredictable times. While sounds used in communication can also be used by
predators and parasites, they have evolved to be heard by another animal. Therefore
they should be easily localized and produced at a level and for a duration that will allow
the intended receiver to detect and locate the sender from some distance. Being able
to localize a sound source from a distance is an important consideration when using
sound produced by insects to census their populations.
Device Design: The ease and success with which sound can be used in monitoring and
estimating populations will depend upon the above characteristics and understanding
the contexts in which the sounds are produced. The differences in the two categories
of sound will also influence the design of listening devices. Development of listening
devices for specific incidental sounds will be difficult and will require in-depth analysis
of the sounds (Webb et al. 1988). One potential advantage to sounds that are produced
for communication is that an efficient and effective device for listening to the particular
sound in the specific environment in which it is produced has already been developed.
Natural selection has shaped and modified the ears of the receiving animals to be effec-
tive in detecting the sounds. Investigating the properties of the ears of receiving animals
may help develop transducers with similar properties. Perhaps the animals' own ear
can be used as a 'biological microphone' or transducer (Rheinlaender & Romer 1986,
Romer & Bailey 1986).


Sexual and Life-Stage Differences in Sound Production: To accurately census popula-
tions it is necessary to know what individuals in the population are producing sound
and how these relate numerically to the rest of the population. In insects that communi-
cate via sound usually only one sex produces the sound. For crickets it is usually the
adult male, and to estimate a population size by counting calling males one must know
the proportion of adults in the population and the sex ratio of the adults.
Another important consideration for sound censusing is whether different individuals
of a population produce different sounds. For example, the characteristics of the sounds
produced by insects often depend upon the size of the structures that produce the
sound. During the growth of an insect the size of the structures change, and thus, the
characteristics of the sound change with the life stage of the insect.
Different life-stages of an insect often occur at the same time of the year. If only
one stage produces a particular sound, then information about the life history and the
proportion of individuals in each age class must be considered. When certain sounds are
characteristic of each stage then the sounds can be used to estimate the number of
individuals in each life stage class and will provide information about the distribution
of age classes of a population.
Life table data for a hypothetical cricket population, Gryllus hypotheticus are shown
in Table 3. This species has six life stages from egg to adult. The individuals in the
population are distributed amongst the different age classes as shown in column p,.
During a census period 100 calling males are counted. Counting calling males gives a
minimum size of the population, however this is far from a more accurate estimate
(N = 166) calculated using life table statistics of the population. The number of individu-
als in each age class, n,, can be estimated using the following equation:

nx = (N-p,) / p Eq. 1

where n, is the estimate of the number of individuals in the x age class, N (= 100) is

Florida Entomologist 71(4)


Life Prop. of Number
Stage Population Counted Estimatel
x Px N nx

1 0.00 0
2 0.05 8
3 0.05 8
4 0.10 17
5 0.20 33
adult 0.60 100 100
TOTAL 1.00 166

'The estimate for each age class, n,,is calculated from Eq. 1 (see text).

the number of individuals counted producing sound, px is the proportion of the popula-
tion in the x age class, and pc (=0.60) is the proportion of individuals in the counted
age class. This assumes that all individuals in the age class being counted are producing
sound during the census. To make the equation more general the number counted must
be divided by the proportion of individuals of the age clas that are calling at the time
of the census, rc. The equation then becomes

nx = (N-px) / (pere). Eq. 2

For instance, if all adult males are calling during the census and the adult population
has a sex ratio of 50:50 (ie. r, = 0.50), then the estimates of each class increase by a
factor of 2. The total population is 332. As will be seen below, all males are not always
calling during a census, and it becomes necessary to find out what proportion of the
individuals are producing sound during a particular time to estimate the population.

Daily Periods of Sound Production: One of the problems with using sound to census
populations is the lack of information about when to monitor. If reliable estimates of
populations are to be made using the sound produced by insects, then all individuals in
the population, or a constant proportion of those individuals, must produce sound at the
time of censusing. Almost all activities of animals, especially those of insects, follow a
circadian rhythm (Brady 1982). Much has been learned about physiological and environ-
mental influences on these animal rhythms in the laboratory, but only recently have
researchers looked at such rhythms with the animal's ecology in mind (Walker 1983).
If insect sounds are to be used in estimation of population size, understanding these
rhythms becomes necessary.

Species Differences in Rhythms: The periods of sound production are often very specific
to a species and may be very different between closely related species. For a particular
species these activities can be brief, lasting only a few minutes, or they may be spread
throughout the day. For instance, each evening male mole crickets, Scapteriscus acletus
and S. vicinus, begin calling shortly after sunset and continue to call for about an hour
(Forrest 1983). This period also corresponds to female flight activity. Similarly, male
Anurogryllus arboreus, a short-tailed cricket, call for about an hour shortly after sunset
(Walker 1980a).
Males of other crickets call throughout the night and these species differ in the
proportions of individuals calling at any one time. During an evening, about 25% of the
Anurogryllus muticus in the population call at the same time (Walker & Whitesell 1982,


December, 1988

Forrest: Symposium on Agroacoustics 421

Walker 1983). About 50% of the total population of Gryllodes supplicans call simultane-
ously during an evening (Sakaluk 1987). The proportion of male field crickets (Gryllus
integer, G. veletis, and G. pennsylvanicus) that call during an evening varies throughout
the night and is dependent upon population density. Just after sunset 25-80% of the
male population is calling, and there is an increase in calling activity at sunrise when
almost 100% of the males are calling (Cade 1979, French & Cade 1987).
These daily calling periods may be changed by several environmental factors. For
instance, calling may shift from predominantly nighttime to mostly daytime calling if
nighttime temperatures are below those suitable for calling (Alexander & Meral 1967).
Rain may also cause shifts in the calling period (Alexander & Meral 1967, Forrest 1983,
Walker 1983, Walker & Whitesell 1982). Interaction with other species has also been
found to cause shifts in the calling period of some katydids (Greenfield 1988, Latimer
& Broughton 1984).
Characterizing the periods of sound production is useful in determining the proper
times to census an insect population. Once this characterization has been made it be-
comes necessary to understand what the individuals in that population are doing during
the periods of sound production. However, few data are available on sound production
by individual insects under natural conditions.

Individual Differences in Rhythms Within Days: Calling periods sometimes vary be-
tween individuals within a population. Individual male mole crickets, Scapteriscus ac-
letus and S. vicinus, have significantly different times that they begin their evening
calling (Forrest 1983). Walker & Whitesell (1982) studied calling of individual male A.
muticus. They found males that called near burrows generally called during the same
period and for the same duration from night to night, but individuals differed in both
respects. Some males were early evening callers while others called late in the evening.
Other males that were not associated with burrows were more variable in their calling
times and durations, often moving between census periods. One male moved more than
50 m during a night of singing.

Individuals Differences in Rhythms Between Days: Another important aspect of sound
production is the variability in sound production from one day to the next. To reliably
quantify population density using the sounds that are made by individuals in the popu-
lation, the variation in sound production by individuals must be known. The probability
of calling from night to night varied among individual male mole crickets monitored for
periods of as long as a month (Fig. 1). One S. vicinus male called only 26% of the nights
compared with other males that called as much as 79% of the nights during the same
period. For all males combined, S. acletus males called an average of 78% of the suitable
nights (>16C and no rain) and S. vicinus males averaged 76%. All males were kept
outdoors in 19-liter buckets of soil and were provided with enough food so as not to
limit energy needed for calling (see below).

Effects of Density and Nutrition on Sound Production: If sounds are to be used to
measure population size and density, the influence of population size or density on the
production of sound must be known. This too will depend upon the context in which the
sound is produced. For incidental sounds, the number of sounds detected should increase
linearly with increasing density. In other contexts (eg. aggressive sounds), the propor-
tion of the population producing sounds may be low under low density situations and
may then increase dramatically (exponentially) as the population density increases. The
reverse may be true for sounds such as the calling songs of crickets. In this case, the
proportion of the population calling may decrease as the density increases because of
an increase in attacks from neighbors (see Burk 1983 and this symposium), or because
the sounds are energetically expensive to produce and the high density situation makes
another strategy of finding mates more profitable.

Florida Entomologist 71(4)


S. oc/e tus

. .. 0.79
...... 0.68
... .. 0.63




S. vicinus




0 5 10 15 20 25 30 35
Fig. 1. Daily calling probabilities of individual male Scapteriscus acletus and S.
vicinus. Each horizontal line represents data from a single male (N = 16 S. acletus =21
S. vicinus). Solid lines are days males called, dotted lines represent days males did not
call, and open area show days males were not monitored. Males are grouped according
to dates monitored. Top 4 S. acletus and 8 S. vicinus were monitored 10-22 Apr 1979,
middle 8 S. acletus and 5 S. vicinus were monitored 18 May-18 Jun 1979, and bottom
4 S. acletus and 8 S. vicinus were monitored 7 Mar-9 Apr 1980. Numbers at the right
are the proportion of suitable (>160C and no rain) nights that each individual called.


I'''' '; ''

December, 1988

Forrest: Symposium on Agroacoustics

Data for calling in two mole crickets as a function of male density (Shaw 1981,
personal communication) are shown in Figure 2. Male mole crickets were place at vari-
ous densities in 10 m2 outdoor arenas. The number of individuals calling in the arenas
was monitored on successive nights. The maximum percentage of males calling de-
creases exponentially with increasing density and reaches a constant level of about 10%
calling at high densities (Fig. 2). These data can be fit with an exponential decay function
of the form

P(x) = a / (l-e('-x),

Eq. 3

where P(x) is the proportion of the total calling at density x, a is the asymptotic value
of the function as x becomes large, and T is the function's rate of decay. The dotted line
in Figure 2 is the least-squares fit to all data using the above equation (a = 0.11 and
7 = 2.00). Walker (1980b, personal communication) has shown that the presence of
conspecific male and female mole crickets will decrease the proportion of nights indi-
vidual males call by as much as 70%.
Nutrition will also influence sound production. If the sounds to be monitored are
produced because of feeding activity, then abundant food supplies could cause a decrease
in movement associated with acquiring food. Sound production caused by the movement
of the insect would decrease. However, if the food has little nutritional value, then the
individual must consume more and feed more often to make up for the low nutritional
intake (Slansky 982). Sound production would increase. If the sounds are energetically
expensive to produce, poor nutrition may decrease the production of such sounds. Cal-







Fig. 2. Maximum percent of male mole crickets calling in 10 m2 outdoor arenas
plotted as a function of male density in the arena. Dotted line is a least-square fit to
the data using a = 0.11 and T = 2.00 as parameters of Eq. 3 (see text, Shaw 1981,
personal communication).


0 S. vicinus
I U S. ocletus


S "", ...... i l~ ....... ..................



424 Florida Entomologist 71(4) December, 1988

ling male crickets use more than 10 times the energy used during resting (Prestwich &
Walker 1981, Kavanagh 1987). Walker (1980b, personal communication) found that the
proportion of nights that male mole crickets called decreased from about 80 to 30% when
they were deprived of food (Fig. 3).

Seasonal Periods of Sound Production: Besides daily periods of sound production, ani-
mals very often have specific seasonal periods. This is especially true of insects where
certain life stages are only present during specific times of the year. Crickets are a good
example. It has become common practice to use counts of calling crickets to determine
life cycles, seasonal maturation, and adult activity periods of crickets (Alexander 1962,
Alexander & Meral 1967, Walker 1983). Because adult males are the only part of the
population that produce calling songs, the season of sound production generally corres-
ponds to the adult activity of the species. However, in some species males are rarely
heard at certain times of the year even though adult males are abundant in the popula-
tion. This may be caused by the abundance of acoustically orienting parasitoid flies
present at this time of the year (Burk 1982 and refs.). The seasonal activity of sound
production must be considered if acoustic signals are to be used to monitor populations.

Geographic Ranges and Population Spread: Two very practical uses of monitoring insect
populations based upon the sounds produced are determining geographic ranges and
measuring the spread of populations. The use of acoustic signals to determine population
ranges and their spread have been used extensively in insects that use sound to com-

100 F FED


< 60 -

c= 40



S. ocletus S. vicinus

Fig. 3. Bars show how starving influences the nightly calling of male mole crickets.
Percent of nights called is significantly decreased for males that are deprived of food
(hatched bars) compared with males provided food (open bars) (Walker 1980b, personal


Forrest: Symposium on Agroacoustics 425

municate over long distances (Alexander 1962). The signals of particular species are
easily distinguished, and to determine a species geographic range it is a simple matter
of cataloguing the localities where that species sound has been heard. This technique
could also be used for incidental sounds, provided a suitable detection device can be
produced. This application will increase our ability to monitor the spread of migratory
species and pest species that have been introduced to new areas.


Monitoring and estimating populations is important in the management of any pest
species. Because sound of one form or another is produced by all insects, a promising
area of research is to use sound in detecting, censusing, and monitoring insect popu-
lations. An understanding of the sounds, knowing the contexts in which they are pro-
duced, and knowing how the production of sound relates to the ecology and biology of
the animals will improve population estimation. In most instances, the population size
will be underestimated. It is only when insects of different species produce indistin-
guishable sounds or when individuals are very mobile and are counted more than once
that a population would be overestimated. Thus, the use of sound censusing provides
baseline data for estimating population size. Where exact counts are not a necessity,
monitoring the sounds of insects can provide much information about a species life
history. Using sound should become an increasingly important tactic for monitoring
population spread and has application in monitoring the movement of migratory pest


I wish to thank Dr. K. C. Shaw and Dr. T. J. Walker for allowing me to use
unpublished data shown in Figures 2 and 3. I want to thank Dr. Daniel Wojcik for
providing me with many helpful references. Drs. C. 0. Calkins, J. Sivinski, T. J.
Walker, J. C. Webb and an anonymous reviewer offered many suggestions that im-
proved the manuscript. I also thank Dr. Thomas Walden for inviting me to be a part
of this symposium.

detection of grain infested internally with insects. Science 118: 163-164.
ALEXANDER, R. D. 1962. The role of behavioral study in cricket classification. Syst.
Zool. 11: 53-72.
ALEXANDER, R. D., AND G. H. MERAL. 1967. Seasonal and daily chirping cycles in
the northern spring and fall field crickets, Gryllus veletis and G. pennsylvanicus.
Ohio J. Sci. 67: 200-209.
BRADY, J. (ed.). 1982. Biological timekeeping. Cambridge University, Cambridge.
BURK, T. 1982. Evolutionary significance of predation on sexually signalling males.
Florida Entomol. 65: 90-104.
BURK, T. 1983. Male aggression and female choice in a field cricket (Teleogryllus
oceanicus). Pages 97-119 in D. T. Gwynne and G. K. Morris, eds. Orthopteran
mating systems: sexual competition in a diverse group of insects. Westview
Press, Boulder, Colorado.
CADE, W. 1979. The evolution of alternative male reproductive strategies in field
crickets. Pages 343-379 in M. S. Blum and N. A. Blum, eds. Sexual selection and
reproductive competition in insects. Academic Press, New York.
FORREST, T. G. 1983. Calling songs and mate choice in mole crickets. Pages 185-204
in D. T. Gwynne and G. K. Morris, eds. Orthopteran mating systems: sexual
competition in a diverse group of insects. Westview Press, Boulder, Colorado.

426 Florida Entomologist 71(4) December, 1988

FRENCH, B. W., AND W. H. CADE. 1987. The timing of calling, movement, and
mating in the field crickets Gryllus veletis, G. pennsylvanicus, and G. integer.
Behav. Ecol. Sociobiol. 21: 157-162.
GREENFIELD, M. C. 1988. Interspecific acoustic interactions among katydids
Neoconocephalus: inhibition-induced shifts in diel periodicity. Anim. Behav. 36:
KAVANAGH, M. W. 1987. The efficiency of sound production in two cricket species,
Gryllotalpa australus and Teleogryllus commodus (Orthoptera: Grylloidea). J.
Exp. Biol. 130: 107-119.
LATIMER, W., AND W. B. BROUGHTON. 1984. Acoustic interference in bush crickets;
a factor in the evolution of singing in insects? J. Nat. Hist. 18: 599-616.
MARTEN, K., AND P. MARLER. 1977. Sound transmission and its significance for
animal vocalization: I. temperate habitats. Behav. Ecol. Sociobiol. 2: 271-290.
PRESTWICH, K. N., AND T. J. WALKER. 1981. Energetics of singing in crickets:
effect of temperature in three trilling species (Orthoptera: Gryllidae). J. Comp.
Physiol. 143: 199-212.
RHEINLAENDER, J., AND H. ROMER. 1986. Insect hearing in the field: I. the use of
identified nerve cells as 'biological microphones'. J. Comp. Physiol. 158: 647-652.
ROMER, H., AND W. J. BAILEY. 1986. Insect hearing in the field: II. male spacing
behaviour and correlated acoustic cues in the bushcricket Mygalopsis marki. J.
Comp. Physiol. 159: 627-638.
SAKALUK, S. K. 1987. Reproductive behaviour of the decorated cricket, Gryllodes
supplicans (Orthoptera: Gryllidae): calling schedules, spatial distribution, and
mating. Behaviour 100: 202-225.
SHAW, K. C. 1981. Use of calling songs to census mole crickets. Annu. Rept. Mole
Cricket Res. 80-8: 75-78 [Documents, Univ. of Florida Libraries].
SLANSKY, F. 1982. Insect nutrition: an adaptationist's perspective. Florida Entomol.
65: 45-71.
WALKER, T. J. 1980a. Reproductive behavior and mating success of male short-tailed
crickets: differences within and between demes. Evol. Biol. 13: 219-260.
WALKER, T. J. 1980b. Estimating adult populations: song census. Annu. Rept. Mole
Cricket Res. 79-80: 55-58 [Documents, Univ. of Florida Libraries].
WALKER, T. J. 1983. Diel patterns of calling in nocturnal Orthoptera. Pages 45-72 in
D. T. Gwynne and G. K. Morris, eds. Orthopteran mating systems: sexual com-
petition in a diverse group of insects. Westview Press, Boulder, Colorado.
WALKER, T. J., AND J. J. WHITESELL. 1982. Singing schedules and sites for a
tropical burrowing cricket (Anurogryllus muticus). Biotropica 14: 220-227.
WEBB, J. C., D. C. SLAUGHTER, AND C. A. LITZKOW. 1988. Acoustical system to
detect larvae in infested commodities. Florida Entomol. 71: 492-504.
WILEY, R. H., AND D. G. RICHARDS. 1978. Physical constraints on acoustic com-
munication in the atmosphere: implications for the evolution of animal vocaliza-
tions. Behav. Ecol. Sociobiol. 3: 69-94.
WOJCIK, D. P. 1968. Tests for audible and ultrasonic sound production by stored-prod-
uct insects. J. Econ. Entomol. 61: 1414-1417.

Haack et al.: Symposium on Agroacoustics



'North Central Forest Experiment Station, USDA Forest Service,
1407 S. Harrison Road, Room 220,
East Lansing, MI 48823
2Department of Metallurgy, Mechanics, and Materials Science,
College of Engineering. Michigan State University,
East Lansing, MI 48824


Ultrasonic acoustical emissions (AEs) were recorded from trunk samples of eastern
white pine, Pinus strobus L., northern red oak, Quercus rubra L., paper birch, Betula
papyrifera Marsh., and red maple, Acer rubrum L., that were cut in winter (February)
and allowed to dry indoors. Emission rates were determined for waxed and unwaxed
samples at weekly intervals throughout the drying period. Waveform parameters deter-
mined for each AE were duration, counts (oscillations above threshold), rise time,
amplitude, energy, and average frequency (counts/duration). For all tree species, AE
rates from unwaxed samples first increased and then decreased during drying, and they
were several times greater than rates for waxed samples. For white pine, mean dura-
tion, counts, rise time, amplitude, and energy decreased over time while average fre-
quency increased. Such patterns were often lacking among the three hardwood species.
Average waveform parameters differed among species; AEs from white pine were
strongest while those from red oak were weakest. Implications for bark- and wood-bor-
ing insects are discussed.


Emisiones acisticas ultras6nicas (EAU) se registraron en muestras de troncos de
pinos blancos del este, Pinus strobus L., robles rojos del norte, Quercus rubra L.,
abedul de papel, Betula papyrifera Marsh., y en el meple rojo, Acer rubrum L., que
fueron cortados en el invierno (Febrero) y secados dentro de la casa. Se determine la
proporci6n de emisi6n de muestras enceradas y sin encerar a intervalos semanales
durante el period que se secaban. Los parametros determinados del tipo de onda para
cada EAU fueron duraci6n, conteo, (oscilaci6n encima del umbral), tiempo en ascender,
amplitud, energia, y el promedio de la frecuencia (conteo/ duraci6n) Para todas las
species de arboles, la proporci6n de EAU de muestras sin cera primero aument6 y
despues disminuy6 cuando se secaban y fueron varias veces mayor que la proporci6n de
muestras enceradas. Para los pinos blancos, el intermedio de duraci6n, conteo, tiempo
en ascender, amplitud, y energia, disminuy6 con el tiempo mientras que el promedio de
la frecuencia aument6. Tales patrons amenudo estaban ausentes entire las tres species
de madera dura. El promedio de los parametros del tipo de onda diferi6 entire las
species; EAU del pino blanco fueron los mas fuertes mientras que aquellos del roble
rojo fueron los mas debiles. Se discuten las implicaciones para insects taladradores de
la corteza y de la madera.

Water is conducted upwards in the xylem tissue of plants. To accomplish this, xylem
functions at negative pressures, with water being held under tremendous tension

428 Florida Entomologist 71(4) December, 1988

(Kramer 1983, Oertli 1971, Zimmermann 1983). This tension increases dramatically
during drought and at some critical point the hydrogen bonds between the water
molecules break or cavitate (Tyree & Dixon 1983). The breaking of individual water
columns releases energy that results in acoustical emissions (AEs) primarily at ul-
trasonic frequencies (>20 kHz) (Sandford & Grace 1985, Tyree & Dixon 1983, Tyree et
al. 1984b). Each AE is believed to result from cavitation of an individual water column
within the conducting xylem. As drought stress intensifies, AE rate increases with the
largest diameter xylem conduits tending to cavitate first (Sandford & Grace 1985, Tyree
& Dixon 1986). The term conduit is used to include both the single-cell tracheids of
conifers and the multi-cell vessels of hardwoods. Ultrasonic AEs from drought-stressed
plants are a recent discovery, being first reported by Tyree & Dixon in 1983. AEs have
since been recorded from a number of hardwood tree species (Jones & Pefia 1986, Salleo
& Lo Gullo 1986, Sandford & Grace 1985, Tyree & Dixon 1986), conifers (Dixon et al.
1984, Pefia & Grace 1986, Sandford & Grace 1985, Tyree and Dixon 1983, 1986, Tyree
et al. 1984a, 1984b), and herbaceous plants such as corn (Tyree et al. 1986). In the forest
products industry, AEs have been reported to occur in drying lumber (Becker 1982,
Honeycutt et al. 1985, Noguchi et al. 1980, 1983, 1985, Skaar et al. 1980).
Considering that many bark- and wood-boring insects preferentially attack drought-
stressed trees (Mattson & Haack 1987), and that several insects from at least nine
Orders can detect and/or produce ultrasound at similar frequencies (Lewis & Gower
1980, Sales & Pye 1974, Schwartzkopf 1974, Spangler 1988), Mattson & Haack (1987)
hypothesized that some bark- and wood-boring species might perceive and utilize
drought-induced acoustic signals during host colonization. One means to test this
hypothesis requires characterization of the AE rate and waveform pattern from
drought-stressed trees during the period of host colonization, and then testing the
relative attractiveness of substrates pulsed with ultrasound to simulate drought-related
AEs with that of similar nonpulsed substrates.
AEs are sensed by transducers affixed to the bark or wood (xylem) surface. Most
researchers have used wideband transducers that are sensitive over the range 100-1000
kHz. The most commonly reported AE parameter has been emission rate. Besides
simply counting AEs, technologies exist to determine several AE waveform parameters
such as duration, counts, rise time, amplitude, and energy (see Fig. 1). Because (a)
larger xylem conduits tend to cavitate first, (b) frequency is inversely proportional to
conduit size, and (c) AE intensity is proportional to cell size, average frequency theoret-
ically would increase while signal intensity would decrease during an extended period
of drought. Moreover, because xylem cells vary among species in length, width, struc-
ture, and arrangement (Panshin & de Zeeuw 1980), it is possible that different species
or genera of trees will produce unique AE signatures. In the present study, our objec-
tive was to describe emission rates and several AE waveform parameters during an
extended period of dehydration for freshly cut trunk samples of eastern white pine,
Pinus strobus L. northern red oak, Quercus rubra L., paper birch, Betula papyrifera
Marsh., and red maple, Acer rubrum L. Such information will allow us to pulse sub-
strates with ultrasound at meaningful levels.
In conifer xylem, the principal water-conducting units are single-celled tracheids.
Conifers lack vessels. Hardwood xylem is more complex; vessels are the chief water-con-
ducting components. Vessels are tubular structures, consisting of individual cells from
which the end walls have disintegrated. Vessels are much larger than tracheids. The
arrangement of vessels is variable among species, but fixed for a given species.
Hardwoods are often classified as either ring porous or diffuse porous. In ring-porous
species, earlywood vessels are much larger than latewood vessels, thus producing an
abrupt and obvious transition between growth rings. In diffuse-porous species, there
is little change in vessel size across the growth ring (Kramer & Kozlowski 1979). In the

Haack et al.: Symposium on Agroacoustics





Fig. 1. Typical acoustical emission waveform showing parameters commonly meas-
ured above a preset threshold voltage: event duration (time between first and last
threshold crossing), rise time (time from first threshold crossing to peak amplitude),
peak amplitude (the peak voltage of the largest excursion attained by the signal
waveform from an emission event), counts (number of times the signal passes above the
threshold), and energy (total elastic energy released by an emission event; this parame-
ter is calculated in various ways using the values for amplitude and duration).

present study, white pine is a conifer, red oak is a ring-porous hardwood, and paper
birch and red maple are diffuse-porous hardwoods.


Plant Material and Preparation

A small-diameter (6-10 cm) tree of each of the four test species was cut on February
22, 1988 in a forested area in southwestern Michigan. Six, branch-free, 40-cm long
samples were cut from the trunk of each tree and immediately taken to the laboratory
and weighed (Table 1). For each species, two samples were dried to determine initial
moisture content (wet-weight basis), two were end-dipped in melted paraffin to slow
dehydration, and two were left unwaxed. The waxed and unwaxed samples were used
to simulate low and high levels of water stress, respectively. The samples were allowed
to air-dry in the laboratory at room temperature; weight was recorded periodically
during the drying period. At the end of the study, all samples were dried at 75C and
To aid in transducer placement, we removed a small area (ca. 4 cm2) of bark near
the center of each sample, shaved the exposed sapwood flat, affixed the transducer with
couplant, and constructed a support cylinder around the transducer using a hot-melt
adhesive. The support cylinder allowed for exact placement of the transducer on each
sampling day. Any remaining exposed sapwood around the support was coated with
petroleum grease to reduce local dehydration.


430 Florida Entomologist 71(4) December, 1988

AE Monitoring

An acoustical emission is a phenomenon in which transient elastic waves are gener-
ated by the rapid release of energy from a localized source within a material. In our
case, the energy released as a result of cavitation is believed to generate the elastic
waves that we detected. Elastic waves propagate outward from their source and can
be detected with the appropriate transducer as small displacements on the surface of
the specimen. Transducers transform elastic waves into electrical signals that can be
further amplified, filtered, and conditioned by other AE-detecting hardware.
We used AE equipment from Physical Acoustics Corporation (PAC) of Princeton,
NJ. The sensors were 150 kHz resonant transducers (PAC model R15) of the ceramic
piezoelectric type with -68 dB sensitivity. We used PAC model 1220A preamplifiers
with 100-300 kHz bandpass filters and 40 dB gain (0 dB = 100 pV; a decibel is 1/20 of
a logarithmic unit; thus 20 dB is 10 times greater, 40 dB is 100 times, 60 dB is 1000
times, etc.). The AE processor was a PAC model 3000/3004, which was set for an
additional gain of 32 dB. The total gain was thus 72 dB. An event was recorded when
the resulting amplified signal was greater than a 0.3 V threshold voltage; considerable
background noise was detected at 0.1 and 0.2 V. For comparison, others have used total
gains of 67 to 82 dB and thresholds of 0.12 to 0.25 V (Jones & Pefia 1986, Pefia & Grace
1986, Sandford & Grace 1985, Tyree & Dixon 1983, 1986, Tyree et al. 1984a, 1984b,
1986). For each recorded AE, the PAC 3000/3004 measured waveform parameters such
as duration microsecondss), counts (oscillations above threshold), rise time (microsec-
onds), amplitude (dB), and energy (volt seconds). Given that frequency is defined as
hertz (cycles per second), we estimated average frequency for each AE by dividing
counts by duration. We realize that this method will not calculate the true frequency
of a signal in most cases because (1) it assumes equal time intervals between oscillations,
(2) it allows for only one frequency per signal, and (3) a resonant transducer was used
which is most sensitive to its resonant frequency.
AEs were recorded three times weekly as the trunk samples dried. The base of the
transducer was coated with couplant and affixed to each sample within the support
cylinder. AEs were recorded for 5 minutes (or until at least 50 AEs had been recorded)
per sample using the same transducer, preamplifier, and AE-processor channel through-
out. Samples were tested until no AEs had been recorded for two successive sampling

Statistical Analysis

The AE data were analyzed by using the general linear models procedures of the
Statistical Analysis System (SAS Institute 1982). To keep the data set sizes manageable,


Weeks of AE
Moisture content (%) production
Tree Sample (unwaxed only)
Tre After 8 wks
age Lth Diam When First Last
Species (yr) (cm) (cm) cut waxed unwaxed week week

Pine 16 40 7.7 68 66 34 1 7
Oak 39 40 8.6 41 39 30 1 8
Maple 20 40 5.6 51 43 22 2 11
Birch 20 40 6.0 49 47 24 1 7

Haack et al.: Symposium on Agroacoustics

only the first 50 AEs were recorded from each 5-min sampling period. All recorded AEs
were used in determination of AE rate and nearly all in conducting simple correlations
among AE waveform parameters by species. However, in analyses of AE waveform
parameters, only AEs with values of 1 or greater for duration, count, rise time, and
amplitude were used, thus eliminating the weakest signals. Positive and negative linear
trends in the weekly mean values for all AE parameters were tested for each species
by using simple linear regression techniques. ANOVA and multiple range testing were
used to determine significant (P<0.05) differences among the four species for overall
mean values for each AE parameter.


Moisture Content

At the time of felling, moisture content was highest in pine and lowest in oak (Table
1). For all species, replicate samples dried similarly, being within 1% moisture content
on each sampling day. After 8 weeks of air drying, waxed samples had lost little mois-
ture compared with unwaxed samples (Table 1) indicating that waxing the cut ends
reduced dehydration. Unwaxed maple contained 16% moisture after 11 weeks, after
which AE production stopped.

AE Rate

The rate of acoustic emissions was far greater in unwaxed than in waxed samples
(Fig. 2; note differences in scale). In general, considering unwaxed samples only, AE
rate increased rapidly at first, remained steady, and then slowly declined. AE rate was
higher in pine than in any of the hardwood species. Unwaxed pine, oak, and birch
samples produced AEs during the first week of drying, whereas none were detected
from maple until week 2. Similarly, AE production ended in pine, oak, and birch samples
during weeks 7 and 8; however, maple samples produced AEs through week 11 (Table
Waxed pine and birch produced practically no AEs, whereas waxed oak and maple
produced some (Fig. 2). Cracks in the wax coatings of the oak and maple samples were
noted at the end of the study; it is possible that these cracks allowed water loss and
thus AE production.

AE Waveform Characteristics

All correlations among the AE parameters were highly significant (P<.001), and in
general, patterns were similar for each tree species (Table 2). Strong positive correla-
tions existed among duration, count, energy counts, amplitude, and rise time. Weak,
negative correlations were found for frequency with each of the other five AE paramet-
ers, indicating that signals of lower intensity were often of higher average frequency
(as defined here).
Weekly mean values for duration, counts, rise time, amplitude, energy, and average
frequency are presented by species for the unwaxed samples in Fig. 3. Data points
represent an average of 55-255 individual AEs.
Mean duration tended to decrease during the drying cycle for pine, oak, and birch,
whereas maple remained rather constant (Fig. 3; Table 3). The greatest decline occurred
in pine. For all weeks combined, AEs from pine were the longest while those from oak
were the shortest in duration (Table 4). Noguchi et al. (1985) recorded durations of
50-200 sec for two hardwood species and one conifer.






6 F- ....... .......... ... ....... ... ...

Florida Entomologist 71(4)


3 .......... .......... ......

Fig. 2. AE rate (events/minute) recorded with a 150 kHz resonant transducer from
unwaxed (top) and waxed (bottom) trunk samples cut from four tree species in February
and allowed to dry indoors in a heated building: threshold voltage = 0.3 V, total system
gain = 72 dB.

Mean counts per AE declined during the drying cycle for pine and oak, but no
consistent linear pattern was found for either maple or birch (Fig. 3; Table 3). The most
dramatic decline occurred in pine. For all weeks combined, AEs from pine had the most
counts while those from oak had the fewest (Table 4).


4 .

.- . .... ..

~\Y I.1111 L\\~m n\l-YI1 .L\~.m

December, 1988

2- ........ ..


Haack et al.: Symposium on Agroacoustics


Duration Counts Rise time Amplitude Energy Frequency

Eastern white pine (below right, N = 1645 AEs)
Duration .97 .66 .80 .87 -.26
Counts .93 .63 .75 .84 -.15
Rise time .67 .65 .48 .60 -.27
Amplitude .72 .66 .49 .71 -.28
Energy counts .84 .80 .68 .61 -.21
Frequency -.31 -.14 -.33 -.31 -.27
Northern red oak (above left, N = 1619 AEs)
Red maple (below right, 2259 AEs)
Duration .97 .71 .72 .88 -.28
Counts .97 .70 .66 .84 -.15
Rise time .73 .71 .44 .66 -.28
Amplitude .81 .76 .55 .60 -.32
Energy counts .89 .86 .72 .69 -.25
Frequency -.29 -.13 -.32 -.31 -.27
Paper birch (above left, N = 655 AEs)

A decline in mean rise time was noted for pine and maple but not for either oak or
birch (Fig. 3; Table 3). Pine demonstrated the greatest decline in rise time. For all
weeks combined, rise time was longest for pine and shortest for oak (Table 4).
Mean amplitude decreased over time in pine, oak, and birch, but not maple (Fig. 4;
Table 3). Pine showed the steepest decline. For all weeks combined, pine and birch had
the largest mean amplitude while oak and maple had the smallest (Table 4).
Mean energy tended to decrease during the drying cycle for pine and maple, but no
such trend was observed for oak or birch (Fig. 4; Table 3). The greatest decline in
energy occurred in pine. For all weeks combined, mean energy was highest in pine and
lowest in oak (Table 4).
Mean average frequency tended to increase during the drying cycle in pine, but did
not show a consistent pattern in oak, maple, or birch (Fig. 4; Table 3). For all weeks
combined, average frequency was broadly similar in all species (Table 4), being close to
the resonant frequency of the transducer used (i.e., 150 kHz).

Seasonal Changes in Xylem Moisture Content

The winter-time moisture contents reported here are typical of conifers and
hardwoods in north-temperate forests. Moisture content of sapwood is generally higher
in conifers than in hardwoods, reaching a maximum in winter and a minimum in mid-
summer. In hardwoods, on the other hand, moisture content reaches a maximum in
spring and a minimum in autumn or in mid-winter; it is lower in winter in ring-porous
types than in diffuse-porous hardwoods (Clark & Gibbs 1957, Gibbs 1958). Therefore,
the relative ranking of species by moisture content will vary depending on the season
during which a study is conducted.

Variation in Xylem Conduit Size

There is much within-tree, between-tree, and between-species variation in size of
xylem conduits. Three important trends in conduit size are (a) size decreases with

Florida Entomologist 71(4)

December, 1988

I 1 I I I I I I i I i
1 2 3 4 6 7 8 9 10 11



1 2 3 4 6 6 7 8 9 10 11


2 3 4 6 6 7 8 9


10 11

Fig. 3. Mean duration, counts, and rise time for AEs recorded with a 150 kHz
resonant transducer from unwaxed trunk samples cut from four tree species in February
and allowed to dry indoors in a heated building: threshold voltage = 0.3 V, total system
gain = 72 dB, N = 1645 AEs for pine, 2259 for maple, 655 for birch, and 1619 for oak.





Haack et al.: Symposium on Agroacoustics


Species Duration Counts Rise time Amplitude Energy Frequency

Pine 0.001 0.002 0.001 0.001 0.001 0.010
Oak 0.028 0.008 0.245 0.047 0.069 0.283
Maple 0.147 0.208 0.035 0.378 0.021 0.801
Birch 0.100 0.183 0.536 0.035 0.435 0.826

increasing height within the tree, (b) the number of conduits per unit area (tangential
section) increases with sampling height, and (c) average conduit length increases with
tree age (Bailey 1958, Fegel 1941, Zimmermann 1978, Zimmermann & Potter 1982).
Vessel length and diameter are positively correlated (Greenidge 1952, Zimmermann &
Jeje 1981). Xylem conduits are shortest in conifers, relatively long in diffuse-porous
hardwoods, and can be extremely long in ring-porous hardwoods. For example, white
pine tracheids are up to 4-5 mm long (Bailey & Tupper 1918), whereas the longest
vessels in maple, birch, and oak are typically 25-35 cm, 36-40 cm, and 10-11 m, respec-
tively (Zimmermann & Jeje 1981). However, only a few vessels belong to the longest
length class. In red maple, for example, although some vessels reached 42 cm in length,
over 50% were in the 0-4 cm class (Zimmermann & Potter 1982). Therefore, in addition
to season of year, AE studies will be influenced by sampling location within the tree,
tree age, and sample length.

AE Rate

The higher AE rate in unwaxed versus waxed samples reflected the former's greater
rate of water loss, given that the evaporated water resulted from cavitated xylem
conduits. The differences in AE rates that we recorded between waxed and unwaxed
specimens are similar to differences in AE patterns from well-watered and drought-


Duration Counts Rise time Amplitude2 Energy Frequency2 AEs3
Species (psec) (N) (ixsec) (dB) (V sec) (kHz) (N)

Pine 43.6 a1 7.2 a 13.3 a 41.4 a 3.0 a 168 a 1069
(8-187) (2-33) (1-95) (30-73) (1-30) (68-583)
Oak 30.9 d 5.2 d 10.0 d 39.6 b 2.2 d 171 a 801
(7-230) (2-36) (1-44) (30-66) (1-11) (27-444)
Maple 39.2 b 6.6 b 12.5 b 40.0 b 2.8 b 172 a 1264
(9-269) (2-42) (1-68) (25-79) (2-42) (30-500)
Birch 35.1 c 5.7 c 10.8 c 41.1 a 2.5 c 164 b 382
(8-172) (2-28) (1-42) (31-65) (1-13) (62-278)

'Means followed by the same letter (within columns) are not significantly different at the P < 0.05 \evel (\Dunans
multiple range test).
'For amplitude, 0 dB = 100 iV; frequency was calculated as (counts)/(duration) for each AE.
and amplitude were used. AEs were sensed with a 150 kHz resonant transducer.iac d ,
and amplitude were used. AEs were sensed with a 150 kHz resonant transducer.


Florida Entomologist 71(4)

December, 1988

L 41

T 39
D 37
36 I I -I I I --
1 2 3 4 6 6 7 8 9 10 11


0 I I t
1 2 3 4 6 6 7 8 9 10 11
R 220 -- RED MAPLE
Q 210 -
E 200-

C 190-

180 I I I f | I I -
1 2 3 4 6 7 8 9 10 11

Fig. 4. Mean amplitude, energy, and average frequency for AEs recorded with a
150 kHz resonant transducer from unwaxed trunk samples cut from four tree species
in February and allowed to dry indoors: threshold voltage = 0.3 V, total system g";-
= 72 dB, N = 1645 AEs for pine, 2259 for maple, 655 for birch, and 1619 for oak.


Haack et al.: Symposium on Agroacoustics

stressed conifers (Pefia & Grace 1986) and hardwoods (Jones & Pefia 1986). The increase
and eventual decrease in AE rate noted here is similar to the AE rate recorded from
drying lumber (Becker 1982, Honeycutt et al. 1985, Noguchi et al. 1980, 1983, 1985,
Skaar et al. 1980).
AE rate and the time over which AEs are produced are known to increase with
increasing specimen size (Becker 1982, Jones and Pefa 1986). In our study, because
specimen size was kept within a narrow range, we feel that differences in AE rate
reflect primarily variation in conduit size and moisture content, i.e., more water-filled
xylem conduits cavitated within the "listening" distance of the transducer in pine than
in the hardwood samples. However, it is possible that many of the AEs produced by
the hardwoods could have been above or below the bandwidth we examined (100-300
kHz). For example, Noguchi et al. (1985) reported that AE rate was greater in
hardwoods than conifers but they recorded AEs between 50 and 150 kHz. Additionally,
had testing been done during spring or summer, hardwood AE rate would probably
have been higher because more cells would have been water-filled, and conversely pine
AE rate may have been lower for similar reasons.

AE Waveform Parameters

Although most AE parameters varied among species, it is not known how similar
the findings would be in another experiment if we sampled other trees of the same
species, other tree locations, longer or shorter samples, during other seasons of the
year, or used other transducers that were sensitive to different frequency bands. For
example, given the above information on within-tree variation in xylem conduit size and
that frequency is inversely correlated with conduit size, we hypothesized that the aver-
age frequency would increase with increasing within-tree sampling height and sample
length as well as with lowered tree age.
Highest mean AE intensity (i.e., energy) was measured in pine. This may have
occurred because pine had the highest moisture content and the smallest xylem con-
duits. That is, more conduits cavitated within the listening range of the transducer on
pine than on any of the hardwoods. Sandford & Grace (1985) reported that the listening
distance was greater on hardwoods than conifers. If true, this could also help explain
lowered AE intensity on hardwoods because more distant AEs, which weaken as they
propagate, would be averaged in the calculations. Average intensity is also known to
decrease as sample diameter increases (Tyree et al. 1984b). As just mentioned, sound
attenuates as it travels through wood. We do not know the distances that AEs traveled
to reach the transducer in this study. Given two AEs of similar magnitude that occur
at different distances from a transducer, the more distant signal will be sensed as
Only in pine was there a significant trend of increasing average frequency with
drying time. Such evidence supports the contention that large conduits cavitate earlier
than smaller ones (Tyree & Dixon 1986, Tyree et al. 1984b). The values we reported
for average frequency are crude estimates based on counts and duration. Individual
AEs may have more than one major frequency component (Noguchi et al. 1985, Tyree
& Dixon 1983). Detailed spectrum analyses are needed to determine the actual fre-
quency components produced by drought-stressed plants, and how these spectra change
as drought intensifies.
Considering that oaks have the largest xylem conduits of the species tested, it is
puzzling that average frequency was not substantially lower in oak. However, because
the samples were only 40-cm long, more than 80% of the earlywood vessels may have
been severed (see Fig. 12 in Zimmermann & Jeje 1981). Moreover, Zimmermann (1983)
states that most earlywood vessels of ring-porous hardwoods cavitate during winter.

438 Florida Entomologist 71(4) December, 1988

Because samples were collected in February, most earlywood vessels probably had
already cavitated. Thus, the AEs recorded from our oak samples, most likely resulted
from cavitations occurring in the much shorter (mostly <25 cm) latewood vessels (Zim-
mermann & Jeje 1981).

Implications for Bark- and Wood-Feeding Insects

Visual and olfactory cues are considered of primary importance in host finding by
bark- and wood-feeding insects (Haack & Slansky 1987). It is well recognized, however,
that ultrasound in the range 20-200 kHz is produced and/or detected by insects in
diverse groups such as Diptera (Saini 1984), Lepidoptera (Spangler 1988). Neuroptera
(Olesen & Miller 1979), Orthoptera (Silver et al. 1980), and Trichoptera (Silver & Halls
1980). Two species of Ips bark beetles (Scolytidae) are known to produce low-frequency
ultrasound (Swaby & Rudinsky 1976, Wilkinson et al. 1967). We believe it highly plaus-
ible that some insects could detect and utilize drought-induced AEs during host finding
and accepting. It is not known, however, to what distance ultrasound radiates from
stressed trees into the surrounding air. If insects do utilize tree-produced AEs, we
believe it is likely that they would sense them primarily after landing on the bark or
after initiating boring. In support of such a scenario is evidence that mate calling by a
species of pyralid moth involves using pheromones for long-range communication and
ultrasonic pulses for short-range orientation (Spangler et al. 1984).
To be meaningful to insects, host-produced AEs should convey information that is
consistent and specific. AEs are consistent in that they are associated with drought-
stressed trees; AEs are typically lacking in well-watered trees and in dry host material.
More specific information could be ascertained from AE waveform parameters such as
intensity and frequency. Since these waveform parameters tend to change as drought
intensifies, certain combinations of them could signify susceptibility. We are currently
measuring a number of AE parameters on drought-stressed potted trees that are ex-
posed to bark beetle attack. Our future studies will test whether substrates artificially
pulsed with AEs that simulate the drought-stressed condition are any more attractive
to bark beetles than are unpulsed controls. Besides visual, olfactory, and gustatory
cues, it may be learned that certain insects utilize plant-produced vibrational cues in
host recognition or acceptance. It is already known that certain Homoptera communi-
cate by sending and perceiving vibrational signals through their host plants (Michelsen
et al. 1982).


The authors wish to thank Ned Clapp (Oak Ridge National Laboratory), Donald
Dickmann (Michigan State University), John Ferguson (Michigan State University),
and Marcia Patton-Mallory (USDA Forest Products Laboratory) for critically reviewing
an earlier draft of this paper; and the U.S. Department of Agriculture, Insect Pest
Science Program, for partially funding these studies. This research was conducted in
cooperation with Michigan State University.
Mention of a commercial or proprietary product does not constitute an endorsement
by the U.S. Department of Agriculture.


BAILEY, I. W. 1958. The structure of tracheids in relation to the movement of liquids,
suspensions, and undissolved gases. Pages 71-82 in K. V. Thimann (ed.), The
Physiology of Forest Trees. Ronald Press, New York.

Haack et al.: Symposium on Agroacoustics 439

BAILEY, I. W., AND W. W. TUPPER. 1918. Size variation in tracheary cell: I. A
comparison between the secondary xylems of vascular cryptogams, gymno-
sperms, and angiosperms. Proc. Am. Acad. Arts Sci. 54: 149-204.
BECKER, H. F. 1982. Acoustic emissions during wood drying. Holz als Roh- und
Werkstoff 40: 345-350.
CLARK J., AND R.. D. GIBBS. 1957. Studies in tree physiology. IV. Further investi-
gations of seasonal changes in moisture content of certain Canadian forest trees.
Canadian J. Bot. 35: 219-253.
DIXON, M. A., J. GRACE, AND M. T. TYREE. 1984. Concurrent measurements of
stem density, leaf and stem water potential, stomatal conductance and cavitation
on a sapling of Thuja occidentalis L. Plant Cell Environ. 7: 615-618.
FEGEL, A. C. 1941. Comparative anatomy and varying physical properties of trunk,
branch, and root wood in certain northeastern trees. Bull. New York State Coll.
For. Syracuse Univ. Vol. 14, No. 2b, Tech. Publ. No. 55: 1-20.
GIBBS, R. D. 1958. Patterns of seasonal water content of trees. Pages 43-69 in K. V.
Thimann (ed.), The Physiology of Forest Trees. Ronald Press, New York.
GREENIDGE, K. N. H. 1952. An approach to the study of vessel length in hardwood
species. American J. Bot. 39: 570-574.
HAACK, R. A., AND F. SLANSKY JR. 1987. Nutritional ecology of wood-feeding Col-
eoptera, Lepidoptera, and Hymenoptera. Pages 449-486 in F. Slansky Jr. and J.
G. Rodriguez (eds.), The Nutritional Ecology of Insects, Mites, and Spiders.
Wiley, New York.
HONEYCUTT, R. M., C. SKAAR, AND W. T. SIMPSON. 1985. Use of acoustic emissions
to control drying rate of red oak. For. Products J. 35 (1): 48-50.
JONES, H. G., AND J. PENIA. 1986. Relationships between water stress and ultrasound
emission in apple (Malus x domestic Borkh.). J. Exp. Bot. 37: 1245-1254.
KRAMER, P. J. 1983. Water Relations of Plants. Academic Press, New York.
KRAMER, P. J., AND T. T. KOZLOWSKI. 1979. Physiology of Woody Plants. Academic
Press, New York.
LEWIS, D. B., AND D. M. GOWER. 1980. Biology of Communication. Wiley, New
MATTSON, W. J., AND R. A. HAACK. 1987. The role of drought in outbreaks of
plant-eating insects. BioScience 37: 110-118.
MICHELSEN, A., F. FINK, M. GOGALA, AND D. TRAUNE. 1982. Plants as transmis-
sion channels for insect vibrational songs. Behav. Ecol. Sociobiol. 11: 269-281.
NOGUCHI, M., Y. KAGAWA, AND J. KATAGIRI. 1980. Detection of acoustic emissions
during hardwoods drying. J. Japan Wood Res. Soc. 26: 637-638.
NOGUCHI, M., Y. KAGAWA, AND J. KATAGIRI. 1983. Acoustic emission generation
in the process of drying hardwoods. J. Japan Wood Res. Soc. 29: 20-23.
NOGUCHI, M., S. OKUMURA, AND S. KAWAMOTO. 1985. Characteristics of acoustic
emissions during wood drying. J. Japan Wood Res. Soc. 31: 171-175.
OERTLI, J. J. 1971. The stability of water under tension in the xylem. Z. Pflanzen-
physiol. 65: 195-209.
OLESEN, J., AND L. A. MILLER. 1979. Avoidance behavior in green lacewings. II.
Flight muscle activity. J. Comp. Physiol. 131: 121-128.
PANSHIN, A. J., AND C. DE ZEEUW. 1980. Textbook of Wood Technology. 4th ed.
McGraw-Hill, New York.
PENA J., AND J. GRACE. 1986. Water relations and ultrasound emissions of Pinus
sylvestris L. before, during and after a period of water stress. New Phytol. 103:
SALES, G., AND D. PYE. 1974. Ultrasonic Communication by Animals. Wiley, New
SALLEO, S., AND M. A. Lo GULLO. 1986. Xylem cavitation in nodes and internodes
of whole Chorisia insignis H.B. et K. plants subjected to water stress: Relations
between xylem conduit size and cavitation. Ann. Bot. 58: 431-441.
SANDFORD, A. P., AND J. GRACE. 1985. The measurement and interpretation of
ultrasound from woody stems. J. Exp. Bot. 36 (163): 298-311.
SAINI, R. K. 1984. Acoustic emissions of tsetse. Insect Sci. Applic. 5: 341-344.

Florida Entomologist 71(4)

SAS INSTITUTE. 1982. Sas User's Guide: Statistics. Raleigh, NC.
SCHWARTZKOPF, J. 1974. Mechanoreception. Pages 273-352 in M. Rockstein (ed.),
The Physiology of Insecta. Vol. 2, 2nd ed. Academic Press, New York.
SILVER, S. C., AND J. A. T. HALLS. 1980. Recording the sounds of hydropsychid
larvae-A cautionary tale. J. Comp. Physiol. 140: 159-161.
SILVER, S., K. KALMRING, AND R. KUHNE. 1980. The responses of central acoustic
and vibratory interneurones in bush crickets and locust to ultrasonic stimulation.
Physiol. Entomol. 5: 427-435.
SKAAR, C., W. T. SIMPSON, AND R. M. HONEYCUTT. 1980. Use of acoustic emissions
to identify high levels of stress during oak lumber drying. For. Products J. 30
(2): 21-22.
SPANGLER, H. G. 1988. Moth hearing, defense and communication. Annu. Rev. En-
tomol. 33: 59-81.
SPANGLER, H. G., M. D. GREENFIELD, AND A. TAKESSIAN. 1984. Ultrasonic mate
calling in the lesser wax moth. Physiol. Entomol. 9: 87-95.
SWABY, J. A., AND J. A. RUDINSKY. 1976. Acoustic and olfactory behavior of Ips
pini (Say) (Coleoptera: Scolytidae) during host invasion and colonization. Z. Ang.
Entomol. 81: 421-432.
TYREE, M. T., AND M. A. DIXON. 1983. Cavitation events in Thuja occidentalis L.?
Ultrasonic acoustic emissions from the sapwood can be measured. Plant Physiol.
72: 1094-1099.
TYREE, M. T., AND M. A. DIXON. 1986. Water stress induced cavitation and em-
bolism in some woody plants. Physiol. Plant. 66: 397-405.
TYREE, M. T., M. A. DIXON AND R. G. THOMPSON. 1984a. Ultrasonic acoustic
emissions from the sapwood of Thuja occidentalis measured inside a pressure
bomb. Plant Physiol. 74: 1046-1049.
TYREE, M. T., M. A. DIXON, E. L. TYREE, AND R. JOHNSON. 1984b. Ultrasonic
acoustic emissions from the sapwood of cedar and hemlock: An examination of
three hypotheses regarding cavitation. Plant Physiol. 75: 988-992.
Detection of xylem cavitation in corn under field conditions. Plant Physiol. 82:
1967. Stridulation and behavior in two southeastern Ips bark beetles (Coleopt-
era: Scolytidae). Florida Entomol. 50: 185-195.
ZIMMERMANN, M. H. 1978. Hydraulic architecture of some diffuse-porous trees.
Canadian J. Bot. 56: 2286-2295.
ZIMMERMANN, M. H. 1983. Xylem Structure and the Ascent of Sap. Springer-Verlag,
ZIMMERMANN, M. H., AND A. A. JEJE. 1981. Vessel-length distribution in stems of
some American woody plants. Canadian J. Bot. 59: 1882-1892.
ZIMMERMANN, M. H., AND D. POTTER. 1982. Vessel-length distribution in branches,
stems, and roots of Acer rubrum. IAWA Bull. 3: 103-109.

December, 1988


Hagstrum et al.: Symposium on Agroacoustics


U.S. Department of Agriculture, Agricultural Research Service
U.S. Grain Marketing Research Laboratory
Manhattan, Kansas 66502

Insect Attractants, Behavior and Basic Biology Research Laboratory,
Agricultural Research Service, U.S. Department of Agriculture,
Gainesvllle, Florida 32604


The possibility of using insect-produced sounds to estimate insect populations with-
out removing grain samples was investigated. The number of insect-produced sounds,
heard with a piezoelectric microphone pushed into the grain, increased as the number
of lesser grain borer, Rhyzopertha dominica (F.), larvae increased. The probabilities
of detection and the accuracy of estimation of insect densities with the acoustical method
were comparable to those obtained with a standard grain trier.


Se investig6 la posibilidad de usar sonidos producido por insects para estimar las
poblaciones de insecto sin tener que remover muestras de granos. El nfimero de sonidos
producido por insects, oidos con un micr6fono piezoel6ctrico empujado dentro del
grano, aument6 a media que el nfmero de larvas de Rhyzopertha dominica (F.),
aument6. Las probabilidades de detecci6n y la exactitud del estimado de la densidad de
insects con el metodo acfstico fue comparable a aquellos obtenidos en un probador
patron de grano.

Nondestructive methods of detecting insects feeding inside kernels of grain during
storage include x-ray of insects within the grain (Milner et al. 1950), and measurement
of insect-produced carbon dioxide (Bruce et al. 1982) or sounds (Vick et al. 1988). The
initial cost of the x-ray machine and the ongoing cost of the x-ray film and chemicals to
develop the x-ray film are high and examination of each of the individual grains on the
x-ray film for insects is labor intensive. While the cost and labor might be lower for
carbon dioxide and acoustical methods than for the x-ray method, x-ray has the added
advantages of 1) detection of both live and dead insects (although live insects cannot be
distinguished from dead insects) and 2) identification of the species and stage of insect
detected. However, eggs and small larvae are generally difficult to distinguish from
denser portions of the grain. Also, not all carbon dioxide or sound detected in grain are
produced by insects. Adams et al. (1953) suggested that the acoustical method might
be used to monitor ". . grain within storage bins for infestation without sampling or
removing the grain from the bins in much the same manner as permanent thermocouple
systems are now used for checking the heating of grain in storage." This potential for
automation of insect monitoring with the acoustical method may be a major advantage
over the carbon dioxide and x-ray methods.

Florida Entomologist 71(4)

We investigated whether the insect-produced sounds, heard with a piezoelectric
microphone pushed into grain, can be used to detect insect infestations and to estimate
insect densities as efficiently as other methods.

The acoustical system shown in Figure 1 was used for discovery and estimation of
the densities of lesser grain borer, Rhyzopertha dominica (F.), larvae detecting the
feeding sounds they produce. The system was composed of a durable piezoelectric micro-
phone (#9D0576 BNF Enterprises, Peabody, Mass.) mounted on the end of a probe
which was pushed into the grain, a battery operated amplifier (Insecta-scope, Sound
Technologies Inc., Kilgore, TX) and earphones for listening to insect-produced sounds.
A Krohn-Hite model 3700 filter (not shown in Figure 1) was used between the amplifier
and earphones to remove frequencies below 1000 hz and above 3000 hz. Feeding sounds
were recorded on a Technics magnetic tape recorder, Model RS-B16, equipped with
dBx. Magnetic tape recordings of these sounds were analyzed later with a Fast Fourier
Transformation (FFT) instrument (Nicolet Model 660A) and a Hewlett Packard vectra
computer to compare earphone with instrument counts. The FFT determined the fre-
quency content of the sounds, whereas instrument counts with the Hewlett Packard
Vectra computer coupled through a Hewlett Packard universal counter Model 5316A


Fig. 1. Acoustical insect detection system including a probe with piezoelectric micro-
phone (Top), battery operated amplifier (Bottom Right), and earphones (Bottom Left).

December, 1988

Hagstrum et al.: Symposium on Agroacoustics

determined the number of voltage spikes in a predetermined time interval (Webb et al.
Insect populations of several densities were prepared by diluting lesser grain borer
cultures with clean wheat. The actual densities and age structure of the lesser grain
borer populations were determined at each density level by x-raying 90 ml samples of
wheat. The acoustical system was used 1) to estimate insect densities by counting the
number of sounds per unit of time produced by insects at each of nine locations during
a one minute interval in a grain mass (10 cm deep by 14 cm diameter) in a four-liter jar
or 2) to determine the probability of detection from the fraction of nine locations in a
grain mass (10 cm deep by 30 cm diameter) in a 20-liter can at which sounds were heard
during 20-s intervals. Both containers were set on 10 cm thick synthetic foam inside a
40 cm diam cylinder of 5 cm thick synthetic foam to dampen background sound. The
acoustical probe was inserted 3 cm deep in the grain at the centers and at eight equidis-
tant locations halfway between the centers and edges of the containers. A series of 1:1
dilutions was repeated four times in each size of container.

-0.18X -1.64X
Y=500(1-(1.22e -0.22e )500

,40 r =0.9601, df==3

/ z
Z 300 C
0 K
u 20- -
6 -200 8
-0.023X 200
I Y=50(1-e ) c
2 Z
10- r =0.9648, df=16 1

0-" *-0
0 10 20 30 40 50 60
Fig. 2. Relationship between insect density and the number of insect-produced
sounds counted with instrument (x) by Vick et al. (1988) or heard in the present study
with earphones (.).

Florida Entomologist 71(4)

The number of sounds heard with the piezoelectric microphone pushed into the grain
increased as the density of lesser grain borer larvae increased (Fig. 2). This is consistent
with the results of Vick et al. (1988) using an acoustical system which required that
grain samples be taken. In both studies, the relationship between the number of sounds
or voltage spikes counted and the number of lesser grain borer larvae present was not









r =0.4838, df=176


0 0

0 0


0S 4
@0 0


10 20 30 40

Fig. 3. Relationship between earphone counts of insect-produced sounds and instru-
ment counts of insect-produced sounds in the present study.


December, 1988

Hagstrum et al.: Symposium on Agroacoustics

linear. The nonlinear increase in insect-produced sounds is probably the results of an
increased simultaneous occurrence of sounds of insects which cannot be distinguished
as separate sounds.
In the present study, the positive correlation between the number of insect-produced
sounds counted with earphones and the number counted with the instrument indicates
that our use of earphones instead of instrumentation has not altered the relationship
observed between insect sounds and insect density (Fig. 3). The slope of the regression
indicates that each insect-produced sound counted with the earphones was actually
composed of an average of 3.76 instrument counts. This means that the instrument
counts from Vick et al. (1988) must be divided by 3.76 in comparing the two data sets.
Even with this adjustment, the counts in this earlier study increased much more rapidly
as insect density increased than in the present study. In Vick's earlier study, the
number of counts per insect increased rapidly between 0 and 10 insects per 100 ml
sample of wheat, but the counts increased more slowly as insect density increased from
10 to 20 insects per sample. Because we are estimating increases in the number of
insects by increases in the number of sounds counted, insect densities are estimated

107- LOGY=1.77LOGX+0.43

106 r20.9562, df=57

10 ete 0
Lu 10- *


K x LOGY=1.09LOGX+0.19
1 x
S." r = 0.7123, df=78

0.1 1 10 10 2 10 3
Fig. 4. Regressions of the logarithm of variance in the number of counts or insects
(Y) against logarithm of mean (X) number of counts or insects are given for present
study (.) and an earlier study (x) by Vick et al (1988). Regression line for other insect
sampling methods from Hagstrum et al. (1988) is shown as dashed line for comparison.

Florida Entomologist 71(4)

December, 1988

b 1.0-
0 0.8

o 4 0
< C0.4- /

LL Y=1-(0.96e +43X0.04e 021X)

I2 r =0.9393, df =7

0 1 2 3 4 5 6 7
Fig. 5 Regression of the fraction of samples with insects or sounds against mean
number of lesser grain borer larvae per sample. The regression line for other sampling
methods from Hagstrum et al. (1988) is given as dashed line for comparison.
better between 1 and 10 insects per 100 ml sample of grain than at higher densities. In
the present study, over a much broader range of 0 to 60 insects per 100 ml sample of
grain, the rate of increase in counts for each increase in insect density was intermediate
between rates observed for the 0 to 10 and 10 to 20 density ranges in the earlier study.
The relationship between the mean number of insects or insect-produced sounds per
sample and the sample-to-sample variance (Fig. 4) provides a measure of insect distri-
bution and a means of calculating the accuracy of estimation of insect populations
(Hagstrum 1987). The relationship between the variance and mean for the acoustical
method are very similar to the relationship between variance and mean for other insect
sampling methods. Such similarities indicate that the accuracies of the acoustical method
will be similar to accuracies for other methods.
We also determined the fraction of samples with insect-produced sounds because
this represents the probability of detection. The increase in the fraction of samples with
insect-produced sounds as insect density increased is described by the double logarithm
model (Fig 5). The similarity of relationship between probability of detection and insect
density for acoustical method to that for other insect sampling methods is also shown.
Because it is simpler to record the number of locations at which insect-produced sounds
are are heard (presence or absence sampling) than to count the number of insect-pro-
duced sounds at each location, these curves may also be useful in estimating insect
densities from the fraction of sample locations with insects.

Ryker: Symposium on Agroacoustics

These studies suggest that with further development the acoustical method might
provide a quick and easy way of detecting and perhaps even estimating insect popula-
tions in stored grain. In our laboratory studies, the probability of detection and accuracy
of estimation with an acoustical method appear to be quite similar to those for other


Mention of a commercial or proptietary product in this paper does not constitute
endorsement by the USDA.


Aural detection of grain infestation internally with insects. Science 118: 163-164.
BRUCE, W. A., M. W. STREET, R. C. SEMPER, AND DAVID FULK. 1982. Detection
of hidden insect infestations in wheat by infrared carbon dioxide gas analysis.
Advances in Agric. Tech., South. Series, No. 26, U.S. Dept. Agric., 1-8.
HAGSTRUM, D. W. 1987. Seasonal variation of stored wheat environment and insect
populations. Environ. Entomol. 16: 17-83.
HAGSTRUM, D. W., R. L. MEAGHER, AND L. B. SMITH. 1988. Sampling statistics
and detection or estimation of diverse populations of stored-product insects. En-
viron. Entomol. 17: 377-380.
MILNER, M., M. R. LEE, AND R. KATZ. 1950. Application of x-ray techniques to the
detection of internal insect infestation of grain. J. Econ. Entomol. 43: 933-935.
VICK, K. W., J. C. WEBB, B. A. WEAVER, AND C. LITZKOW. 1988. Sound-detection
of stored-product insects that feed inside kernels of grain. J. Econ. Entomol. 81:
WEBB, J. C., C. A. LITZKOW, AND D. C. SLAUGHTER. 1988. A computerized acous-
tical larval detection system. Applied Engineering in Agriculture. 4: 268-274.


478 Willow Street
Ashland, Oregon 97520


The utility of recording, monitoring, and manipulating acoustic signals of destructive
bark beetles and some methods of bioassay and analysis related to pheromone research
are discussed. J. A. Rudinsky and his research group at Oregon State University
utilized particular chirps of males and females to acoustically stimulate pheromone re-
lease, for bioassays of odors as possible pheromones, and as indicators of the behavior
of beetles hidden under the bark. A summary of the acoustic signals of five species of
Dendroctonus, D. pseudotsugae, D. ponderosae, D. brevicomis, D. valens, and D. fron-
talis, is presented.


Se discuten la utilidad de grabar, chequear, y manipular las sefiales acusticas de
escarabajos destructores de cortezas y algunos m6todos de bio-ensayos y analisis re-

448 Florida Entomologist 71(4) December, 1988

lacionados con investigaciones de feromonas. J. A. Rudinsky y su grupo investigator
en la Universidad del Estado de Oregon utiliz6 chirridos particulares de machos y
hembras para estimular acusticamente la liberaci6n de feromonas, para bio-ensayos de
olores como possible feromonas, y como indicadores del comportamiento de los es-
carabajos escondidos debajo de la corteza. Se present un sumario de las sefiales acis-
ticas de cinco species de Dendroctonus, D. pseudotsugae, D. ponderosae, D. bre-
vicomis, D. valens, y D. frontalis.

In keeping with the purpose of this agroacoustic symposium, this paper considers
the practical results obtained from the study of insect acoustics by J. A. Rudinsky and
his co-workers at Oregon State University. It describes some of the laboratory methods
developed to study destructive bark beetles and brings together and reviews the acous-
tic signals of five species of Dendroctonus. The signaling repertoire of several of these
beetles has been obscured until now by non-uniform illustration and by being scattered
piecemeal through several papers as different types of signals, i.e., stress, attractant,
rivalry, territorial, and courtship signals, were described.
Virtually every important timber species of conifer in North America, excluding
redwood and sequoia, are attacked by bark beetles, Family Scolytidae (Stark 1982).
The economic impact of one species, the Douglas-fir beetle, Dendroctonus pseudotsugae
Hopkins, was $1.5 million for the states of Oregon and Washington in a non-epidemic
year, 1980; and the mountain pine beetle, Dendroctonus ponderosae Hopkins, was re-
sponsible for over $13 million of losses of pines the same year (Ruderman 1980). Four
epidemics of the Douglas-fir beetle between 1950 and 1969 killed 7.4 billion boardfeet
of prime timber valued at over $3 billion (Furniss & Orr 1970). Losses of timber in the
southern U.S. due to the depredations of the southern pine beetle, Dendroctonus fron-
talis Zimmermann, reach similar proportions (Bronson 1986). Attempts to control infes-
tations of bark beetles are made difficult because the insects are inaccessible except
during a brief flight period, spending their lives under the bark of trees (Rudinsky
1962). Therefore, research focused upon pheromones as a possible means of control
(Rudinsky 1963).
Stridulation by species of Dendroctonus was mentioned as early as 1909 (Hopkins
1909), and chirping was identified by Chapman (1955) as a characteristic of males. Allen
et al. (1958) published oscillograms of a pair of D. pseudotsugae beetles chirping beneath
the bark of a Douglas-fir tree. The elytro-abdominal stridulatory apparatus and method
of sound production of several species of Dendroctonus was described and illustrated
by scanning electron micrographs by Michael & Rudinsky (1972). Wilkinson et al. (1967)
reported that females of Ips calligraphus (Germar) attracted to nuptial chambers chirp
to polygynous males guarding the entry. Barr (1969) demonstrated similar behavior for
Ips paraconfusus (LeConte). She proposed that the gender that invades the host is
silent, and the gender that is attracted to the invader has well developed stridulation
in sound producing species. Males of polygynous species colonize the trees, and females
chirp at the entry; conversely, females of monogynous beetles, such as species of Den-
droctonus, colonize and males chirp. Barr also described three different types of
stridulatory organs occurring on numerous species of bark beetles.
Alexander et al. (1963) reported loud chirps coming from the burrow of a skin beetle,
Trox suberosus Fabricius, and generalized that the chirps of most beetles would likely
be of low intensity, close range sounds. The long range signals of beetles, analogous to
the songs employed by crickets and other Orthoptera, were expected to be pheromones,
chemical signals (Ryker 1975). With this expectation, the idea of studying acoustic
signals of economically impacting species of bark beetles seemed unrewarding. How-
ever, the chemical communication system of the Douglas-fir beetle, our number one

Ryker: Symposium on Agroacoustics

target insect at Oregon State University in the 1970's, was so complex and difficult to
analyze that it became apparent that an understanding of the acoustic signals might be
useful in manipulating the beetle's signaling behavior for pheromone analysis (Rudinsky
& Michael 1972).
Rudinsky (1968) made a discovery that underlined the importance of acoustic signals
for understanding the behavior of the Douglas-fir beetle. He had attracted thousands
of flying beetles to caged logs infested with female Douglas-fir beetles. After he placed
a male with each female in her burrow (gallery), the attractiveness of the logs was
completely extinguished within minutes. The effect was so dramatic that Rudinsky
hypothesized the release of an antiaggregation pheromone, which he called "the mask."
Further, he found that the male did not have to enter the female's burrow for the
pheromone mask to be produced, but it did have to be able to chirp. Surgically silenced
males had no effect, but normal males that chirped continually while on the screen over
the entry triggered the masking effect (Rudinsky 1968, 1969, Rudinsky & Ryker 1977).
Acoustic signals still are not considered promising as population control tools, but
they became invaluable for analysis of the various pheromone components. Indeed, the
communication system of Dendroctonus bark beetles cannot be understood or explained
without detailed knowledge of the acoustic signals. Acoustic signals were of value: 1)
as stimuli to cause males or females to release pheromones for analysis; 2) as indicator
responses of beetles during behavoral bioassays of possible chemical stimuli; 3) as indi-
cators of the timing of release of pheromones during natural interaction between the
male and female; 4) as indicators of passive vs aggressive behavior during natural beetle
interactions; and 5) as releasing signals in the chain of stimuli and responses that allow
destructive bark beetles to select host trees, to attack en masse, and to regulate the
density of the attack, preventing overcrowding and subsequent starvation (Rudinsky
1968, Rudinsky & Ryker 1977, Alcock 1981, Ryker 1984). These studies have resulted
in the patenting of methylcyclohexenone as a control pheromone for the Douglas-fir
beetle (Rudinsky 1974) and the identification of endo-brevicomin and verbenone as anti-
aggregation pheromones with potential as control substances for the mountain pine
beetle (Ryker & Yandell 1983, Borden et al. 1987).


Test beetles were obtained each spring by cutting sections of logs from infested
trees, holding them in the warmth of a greenhouse while the brood matured, and then
storing the infested logs in walk-in coolers at 4C until needed. All species of Dendroc-
tonus can be sexed by the presence of a sclerotized plectrum on the seventh abdominal
tergite only in males (Michael & Rudinsky 1972). Females were introduced into holes
drilled into the bark of freshly cut logs and given 36 h to begin a burrow and produce
attractive frass (bark shreds and fecal pellets containing tree odors and pheromones).
At this point, a Hewlett-Packard 15119A condenser microphone was placed directly
above the entry, leaving space for the tiny male to walk beneath it and enter the burrow
(Fig. 1). The signal was amplified by a Princeton Applied Research 113 low noise
preamplifier, and recorded on a Nagra 4.2L tape recorder at 38 cm/sec tape speed.
Recording system components all showed an essentially flat response from about 0.02
to 22 kHz, and frequencies below 0.3 kHz were filtered by the preamplifier settings to
minimize stray noise. Signals were monitored via earphones and an oscilloscope, and
signal parameters were measured on a Tektronix 5103N storage oscilloscope. Stored
tracings were photographed by a Polaroid camera (Rudinsky & Ryker 1976).


Pheromone odors were trapped on Porapak-Q for gas chromatographic/mass spec-

450 Florida Entomologist 71(4) December, 1988


coe c cc11



Fig. 1. Method of tape recording and monitoring acoustic signals of Dendroctonus
bark beetles on a section of log.

trometric analysis easily by placing together a male and female Douglas-fir beetle in a
short, paper-lined glass vial and trapping odors from purified air flowing over 30 to 50
such vials within a larger glass container. The male would chirp and jostle the female,
and they would each release pheromones. Two males confined in a single vial would
chirp and fight, and would also release pheromones. However, females would not release
their pheromones if placed together. To trigger pheromone release by single females,
Rudinsky stimulated them with recorded male attractant chirps played back through a
piezoelectric ceramic disk. The disk was pressed to a silicone rubber gasket over an
opening in the glass chamber facing the screened ends of the vials, where it acted as a
transducer for recorded chirps (Rudinsky et al. 1973, Ryker et al. 1979) (Fig. 2). Only
the chirp of males near the females' burrow successfully triggered females to release
their pheromones (Rudinsky et al. 1973).

Ryker: Symposium on Agroacoustics 451

cable to

to air
GC/MS intake
and computer ^ 1 '

Top View

Porapak chamber
trap with vials

Fig. 2. Glass chamber used to trap pheromones via acoustic stimulation (Ryker et
al. 1979) modified from earlier chamber (Rudinsky et al. 1973). The small vial shown
has the paper removed from the inside; the wire mesh is stainless steel. A single female
is in each vial in this example.


Jantz & Rudinsky (1965) tested the responses of male beetles walking on screening
above a tiny vial containing either natural attractive frass or a dilute solution of synthe-
sized chemical odors. Their technique was developed further to include monitoring of
the presence or absence of chirping (Rudinsky & Michael 1972), and finally the identifi-
cation of the type of chirp elicited (Rudinsky & Ryker 1976) (Fig. 3). This technique
was very helpful in determining when the beetles switched from being attracted to
being repelled as the concentration of certain pheromones increased. For example, the
pheromone methylcyclohexenone at only 0.002% concentration in a solution of several
other attractants (evaporating at about 1 ng/h) stimulated walking males to double their
turning and digging behavior above the test vial, and to double their tendency to emit
attractant chirps. Increasing the concentration (and evaporation rate) of methylcyc-
lohexenone 100 times stimulated the male beetles to pass by the vial without stopping
and to give aggressive (rivalry) chirps. Flying beetles showed a similar inhibition to
attractive traps and logs in the forest in the presence of higher concentrations of this
pheromone (Rudinsky & Ryker 1976, 1980). The chirping response of male beetles was
similarly used to bioassay candidate odors and concentrations of synthesized
pheromones with other species of Dendroctonus (Michael & Rudinsky 1972, Rudinsky
& Ryker 1977, Rudinsky et al. 1974, Ryker & Yandell 1983).


The Douglas-fir beetle, D. pseudotsugae Hopkins, emits five known types of sounds
(Fig. 4). Females click intermittently in their burrows in the bark. Clicks appear to be

Florida Entomologist 71(4)





Fig. 3. Olfactory walkway for testing of various concentrations of candidate
pheromones, modified for acoustic monitoring (Rudinsky & Ryker 1977).

territorial signals because other females turn away from established, clicking females
as they select a place to cut an entry hole. When she is disturbed by digging or scratch-
ing sounds, a female increases the frequency of clicking (Rudinsky & Michael 1973).
Thus she clicks when the male digs into the entry, causing him to release concentrated
methylcyclohexenone as an antiaggregation pheromone (Rudinsky et al. 1976).
The male attractant chirp is produced by males responding to the pheromones in
female frass at the burrow entry (Rudinsky & Michael 1972). A dilute solution of synthe-
sized tree terpenes and pheromones, with methylcyclohexenone in trace amounts, dupli-
cates the effect of frass (Rudinsky 1973, Rudinsky & Michael 1972, Rudinsky & Ryker
1976). This male chirp also signals the female to actively release additional pheromone,
including concentrated methylcyclohexenone (Rudinsky et al. 1973), and to stridulate
(clicking) loudly (Rudinsky & Michael 1973).
Male Douglas-fir beetles emit an interrupted chirp, the aggressive chirp (= rivalry
chirp), when they meet another beetle in a burrow, on which occasion they invariably
attack (Fig. 4). Males fight head to head. The female faces away from the male and
pushes backward with the heavily armed posterior portion of her elytra, attempting to
force the male out of her burrow (Rudinsky & Ryker 1976, Ryker 1984). This is aggres-

December, 1988

Ryker: Symposium on Agroacoustics





II Ih 1L

0.1 second

Fig. 4. Drawings of oscillograms of stridulations of Dendroctonus pseudotsugae, the
Douglas-fir beetle. All oscillograms in this paper are to the same time scale; the shortest
tracings each have a 0.1 sec. sweep duration.

1 11 '1 1
1-1 11v


454 Florida Entomologist 71(4) December, 1988

sive courtship. When unmated beetles or two males meet, both sexes release concen-
trated methylcyclohexenone. This pheromone, even in the absence of another beetle,
stimulates males to give the aggressive chirp and to pass by rather than to stop and
dig (Rudinsky & Ryker 1976).
The male Douglas-fir beetle courts the female aggressively, with much jostling,
biting, and aggressive chirping for about an hour, after which he emits the courtship
chirp (Fig. 4). While giving the courtship chirp, the male strokes the female gently.
Copulation follows within minutes (Rudinsky & Ryker 1976). Males give the stress chirp
when disturbed (Fig. 4) (Rudinsky & Michael 1972).
The mountain pine beetle, D. ponderosae Hopkins, males and females each emit two
chirps (Fig. 5). Females click intermittently in their burrows in the bark; this is pre-
sumed to have a territorial function (Rudinsky & Michael 1973). The female produces
a multi-impulse chirp when the entry is disturbed by digging or scratching, and when
defending her burrow against intruders, a territorial function. She ceases to chirp only
when touched by a chirping male (Ryker & Rudinsky 1976a). If the male beetle does
not chirp correctly, the female attacks and repels him from her burrow if possible,
chirping continually.
Male mountain pine beetles give the attractant chirp, an interrupted chirp, when
stimulated by the odor of female frass or a synthetic pheromone mixture (Michael &
Rudinsky 1972, Ryker & Yandell 1983, Yandell 1984). This chirp also accompanies
aggressive behavior whenever unmated beetles meet or when rival males fight
(Rudinsky et al. 1974). A male also gives this chirp while digging through frass to enter
the female's burrow and while attacking his future mate; this aggressive phase of
courtship lasts only a few minutes before mating (Ryker & Rudinsky 1976a).
Mountain pine beetle males produce the simple chirp both when disturbed (stress)
and during courtship. Within five minutes of contacting the female, the male ceases to
attack the female, begins to stroke her, and gives simple chirps. Copulation follows
(Ryker & Rudinsky 1976a).
The western pine beetle, D. brevicomis LeConte, female has two and the male has
three chirps (Fig. 6). Females click intermittently (about 8 clicks per minute) when
alone in their burrows. When disturbed by other females boring nearby (Rudinsky &
Michael 1973), or when a male enters her burrow, they give a multi-impulse chirp about
twice per second.
Males emit an interrupted chirp with two or three subchirps when attracted to
female frass attractantt) (Rudinsky & Michael 1974) and when courting the female (Fig.
The male rivalry chirp is an interrupted chirp with an average of about five subchirps
(Fig. 6) and is given during fighting (Rudinsky & Michael 1974). Males also emit a
simple stress chirp when disturbed.
The red turpentine beetle, D. valens LeConte, male has two distinctive chirps and
the female emits several rather variable sounds (Fig. 7). Males produce a simple chirp
when attracted to female frass; the stress chirp given by the male when disturbed is
not measurably different (Ryker & Rudinsky 1976b).
The "agreement" chirp is a grating sound emitted by a female when a male is digging
and chirping in her entry. This is possibly a territorial signal. When first contacted by
a male, females produce a much shorter and less variable chirp, the "greeting" chirp.
Females also produced stress chirps, containing only about five pulses of sound (Fig.
7). This is the only species of Dendroctonus known to do so.
The rivalry or aggressive chirp of male beetles is produced during fighting, as well
as during the first 30 seconds of courtship. The entire train of sounds is an interrupted
chirp, made by a single motion of the abdomen (Ryker & Rudinsky 1976b).

Ryker: Symposium on Agroacoustics



if 1 IIh 1liii




Fig. 5. Drawings of oscillograms of stridulations of Dendroctonus ponderosae, the
mountain pine beetle.



Florida Entomologist 71(4)






Fig. 6. Drawings of oscillograms of stridulations of Dendroctonus brevicomis, the
western pine beetle.

December, 1988

Ryker: Symposium on Agroacoustics






Fig. 7. Drawings of oscillograms of stridulations of Dendroctonus valens, the red
turpentine beetle.

The southern pine beetle, D. frontalis Zimmermann, is diminutive compared to the
other species of Dendroctonus, but it is a remarkable chirper. Males readily produce
many stress chirps when disturbed (Fig. 8). The male attractant chirp is an interrupted
chirp produced when the male is digging in the frass of a virgin female (Rudinsky &
Michael 1974).



Florida Entomologist 71(4)

-... 11 ---




Fig. 8. Drawings of oscillograms of stridulations of Dendroctonus frontalis, the
southern pine beetle.

A female produces a series of chirps (about eight per second) when a male enters
her burrow. Female chirps are probably territorial signals (Rudinsky & Michael 1973).
Males defend their burrows, fighting other males and producing long, fast series of
rivalry chirps (Fig. 8). These chirps are high-pitched, brief, and are delivered at a rate
of about 16 per second (Rudinsky & Michael 1974, Rudinsky et al. 1974).

All five species of Dendroctonus reviewed in this paper show aggressive behavior
in both sexes. Colonizing females compete for limited and ephemeral host material, have
territorial clicks or chirps, and attack strangers. Males both fight over control of a
female's burrow and test the female for fitness by attacking her (Alcock 1981, Rudinsky
& Ryker 1976, Ryker 1984). Specific chirps are associated with fighting; these signals
either cease or are replaced by a distinctive courtship signal before mating. This
suggests that aggressive chirps are important to the reproductive success of male Den-
droctonus. Excluding D. ponderosae, these species of Dendroctonus also have a distinc-
tive signal to the female as they locate the entry of her burrow attractantt chirp). This
signal has been shown in two species to trigger an interaction between the female and
the male that results in release of antiaggregation pheromones (Rudinsky et al. 1976),
thus protecting their brood resources from pressure of over-population and warning

December, 1988

Ryker: Symposium on Agroacoustics 459

flying beetles to seek more sparsely settled trees. Alcock (1981) gives a convincing
argument for individual selection vs signal selection for the benefit of the population.
Such closely interlocked acoustic and chemical signals are most complex in species of
Dendroctonus bark beetles that attack and overwhelm living trees en masse. Similarly
complex development of acoustic and chemical signals, especially territorial and aggres-
sive sounds, might be expected in other highly competitive species of bark beetles
(Swedenborg et al. in press).


A list of acoustic studies of bark beetles other than Dendroctonus, arranged by host
tree, follows:

On pine: Ips calligraphus (Germar), Wilkinson et al. (1967); Ips paraconfusus
(LeConte), Barr (1969); Ips pini (Say), Oester & Rudinsky (1975), Swaby & Rudinsky

On spruce: Ips concinnus (Mannerheim), Oester & Rudinsky (1975, 1979); Ips plastog-
raphus (LeConte), Oester & Rudinsky (1979); Ips tridens (Mannerheim), Oester and
Rudinsky (1975, 1979); Ips typographus L, Rudinsky (1979); Polygraphus rufipennis
Kirby, Rudinsky et al. (1978); Hylurgops rugipennis (Mannerheim), Oester et al. (1978).

On Douglas-fir: Pseudohylesinus nebulosus (LeConte), Oester et al. (1981).

On ash: Hylesinus oleiperda Fabricius, Rudinsky & Vallo (1979); Leperisinus califor-
nicus Swaine and L. oregonus Blackman, Vernoff & Rudinsky (1980); Leperisinus fra-
xini Panzer, Rudinsky & Vallo (1979).

On apple: Scolytus mali Bechst, Rudinsky et al. (1978b).

On elm: Hylurgopinus rufipes (Eichhoff), Swedenborg et al. (in press).


I thank Dr. R. T. Walden of the National Center for Physical Acoustics for inviting
me to participate in the National Agroacoustics Symposium and for providing financial
assistance. The research described herein was supported by a series of National Science
Foundation grants to the late Prof. J. A. Rudinsky from 1968 to 1982. Figures 1-3 were
drawn by Bonnie Hall. I thank the Entomological Society of America for permission to
use Figure 2, and I thank the Centre National de la Recherche Scientifique, Paris, for
permission to use Figure 3.


ALCOCK, J. 1981. Natural selection and communication among bark beetles. Florida
Entomol. 65: 17-32.
ALLEN, D. G., R. R. MICHAEL, AND S. A. STONE. 1958. Sounds of Douglas fir beetle
activity. Ore. For. Lands Res. Center, Res. Note No. 36. P. O. Box 571, Corval-
lis, OR. 97339.
ALEXANDER, R. D., T. E. MOORE, AND R. E. WOODRUFF. 1963. The evolutionary
differentiation of stridulatory signals in beetles (Insecta: Coleoptera). Anim.
Behav. 11: 111-115.
BARR, B. A. 1969. Sound production in Scolytidae (Coleoptera) with emphasis on the
genus Ips. Canadian Entomol. 101: 636-672.

460 Florida Entomologist 71(4) December, 1988

A. C. OEHLSCHLAGER. 1987. Response of the mountain pine beetle, Dendroc-
tonus ponderosae Hopkins (Coleoptera: Scolytidae), to five semiochemicals in
British Columbia lodgepole pine forests. Canadian J. For. Res. 17: 118-128.
BRONSON, G. 1986. The bug that's eating Texas. Forbes, April 7, 1986.
CHAPMAN, J. A. 1955. Sex determination by stridulation sounds in the Douglas fir
beetle, Dendroctonus pseudotsugae. Bi-Mon. Prog. Rep. Div. Forest Biol., Ot-
tawa 11: 2.
FURNISS, M. M., AND P. W. ORR. 1970. Douglas-fir beetle. Fo-. Pest leaflet 5,
U.S.D.A. For. Service, Portland.
HOPKINS, A. D. 1909. Contributions toward a monograph of the scolytid beetles. I.
The Genus Dendroctonus. U.S. Dep. Agric. Bur. Ent. Tech. Ser. 17, Part I.
JANTZ, O. K., AND J. A. RUDINSKY. 1965. Laboratory and field methods for assaying
olfactory responses of the Douglas-fir beetle, Dendroctonus pseudotsugae Hop-
kins. Canadian Entomol. 97: 935-941.
MICHAEL, R. R., AND J. A. RUDINSKY. 1972. Sound production in Scolytidae:
specificity in male Dendroctonus beetles. J. Insect Physiol. 18: 2189-2201.
OESTER, P. T., AND J. A. RUDINSKY. 1975. Sound production in Scolytidae: stridu-
lation by "silent" Ips bark beetles. Z. Angew. Entomol. 79: 421-427.
OESTER, P. T., AND J. A. RUDINSKY. 1979. Acoustic behavior of three sympatric
species of Ips (Coleoptera: Scolytidae) co-inhabiting Sitka spruce. Z. Angew.
Entomol. 87: 398-412.
OESTER, P. T., J. A. RUDINSKY, AND L. C. RYKER. 1981. Olfactory and acoustic
behavior of Pseudohylesinus nebulosus (Coleoptera: Scolytidae) on Douglas-fir
bark. Canadian Entomol. 113: 645-650.
OESTER, P. T., L. C. RYKER, AND J. A. RUDINSKY. 1978. Complex male premating
stridulation of the bark beetle Hylurgops rugipennis (Mann.). Coleopt. Bull. 32:
RUDERMAN, F. K. 1980. Current Stumpage Prices, 1980. U.S.D.A. For. Ser.,
P.N.W. For. & Range Exp. Sta., Portland.
RUDINSKY, J. A. 1962. Ecology of the Scolytidae. Ann. Rev. Ent. 7: 327-348.
RUDINSKY, J. A. 1963. Response of Dendroctonus pseudotsugae Hopkins to volatile
attractants. Boyce Thompson Inst. 22: 22-38.
RUDINSKY, J. A. 1968. Pheromone-mask by the female Dendroctonus pseudotsugae
Hopkins, an attractant regulator. Pan-Pac. Entomol. 44: 248-250.
RUDINSKY, J. A. 1969. Masking of the aggregation pheromone in Dendroctonus
pseudotsugae Hopkins. Science 166: 884-885.
RUDINSKY, J. A. 1973. Multiple functions of the Douglas-fir beetle pheromone 3-
methyl-2-cyclohexen-l-one. Environ. Entomol. 2: 579-585.
RUDINSKY, J. A. 1974. Method for the control of the Douglas-fir beetle using 3-
methyl-2-cyclohexen-l-one. U.S. Patent 3,830,935.
RUDINSKY, J. A. 1979. Chemoacoustically-induced behavior of Ips typographus. Z.
Angew. Entomol. 88: 537-541.
RUDINSKY, J. A., AND R. R. MICHAEL. 1972. Sound production in Scolytidae:
chemostimulus of sonic signal by the Douglas-fir beetle. Science 175: 1386-1390.
RUDINSKY, J. A., AND R. R. MICHAEL. 1973. Sound production in Scolytidae: stridu-
lation by female Dendroctonus beetles. J. Insect Physiol. 19: 689-705.
RUDINSKY, J. A., AND R. R. MICHAEL. 1974. Sound production in Scolytidae:
"rivalry" behaviour of male Dendroctonus beetles. J. Insect Physiol. 20: 1219-
RUDINSKY, J. A., AND L. C. RYKER. 1976. Sound production in Scolytidae: rivalry
and premating stridulation of male Douglas-fir beetles. J. Insect Physiol. 22:
RUDINSKY, J. A., AND L. C. RYKER. 1977. Olfactory and auditory signals mediating
behavioral patterns of bark beetles. Pages 195-209 in V. Labeyrie (ed.), Comporte-
ment des Insectes et Milieu Trophique. No. 625, Colloques Internationaux de
Centre National de la Recherche Scientifique, Paris.

Ryker: Symposium on Agroacoustics

RUDINSKY, J. A., AND L. C. RYKER. 1980. Multifunctionality of Douglas-fir beetle
pheromone 3,2-MCH confirmed with solvent di-butyl phthalate. J. Chem. Ecol.
6: 193-201.
RUDINSKY, J. A., AND V. VALLO. 1979. The ash bark beetles Leperisinus fraxini
and Hylesinus oleiperda: stridulatory organs, acoustic signals, and pheromone
production. Z. Angew. Entomol. 87: 417-429.
RUDINSKY, J. A., P. T. OESTER, AND L. C. RYKER. 1978a. Gallery initiation and
male stridulation of the polygamous spruce bark beetle Polygraphus rufipennis
Kirby. Ann. Entomol. Soc. Am. 71: 317-321.
RUDINSKY, J. A., V. VALLO, AND L. C. RYKER. 1978b. Sound production in
Scolytidae: attraction and stridulation of Scolytus mali (Coleoptera: Scolytidae).
Z. Angew. Entomol. 86: 381-389.
production in Scolytidae: 3-methyl-2-cyclohexen-l-one released by the female
Douglas-fir beetle in response to male sonic signal. Environ. Entomol. 2: 505-
aggregative-rivalry pheromone of the mountain pine beetle, and a new arrestant
of the Southern pine beetle. Environ. Entomol. 3: 90-98.
GAN. 1976. Sound production in Scolytidae: female sonic stimulus of male
pheromone release in two Dendroctonus beetles. J. Insect Physiol. 22: 1675-
RYKER, L. C. 1975. Calling chirps in Tropisternus natator (D'Orchymont) and T.
lateralis nimbatus (Say) (Coleoptera: Hydrophilidae). Entomol. News 86: 179-
RYKER, L. C. 1984. Acoustic and chemical signals in the life cycle of a beetle. Sci.
Amer. 250: 112-123.
RYKER, L. C., AND J. A. RUDINSKY. 1976a. Sound production in Scolytidae: aggres-
sive and mating behavior of the mountain pine beetle. Ann. Entomol. Soc. Am.
69: 677-680.
RYKER, L. C., AND J. A. RUDINSKY. 1976b. Sound production in Scolytidae: acoustic
signals of male and female Dendroctonus valens LeConte. Z. Angew. Entomol.
80: 113-118.
RYKER, L. C., AND K. L. YANDELL. 1983. Effect of verbenone on aggregation of
Dendroctonus ponderosae Hopkins (Coleoptera: Scolytidae) to synthetic attrac-
tant. Z. Angew. Entomol. 96: 452-459.
RYKER, L. C., L. M. LIBBEY, AND J. A. RUDINSKY. 1979. Comparison of volatile
compounds and stridulation emitted by the Douglas-fir beetle from Idaho and
western Oregon populations. Environ. Entomol. 8: 789-798.
STARK, R. W. 1982. Generalized ecology and life cycle of bark beetles. Pp. 21-45 in
J. B. Mitton and K. B. Sturgeon (eds.), Bark Beetles in North American Conif-
ers. Univ. Texas Press, Austin.
SWABY, J. A., AND J. A. RUDINSKY. 1976. Acoustic and olfactory behavior of Ips
pini (Say) during host invasion and colonization. Z. Angew. Entomol. 81: 421-
SWEDENBORG, P. D., R. L. JONES, AND L. C. RYKER. Stridulation and pairing bavior
of the native American elm bark beetle Hylurgopinus rufipes (Coleoptera:
Scolytidae). Canadian Entomol. (In press)
VERNOFF, S., AND J. A. RUDINSKY. 1980. Sound production and pairing behavior
of Leperisinus californicus Swaine and L. oregonus Blackman (Coleoptera:
Scolytidae) attacking Oregon ash. Z. Angew. Entomol. 90: 58-74.
1967. Stridulation and behavior in two southeastern Ips bark beetles (Coleopt-
era: Scolytidae). Florida Entomol. 50: 185-195.
YANDELL, K. L. 1984. Sound production of Dendroctonus ponderosae Hopkins (Col-
eoptera: Scolytidae): a comparison of populations from three host pines in Oregon.
Z. Angew. Entomol. 97: 180-187.

Florida Entomologist 71(4)

December, 1988


Insect Attractants, Behavior, and Basic Biology Research Laboratory,
Agricultural Research Service, U.S. Department of Agriculture,
Gainesville, Florida 32604


Many flies, including tephritid fruit flies, produce acoustic courtship signals; how-
ever, the message these signals transmit is not always clear. Courtship in general is
often considered to be a mechanism for species recognition and the prevention of hyb-
ridization. Such a proposition suffers from the rarity of sympatric character displace-
ment, the complexity of displays in some species with few close sympatric relatives and
the simplicity of courtships in other species with many close sympatric relatives. The
theory that displays are sexually selected advertisements of male qualities that females
prefer in a mate faces none of these objections. The acoustic signals of male Caribbean
fruit flies can be examined from a sexual selection perspective. Females appear to
prefer large mates. Both the "calling" and "precopulatory" songs have characteristics
that are correlated to male size and females are more likely to react to songs typical of
large males. Studies of fruit fly acoustics may serve agriculture either in the develop-
ment of attractants or by providing criteria to monitor the sexual competitiveness of
insects reared for sterile release programs.


Muchas moscas, incluyendo moscas tefriticas, produce sefiales acfisticas de cortejo;
sin embargo, el mensaje transmitido por estas senales no es siempre claro. En general,
el cortejo es a menudo considerado ser un mecanismo de reconocimiento de species y
para prevenir la hibridaci6n. Tal proposici6n sufre de la rareza de la destituci6n de la
carateristica simpatrica, lo complejo de la demostraci6n de algunas species con pocos
parientes cercanos simpAtricos, y la simplicidad de cortejos en otras species con muchos
parientes cercanos simpAtricos. La teoria que demostraciones del macho son un anuncio
de sus selectas cualidades sexuales que las hembras prefieren en su consorte present
ninguna de esta objecciones. Las sefales acsticas de machos de la mosca del Caribe
pueden ser examinadas desde una perspective de selecci6n sexual. Las hembras parecen
preferir machos grandes. Cantos "llamantes" y "precopulatorios" tienen caracteristicas
que estin correlacionadas al tamafio del macho y es mds probable que las hembras
reaccionen al canto tipico de machos grandes. Estudios acasticos de las moscas de fruta
pudieran servir a la agriculture ya sea en el desarrollo de atrayentes o proporcionando
un criterio para chequear la competividad sexual de insects esteriles criados para
programs de liberaciones.

Many male flies, including tephritid fruit flies, make sexual courtship sounds. The
contexts in which these songs are produced are often well described, but the information
being transmitted to listening flies is less well understood. Acoustic and other courtship
displays, such as wing waving, colors, and scents, are sometimes considered species
isolating mechanisms, ie., means by which creatures of the same species and opposite
sex recognize each other. Calls in translation might be nothing more than, "I am a male
species A-I am a male species B," etc.
A model for the evolution of such a song in a fruit fly might start with allopatric
speciation (see Mayr 1963). First, a population of flies becomes geographically divided.
Suppose, for example, that a fly cannot cross a mountain which arises to bisect its

Sivinski: Symposium on Agroacoustics 463

range. Environments and, hence, selection pressures on the two sides of the mountain
may not be identical; perhaps one is warmer or wetter or contains a novel predator. In
addition, different mutations might arise in the separated regions from alleles whose
frequencies differed in the first place because of genetic bottlenecks so that the raw
material of evolution is also different. Over time, the genomes of the two populations
diverge. When erosion breeches the mountain and the flies from the two sides mingle
there are reproductive consequences arising from their previous isolation-hybrids be-
tween the two types may be less fit than offspring produced by same-type parents.
Thus flies that only mate with members of their own isolate are at a reproductive
advantage and will eventually replace those that do not. After all, the latter are spend-
ing time and resources on offspring (should zygotes even develop) that will bear fewer
or no grandchildren. Flies that are clearly of a particular type and those that prefer
mates that are clearly of their own type can "collaborate" in evolving features, de novo
or through exaggeration, whose role is to explicitly say "I am a species A."
This model is appealing because the signals of so many species are recognizably
different, even to humans. One can easily assume that these signals are different be-
cause species are different. However, the argument faces objections when applied to
male-produced displays. First, it makes the largely unmet prediction of sympatric
character displacement. That is, animals sometimes have similar signals in the parts of
their ranges that do not overlap. Where the ranges do overlap, the need for species
recognition should force the signals to diverge. Such cases of character displacement
are rarely found (e.g. Walker 1974). Second, some of the most complex and elaborate
courtships occur in species that have few close relatives and who would have little
chance to err by choosing an almost, but not quite, proper mate (Otte 1972, Alcock &
Pyle 1979, Thornhill & Alcock 1983, West-Eberhard 1984). A mirror image phenomenon
can be observed by anyone watching the flies on a fresh pat of dung or other resource
that concentrates insects. Closely related species can often be seen in great density on
the fecal surface. The opportunity for error seems quite high, yet the courtship of these
insects tends to be relatively simple. (see for example phoretic sphaerocerids of the
genus Borborillus, Sivinski 1983, 1984, or many of the dung breeding Sarcophagidae,
pers. obs.).
If it sometimes seems that species isolation is an insufficient explanation for the
variety and complexity of male courtship, what does all its behavioral sound and fury
signify? An alternative explanation arises from how much the sexes invest in their
offspring. Females produce large gametes. The number of offspring they can have is
largely limited by the number of eggs they can generate. Males make small, cheap
gametes and the scope of their paternity is largely a function of how many females can
be mounted and eggs fertilized. While number of matings is generally important to
males, copulatory quantity is often of little concern to females. Rather, females can
enhance their success at reproduction by choosing quality mates, i.e., by incorporating
the best available genes into their offspring or by obtaining a valuable resource, such
as ejaculate-borne proteins from the "wealthiest" male they encounter (Trivers 1972,
Thornhill 1976, Sivinski & Smittle 1987).
Assume, then, that at least some courtship displays are advertisements that males
produce to convince females that they have the qualities females prefer in a mate. One
might imagine that females would be very discriminating, looking for relatively small
differences among suitors. Species could be kept separate incidentally by females en-
gaged in making minute distinctions among a subset of males of her own species. Hence
the lack of large scale character displacement is less surprising. The elaborate courtship
of isolated species poses no difficulty to the female choice/male advertisement model. If
there were only one species in the world, it might still evolve complex communications
between the sexes. The variety and species-uniqueness of signals may simply be due to

Florida Entomologist 71(4)

the improbability of selection ever repeating itself exactly during the evolution of inde-
pendent displays.
There is profit in examining fruit fly songs as "intersexually selected" instruments
of persuasion (Sivinski & Burk 1988). In the Caribbean fruit fly, Anastrepha suspense
(Loew) (Caribfly), there are two male sexual sounds, both produced by wing movement
(Webb et al. 1976, Sivinski & Webb 1985a). The "calling song" is made up of repeated
bursts (pulse trains). It is sung by males on the leaf territories they defend against
interloping males and coincides with the release of pheromone from abdominal glands
and everted anal membranes (Burk 1983, Nation 1972, Sivinski 1988, Chuman et al.
1988). Calling song may be sung in the absence of other flies, although it can be con-
tinued in the presence of a visiting fly. The "precopulatory song" is a continuous sound
made as the male mounts the female and typically lasting until the male genitalia are
completely threaded through the female ovipositor, a period averaging a minute and a
half (Webb et al. 1984).
What qualities might such songs advertise? One trait that is preferred by females
is large size (Burk & Webb 1983). Big males are more likely to mate than smaller rivals
in many species of flies (e.g. Sivinski 1984). It is often unclear, however, if bulk wins
out in competition with other males for access to females or if it is something favored
by females or both. In Caribbean fruit flies the problem is simpler. Females go to males
to initiate courtship and so must either prefer big males or be more likely to sense them.
Are there characteristics of the signals that are correlated to large size? In the
calling song, the interval between pulse trains is such a correlate. Larger males tend
to have shorter intervals (Burk and Webb 1983, see however Webb et al. 1984). The
reason is unknown; perhaps with size come greater resources to sustain what must be
a more expensive signal. The sound pressure level (SPL) of the precopulatory song is
also related to male size, big males being louder (Webb et al. 1984).
Do females use these correlates when choosing a mate? Virgin females are more
responsive (i.e., increase movement) to shorter pulse train interval songs. Their activity
increases when a recording of a short, but not a long, pulse train interval song is
broadcast into their cage (Sivinski et al. 1984). This increased movement is a plausible
response to a sound heard at some distance that directs the hearer toward an attractive
goal. It is interesting that the papaya fruit fly (Toxotrypana curvicauda [Gerstacker])
sings only as it approaches nearby females. These sounds quiet the female (Sivinski &
Webb 1985b). Perhaps these females become still in order to access their approaching
suitors. Also of note is a sexual dimorphism in the response of Caribflies to calling
sounds. Males become quiet in the presence of the shorter pulse train interval songs,
while mated females are not affected by differences in pulse train intervals. This makes
it less likely that virgin females are simply startled by the short interval sound and so
become agitated. If short pulse train interval songs are both more effective and more
energetically expensive, it might be predicted that pulse train intervals would decrease
when a male was "certain" a female was observing. This is the case. An individual male
will shorten his pulse train intervals in the presence of a female. When another male is
close by, the pulse trains themselves are longer in time, and made at a higher frequency,
suggesting a between-male communication role for calling song as well (Sivinski &
Webb 1986).
The precopulatory song is also an important component of courtship. Its absence in
muted (dealated) males leads to a greater number of rejections by females. A calling
sound reproduced by a tape recorder at a SPL of 90 dB (OdB re 20 pL pac) broadcast
increases female acceptance, but a broadcast at 50 dB does not (Sivinski et al. 1984).
Thus, females make sexual decisions on the basis of precopulatory SPL and that SPL
is positively correlated to male size.
What agricultural purpose is served by studies of fruit fly sex sounds and an appre-
ciation that not all males and their displays are equally attractive? Studies of this type,

December, 1988

Sivinski: Symposium on Agroacoustics

not only on sounds but also visual and pheromonal displays, have at least two potential
applications. The first, which is arguably more important in studies of chemical cues,
is the manufacture of attractants and more efficient traps, either to monitor pests or to
control them. Caribbean fruit fly calling sounds are attractive in themselves. Traps
baited only with recorded songs are more effective than silent controls (Webb et al.
1983). It remains to be demonstrated that sounds and pheromone together, for example,
can be more attractive than a more conventional chemical cue alone or, if they are,
whether it is economically feasible to produce such a trap. The second use is in the
quality control of reared insects. A major concern of sterile release programs is that
their reared insects be sexually competitive with the wild rivals they encounter in the
agricultural arena. While sterile releases are often a very efficient means of fruit fly
control, there have been some less spectacular efforts where overflooding ratios have
reached thousands of reared for every wild fly and still not succeeded (Burk & Calkins
1982). A possible reason for these expensive failures (and perhaps some overly expen-
sive successes) is the inadequate courtships of the reared flies. A first step in ensuring
that competitive flies are released is a description of sexual behavior in a species fol-
lowed by periodic comparison of factory reared stock wild flies in mating compatibility
and competitive tests. In this way, waste can be minimized and the efficiency of pro-
grams be enhanced. It may concern A. suspense breeders, for instance, that radiation
can increase the calling pulse train interval of irradiated males (Webb et al. 1987). On
the other hand, years of domestication in Central American stocks have not dramatically
influenced the acoustical signals of the Mediterranean fruit fly Ceratitis capitata (Wied.)
(Sivinski et al. 1988).


Peter Landolt, Norm Leppla, and Steve Wing made numerous improvements in the
manuscript. Thanks to Drs. Thomas Waldon and J. C. Webb for inviting me to partici-
pate in the symposium.


ALCOCK, J., AND D. W. PYLE. 1979. The complex courtship behavior of Physiphora
demandata (F.) Diptera: Ottidae) Zeit. fur Tierpsychol. 49: 352-362.
BURK. T. 1983. Behavioral ecology of mating in the Caribbean fruit fly, Anastrepha
suspense (Loew). Florida Entomol. 66: 330-344.
BURK, T., AND C. 0. CALKINS. 1982. Medfly mating behavior and control
strategies.Florida Entomol. 66: 3-18.
BURK, T., AND J. C. WEBB. 1983. Effect of male size on calling propensity, song
parameters, and mating success in Caribbean fruit flies (Anastrepha suspense
(Loew)). Ann. Entomol. Soc. Am. 76: 678-682.
1988. Variation in the composition of volatiles produced by the male Caribbean
fruit fly Anastrepha suspense (Loew) (Tephritidae). J. Chem. Ecol. (in press)
MAYR, E. 1963. Animal species and evolution. Belknap of Harvard University Press,
Cambridge, Massachusetts.
NATION, J. L. 1972. Courtship behavior and evidence for a sex attractant in the male
Caribbean fruit fly, Anastrepha suspense. Ann. Entomol Soc. Am. 65: 1364-
OTTE, D. 1972. Simple versus elaborate behavior in grasshoppers: an analysis of com-
munication in the genus Syrbala. Behavior 42: 291-322.
SIVINSKI, J. 1983. The natural history of a phoretic sphaerocerid Diptera fauna. Ecol.
Entomol. 8: 419-426.
SIVINSKI, J. 1984. Sexual conflict and choice in a phoretic fly Borborillusfrigipennis
Sphaeroceridae). Ann. Entomol. Soc. Am. 77: 232-235.

Florida Entomologist 71(4)

SIVINSKI, J. 1989. Lekking and the small scale distribution of the sexes in the Carib-
bean fruit fly, Anastrepha suspense (Loew). Florida Entomol. (in press).
SIVINSKI, J., AND T. BURK. 1988. Reproductive and mating behavior, in A. Robinson
and G. Hooper, eds. Fruit flies-their biology, natural enemies, and control.
Elsevier, Amsterdam, The Netherlands.
SIVINSKI, J., AND T. BURK. 1988. Reproductive and mating behavior, in A. Robinson
and G. Hooper, eds. Fruit flies-their biology, natural enemies, and control.
Elsevier, Amsterdam, The Netherlands (in press).
SIVINSKI, J., AND B. SMITTLE. 1987. Male transfer of materials to mates in the
Caribbean fruit fly, Anastrepha suspense (Diptera: Tephritidae). Florida En-
tomol. 70: 233-238.
SIVINSKI, J., AND J. C. WEBB. 1985a. Sound production and reception in the caribfly
Anastrepha suspense. Florida Entomol. 68: 273-278.
AND 1985b. The form and function of th acoustic courtship signals of
the papaya fruit fly, Toxotrypana curvicauda. Florida Entomol. 68: 634-641.
SIVINSKI, J., AND J. C. WEBB. 1986. Changes in Caribbean fruit fly acoustic signal
with social situation (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 79: 146-149.
SIVINSKI, J., T. BURK, AND J. C. WEBB. 1984. Acoustic courtship signals in the
caribfly Anastrepha suspense. Anim. Behav. 32: 1011-1016.
SIVINSKI, J., C. 0. CALKINS, AND J. C. WEBB. 1988. Comparison of acoustic
courtship signals between wild and laboratory reared Mediterranean fruit fly.
Florida Entomol. (in press)
THORNHILL, R. 1976. Sexual selection and paternal investment in insects. Am. Nat.
110: 153-163.
THORNHILL, R., AND J. ALCOCK. 1983. The evolution of insect mating systems.
Harvard Univ. Press, Cambridge, MA.
TRIVERS, R. L. 1972. Parental investment and sexual selection. Pages 136-179 in B.
Campbell, ed. Sexual selection and the descent of man 1871-1971. Aldine-Alter-
ton, Chicago, Illinois.
WALKER, T. J. 1974. Character displacement and acoustic insects. Am. Zool. 14:
WEBB, J. C., T. BURK, AND J. SIVINSKI. 1983. Attraction of female Caribbean fruit
flies Anastrepha suspense (Loew) (Diptera: Tephritidae) to males and male-pro-
duced stimuli in field cages. Ann. Entomol. Soc. Am. 76: 996-998.
1976. The analysis and identification of sounds produced by the male Caribbean
fruit fly, Anastrepha suspense (Loew). Ann. Entomol. Soc. Am. 69: 415-420.
WEBB, J. C., J. SIVINSKI, AND C. LITZKOW. 1984. Acoustical behavior and sexual
success in the Caribbean fruit fly, Anastrepha suspense (Loew) (Diptera: Tep-
hritidae). Environ. Entomol. 13: 650-656.
WEBB, J. C., J. SIVINSKI, AND B. SMITTLE. 1987. Acoustical courtship signals and
sexual success in irradiated Caribbean fruit flies (Anastrepha suspense) (Diptera:
Tephritidae). Florida Entomol. 70: 103-109.
WEST-EBERHARD, M. J. 1984. Sexual selection, social communication, and species
specific signals in insects. Pages 283-324 in T. Lewis, ed. Insect Communication.
Academic Press, London.

December, 1988

Spangler: Symposium on Agroacoustics 467


U.S. Department of Agriculture
Agricultural Research Service
Carl Hayden Bee Research Center
2000 E. Allen Road
Tucson, AZ 85719


Both lesser wax moth, Achroia grisella (F.) and greater wax moth, Galleria mel-
lonella L. males produce sounds using tymbals located on their tegulae. Wing movement
twists one end of a tymbal causing it to buckle and produce an ultrasonic pulse. Both
sexes are equipped with tympanic ears that hear the high-frequency sound. A. grisella
females use the sound to locate males prior to copulation. In contrast, female G. mel-
lonella respond to the sound with wing fanning. This wing fanning sets off a more
complex, three-step behavioral sequence that allows the females to locate males by
male-produced pheromone. Techniques that make use of the moth-produced sounds to
detect and control these pests of bee products include locating calling males with elec-
tronic detectors and using acoustically-baited traps to capture receptive females.


Ambos machos de los gusanos de cera, Achroia grisella (F.) y Galleria mellonella
L., produce sonidos usando timbales localizados en su tegula. El movimiento del ala
dobla una punta del timbal causando que se encorve y produzca una pulsaci6n ul-
tras6nica. Ambos sexos estan equipados con oidos timpanicos que oyen el sonido de alta
frecuencia. Hembras de A. grisella usan el sonido para encontrar al macho antes de
copular. En contrast, hembras de G. mellonella responded al sonido abanicando con
las alas. Abanicando con las alas produce una secuencia mas compleja de tres etapas
que le permit a las hembras encontrar los machos por feromonas producidas por los
machos. La t6cnica que hace uso de los sonidos producidos por las alevillas para detectar
y controlar estas plagas de products de las abejas incluye el localizar a los machos que
llaman usando detectores electr6nicos y usando trampas de cebos ac6sticos para cap-
turar las hembras receptoras.

Many insects possess organs that serve as receivers of airborne sounds. In Hemipt-
era, Orthoptera and Diptera these organs function primarily in intraspecific communica-
tion. In Lepidoptera, Neuroptera, Dictyoptera and Coleoptera the organs serve to warn
the bearer of the potential threat of a predator. In both groups, however, some insects
use their hearing ability for both communication and defense. For example, while some
Orthoptera use hearing to warn them of approaching insectivorous bats (Moiseff et al.
1978), certain moths have acquired the additional ability to generate sounds for intra-
specific communication.
Although acoustical communication for pair forming is probably uncommon among
moths, recent research has exposed a number of demonstrated or suspected cases (See
review by Spangler 1988). Most moths are equipped with tympanic ears, which appa-
rently resulted from the selective pressure of echolocating insectivorous bats (Fenton
& Fullard 1981). The location of ears, that varies with moth group, can be on the head,
metathorax, first, second or seventh abdominal segment of the insect's body. These ears
provide moths with the hearing ability needed for ultrasonic communication. Two

Florida Entomologist 71(4)

pyralid moths, the greater wax moth, Galleria mellonella L. and the lesser wax moth,
Achroia grisella (F.), specifically consume the products of honey bees, Apis mellifera
L. Both moths have ears which allow them to defend against bats (Spangler and Takes-
sian 1983). Structurally similar to ears of other pyralid moths, these ears are located
on the first abdominal sternum (Coro 1973, Coro & Fernandez 1972, Mullen & Tsao
1971a, b). They are sensitive to a wide range of sound frequencies from ca. 20 kHz to
over 200 kHz (Spangler and Takessian 1983, Spangler 1984a). This hearing ability
suggested that both moths were capable of developing acoustical communication sys-
tems. In fact, A. grisella was the first moth discovered to have such a system (Dahm
et al. 1971, Spangler et al. 1984). However, acoustical communication is not unique to
these was moths. Within the subfamily Galleriinae, to which both wax moths belong,
acoustical communication may be widespread. Galleriinae is divided into three tribes:
Galleriini, Megarthridiini and Tirathabini and contains about 55 genera and 250 species
(Hampson 1917, Pajni & Rose 1977, Whalley 1964). Several members of this group
which feed on agricultural products are of great economic importance. The rice moth,
Corcyra cephalonica (Staint.) and Paralypsa gularis (Zeller) damage a variety of stored
products (Hodges 1979, Smith 1957). Three well-known species consume the stored
products of bees. Besides A. grisella and G. mellonella, Aphomia sociella (L.) feeds
on the products of bumble bee nests. Eldana saccharina Walker, the sugar cane borer,
is another well-known economic pest in this group (Sampson & Kumar 1983). In fact
intraspecific acoustical communication also appears to occur in a number of distant or
unrelated moth groups including Pyralidae, Agaristidae and Noctuidae (Gwynne and
Edwards 1986, Bailey 1978, Surlykke & Gogala 1986).
Moths within the Galleriinae subfamily have different sound-producing structures.
Although both A. grisella and G. mellonella have tymbals which produce single sound
pulses per wing-stroke (Spangler et al. 1984, Spangler 1985a), C. cephalonica has tym-
bals with nine striations, which buckle in sequence to produce up to 36 pulses of sound
per wingbeat (Spangler 1987b) (Fig. 1). C. cephalonica is in the tribe Tirathabini while
A. grisella and G. mellonella are in the tribe Galleriini. E. saccharina and A. sociella,
in the tribe Tirathabini, also have striated tymbals (Zagatti 1985). That tymbals or
other sound producing mechanisms occur only in male insects, suggests that an intra-
specific communication system is present. Only the males of most insects that communi-
cate with sound face the increased risk of parasites or predators attacking the sound
source (Burk 1982, Cade 1976).
Both A. Grisella and G. mellonella wax moths are distributed widely throughout
the tropical and temperate regions of the world. Because the larger, faster growing G.
mellonella seems to out-compete A. grisella in most areas of the United States, it
causes more damage. But where A. grisella predominates, in the extreme Pacific North-
west and at higher elevations, it also causes considerable damage to stored bee comb.
Initial studies on sound and wax moths involved attempts to prevent the females from
ovipositing near bee colonies and stored bee equipment by triggering their defensive
behavior with simulated cries of echolocating bats (Spangler 1984c). However, like
similar previous studies on different moth species (Belton & Kempster 1962, Agee &
Webb 1969), the moth populations were affected, but habituation to the sound prevented
any efficacious control. More recent discoveries that these moths use ultrasound as a
component of their mate-calling systems suggests new possibilities for detection and
control. However, the role of sound in pair forming behavior is different in A. grisella
and G. mellonella. A. grisella males use sound as the primary signal to call females.
Male G. mellonella, in contrast, use sound to stimulate females to fan their wings, an
action which in turn stimulates males to emit pheromone, the primary signal which
attracts females (Spangler 1985a, 1987a).


December, 1988

Spangler: Symposium on Agroacoustics

Fig. 1. Tymbals of four species of galleriine moths: A. Achroia grisella (by S. L.
3uchmann); B. Galleria mellonella; C. Aphomia sociella (from Zagatti 1985); D. Cor-
:yra cephalonica.

Different Roles of Sound in the Pair-Forming Behavior of Two Wax Moth Species

Achroia grisella. Each tegula of male A. grisella bears an anteriorly situated tymbal
Fig. 1). The tegular-wing coupler, a blade like structure, attaches to the lower, distal
order of the tegula, directly below the tymbal (Spangler et al. 1984). When pushed


pi -+r~l'JSt&h~-~~



470 Florida Entomologist 71(4) December, 1988

down by the wing, the coupler causes the tymbal to snap inward (buckle), simultane-
ously producing an ultrasonic pulse (Spangler and Takessian 1986). Release of pressure
on the coupler caused by the wing moving upward, allows the tymbal to snap outward
producing a second ultrasonic pulse. Stroboscopic analysis not only confirmed that the
tymbal snapped inward during the wing downstroke and outward during the upstroke,
but also revealed that the snapping action of the tymbal produced sound about midway
through the downstroke and upstroke (Spangler et al. 1984). Exactly how the insect
accomplishes this precise timing is not known. It appears, however, that moths produce
sound when they elevate the tegular process on the underside of the tegula with a
structure referred to as the "striker" on the costal margin of the forewing (Spangler et
al. 1984). An air chamber, enclosing not only the back side of the tymbal, but also most
of the underside of the tegula, may play a role in the positioning of the striker, although
its function is unknown.
Male A. grisella produce pulses of 100 kHz sound when calling, one pulse per up-
stroke and one pulse per downstroke. The wings, held posterior to the normal flying
position and fanned through about a 45 degree arc, also release a pheromone (Dahm et
al. 1971). The wing action produces a continuous series of sound pulses at a repetition
rate of 80-90/sec (Fig. 2). Males prefer calling in subdued light or darkness, but may be
found calling at any time under a variety of conditions.
The function of the sound is clearly to call potential mates. In tests using simulated
male sound (short pulses of 72 kHz sound at 80/sec), 15 of 21 unmated females ran to
the transducer; 13 fanned their wings. No females moved toward muted, pheromone-
producing males at the opposite end of the arena. In another test, 16 of 20 unmated
females arrived at the sound source plus females, while only one arrived at the muted
males plus silent transducer. In a third test, simulated sound plus females was equally
attractive to unmated females as sound plus muted males (Spangler et al. 1984). Addi-
tional tests over longer distances in greenhouses once again confirmed that simulated




N 64- il





0 12 24 36 48
Time (ms)

Fig. 2. Sonagram of sounds produced by two species of wax moths: upper trace =
Galleria mellonella; lower trace = Achroia grisella.

Spangler: Symposium on Agroacoustics

sound alone attracted 32% of the females and was much more effective than male
pheromone (6%) (Spangler 1984b).
Typically, A. grisella males assume calling positions near food (bee hive or stored
comb) where emerging females are expected to arrive. They stay near the culture if no
bees or only a few bees are present. However, if the bee colony is strong, both sexes
run from the hive immediately after emerging, before their wings have expanded.
Males outside the hive call from positions on the hive or on nearby vegetation (Green-
field & Coffelt, 1983). From either location, females, activated by male pheromone,
search until they locate the sound of a male, then run or fly directly to him (Spangler,
Galleria mellonella. Male G. mellonella also have sound-producing tymbals on their
tegulae which are activated when their forewings push down on the tegular-wing
coupler. The sound is released at the bottom of the wingstroke. As a downward moving
wing approaches the end of the stroke, it exerts pressure on the tegular-wing coupler,
which twists the distal area of the tegula and causes the tymbal to buckle in, producing
an ultrasonic pulse. Just as the wing starts back up, the pressure is released, the tymbal
snaps out, and a second pulse is produced. Since two tymbals buckle in and snap back
out, a train of two, three or four pulses of ultrasound may be released for each wingbeat,
depending on how synchronized the wings are in their movement (Spangler 1985a).
These pulse trains are produced by subsequent wing beats to form a series or phrase.
The number of pulse trains in a phrase ranged from 1 to 18 but averaged 4.03. When
many males are inside an enclosure such as a hive containing comb, but few or no living
bees, the ultrasonic pulses occur frequently. For example, in an abandoned outdoor bee
hive infested with G. mellonella, ultrasound production began about 11 minutes after
sunset and continued for 8 hours or more. isolatedd males still remained silent because
stimulation from other wax moths is requi -ed before they will produce sound (Spangler
G. mellonella males do not fan their wings continuously while they are calling, as
do A. grisella males. Instead, they extend their wings outward about 45 degrees from
their bodies (Flint & Merkle 1983, Spangler 1985a). When wings are in this position,
the upper wing surface is at the same height as the top of the moth's body. If no other
male or female G. mellonella is nearby, a male remains stationary while releasing
enough pheromone to attract females. However, if another male or female G. mellonella
approach the calling male, it may flutter its wings briefly and emit pulses of 75 kHz
The ultrasonic pulses produced by G. mellonella males are known to stimulate only
wing fanning in unmated females (Spangler 1985a). Although a female may run about,
she will not orient directly toward the sound source. Reactions of 100 virgin females in
groups of 5 to bursts of 72-kHz sound at 5-second intervals were tested. Eight moths
began wing fanning in response to the second sound burst at 5 seconds. Forty-five had
responded by 35 seconds; 57 by 1 minute; 77 by 2 minutes; and 83 by 3 minutes. Seven-
teen moths did not respond within the 3-min test period. The average threshold inten-
sity at which simulated male sounds caused virgin females to wing fan was 69 dB. The
frequency within the pulses could be made to vary widely and still elicit virgin female
response. Transducers resonating at 25 kHz and 150 kHz caused the females to wing
fan and emit low frequency sound with energy peaks at about 40 and 80 Hz (Spangler
Males in calling position are sensitive to low-frequency airborne sounds. A recent
study showed calling males to be highly sensitive to sound and vibrations from 35 to
100 Hz, a range which includes most of the sound energy released by a wing-fanning
female. When sound or vibration simulating the sound of a wing-fanning female was
directed toward a male, it responded by increasing the quantities of pheromone it

472 Florida Entomologist 71(4) December, 1988

released from wing glands (Spangler 1987a). Existence of a sexual pheromone in G.
mellonella, known for some time (Leyrer & Monroe 1973), has been demonstrated to
attract females (Finn & Payne 1977).
Males move about and produce ultrasound (Spangler 1986a) and pheromone within
a cavity such as an abandoned bee hive. When a nearby female recognizes the sound
signal, she fans her wings and produces a low-frequency sound which causes the male
to sharply increase the release of pheromone so she can locate him. Outside of the hive,
G. mellonella males are unlikely to produce sound. However, they release a low-level
of pheromone probably sufficient to attract nearby females and they have been observed
producing ultrasound only when two males approach each other closely. When a male
hears the wingbeat sound of a nearby flying female he may flutter his wings to increase
his pheromone dispersal (Spangler 1985a).

Hearing Adaptations and Communication

Whether A. grisella females exhibit behaviors either for defending themselves
against insectivorous bats or responding to calling males depends, in part, on the pulse
repetition rate of the sound they hear. Most females show a sexual attraction above 40
pulses/sec and apparent defensive response below 30 pulses/sec (Spangler et al. 1984).
In contrast, while female G. mellonella wing fan in response to a single pulse and to
repeated pulses with a repetition rate up to about 117/sec, this behavior normally takes
place within the protection of the beehive (Spangler 1985a).
A. grisella females can orient, then run or fly to the sound of a calling male from
distances of about 1 m (Spangler 1984b, Spangler et al. 1984). They lose ability to orient
if one ear is deafened by tearing the tympanum (Spangler & Hippenmeyer 1988). Al-
though they may run and circle in response to sound, they are unlikely to find the
source. Although no mechanism for distinguishing sound frequencies, as in Orthoptera
(Michelsen 1979), has been identified in any moth ear (Miller 1983, Roeder 1967). Some
evidence indicates that G. mellonella can distinguish different frequencies because they
do not appear to respond defensively when subjected to lower frequency ultrasound of
moderate intensity. They do respond with typical defensive behavior at higher frequen-
cies (Spangler 1984a). In contrast, simulations of the very short sound pulses used for
G. mellonella sexual communication affect the female by causing her to wing fan at low
intensities throughout her hearing range (Spangler 1985a).

The Acoustical Environment of Wax Moths

Wax moths of both species usually occur in situations where there are few or no
honey bees (the colony has died, abandoned the hive or the comb has been removed
from the hive). However the moths do emerge and exist in active bee colonies and
females will enter colonies to oviposit. Bees defend their hives against the moths by
evicting both larvae and adults in ways that suggest elaborations of existing cleaning
behavior (Clark 1984).
Bees produce a wide variety of sounds, some of which are in the hearing range of
wax moth ears. However, at the low end of the frequency spectrum there does not
appear to be much opportunity for acoustical interactions (Fig. 3). Male greater wax
moths respond to airborne sounds from a wing-fanning female at the 40 Hz wingbeat
frequency and the second harmonic at 80 Hz (Spangler 1987a). However, they do not
respond much at the lowest frequencies produced by wing-fanning bees. In fact, bees
produce very little sound energy in the 30-100 Hz range which might interfere with the
communication system of greater wax moths.

Spangler: Symposium on Agroacoustics

Sounds and Vibrations



", ..quacking) Stopping
)ueen piping a Stopping
(quacking) -450-
Queen piping

Worker piping -350-

Ventilating S -300

ngbees Dance
Flying bees Dsounds Scent


-100 G mel//one/lla
Golleria mellonella > pheromone
2x release
Sme//one/o 50 behavior
G. mellonella
winbeat (Apis cerona)
Waggles, D-VAV

Fig. 3. The low-frequency sounds which occur around a bee colony, including those
produced and received by the Galleria mellonella. Only ultrasound is important for the
Achroia grisella.

Florida Entomologist 71(4)

A. gr/sello

Ap/s G. me//one//a
me//ifera response

Fig. 4. The repetition rate of ultrasonic pulses produced by wing-fanning bees com-
pared to the response range of two species of wax moths to similar pulses. The open
area represents lower frequency wing-fanning by standing bees while the crosshatched
area represents two pulses per wingbeat.



( 400





. 200



December, 1988


Spangler: Symposium on Agroacoustics

On the other hand bees do produce a broad range of ultrasonic frequencies (Spangler
1986b). This sound is in the form of pulses produced when bees fan their wings, either
while stationary or when flying. The pulses have their highest intensity from about 25
to 50 kHz. Generally there is one loud pulse per wing beat, although sometimes there
is a second pulse of lesser intensity. The pulse repetition rate is typically above the
frequency that female greater wax moths responded to by fanning their wings (Fig. 4).
However, it seems possible that wax moths obtain some information from the bee sound
or respond to it defensively as if a bat were attacking. It is also possible that moths are
able to use the ultrasound produced by wing-fanning bees in locating and/or assessing
the condition of bee colonies to allow females to decide whether to enter the hive or to
oviposit near an opening into the hive.
Although generally honey bees are not sensitive to most airborne sounds and are
unlikely to hear any ultrasound, Apis cerana detect 50 Hz sound, probably from vibra-
tions on the comb, then respond by hissing (Fuchs & Koeniger 1974). Bee hissing,
produced by many bees fanning their wings, is rich in almost continuous ultrasound
(Spangler 1986b). Thus, it would seem that bees, particularly when nesting outside of
cavities, might be able to detect approaching wax moths and disrupt them by triggering
their defensive responses. However, no conclusive evidence has established that the
sound from hissing bees affects any insect (Fuchs & Koeniger 1974). Perhaps this is
because G. mellonella has developed the ability to limit its defensive responses to
bee-produced sound (Spangler 1984a).


Female A. grisella has a sophisticated system for determining the pulse repetition
rate of incoming sound to distinguish male moths from bats. Pheromone released by the
wing glands of calling males may cause females to search, although they are unlikely to
find males without the acoustical signals.
Acoustical techniques have already proven useful in problem situations caused by
A. grisella. Electronic devices designed to detect male sound allow quick inspection of
comb storage facilities for the presence of adult moths (Spangler 1985b). Acoustically
baited traps, designed to monitor for the presence of moths in apiaries or storage
facilities, may also be possible (Spangler 1984b).
If the intensity of simulated male sound is increased above that of male-produced
sound, these acoustically-baited traps might outcompete calling males in attracting
enough unmated females from greater distances to reduce population levels.
The sexual signalling system of G. mellonella, which includes chemical, ultrasonic
and low-frequency sound signals, allows it to function efficiently both inside and outside
enclosures. Because all of these signals seem to operate at a distance of less than one
meter, G. mellonella's communication system is short range. By concentrating their
mating activities on or near bee hives, these moths have no need for a long-range sexual
signalling system. Because G. mellonella produce sound sporadically, the techniques
suggested for monitoring, detecting and controlling A. grisella males using acoustical
equipment seem unlikely to work with G. mellonella. While it is possible to locate A.
grisella males with an electronic detector, isolated male G. mellonella, which don't
produce ultrasound, cannot be detected by sound until a number of males are present.
By that time one can usually detect the odor of their pheromone.
Understanding the biology of these bee pests may lead to yet undiscovered control
techniques. If, for example, the odor and/or acoustical signals that attract wax moths
to bee colonies could be defined, then appropriate methods to lure the moths into traps
could probably be devised.

Florida Entomologist 71(4)


AGEE, H., AND J. C. WEBB. 1969. Effects of ultrasound on capture of Heliothis zea
and Ostrinia nubilalis moths in traps equipped with ultraviolet lamps. Ann.
Entomol. Soc. Amer. 62: 1248-52.
BAILEY, W. J. 1978. Resonant wing systems in the Australian whistling moth
Hecatesia (Agarasidae, Lepidoptera). Nature 272: 444-46.
BELTON, P., AND R. H. KEMPSTER. 1962. A field test on the use of sound to repel
the European corn borer. Entomol. Exp. Appl. 5: 281-88.
BURK, T. 1982. Evolutionary significance of predation on sexually signalling males.
Florida Entomol. 65: 90-104.
CADE, W. 1976. Acoustically orienting parasitoids: fly phonotaxis to cricket song.
Science 190: 1312-13.
CLARK, N. L. 1984. The relationship between the greater wax moth, Galleria mel-
lonella L., and the honey bee, Apis mellifera L. M.S. Thesis, Ariz. State Univ.
62 p.
CORO, F. 1973. Morfologia del 6rgano timpAnico de la polilla del arroz, Corcyra
cephalonica (Stainton) (Lepidoptera: Galleriidae). Cien. Biol. 45: 1-15.
CORO, F., AND A. FERNANDEZ. 1972. Estructura del 6rgano timpAnico de un com-
binado: Diatraea saccharalis (Fabr.), (Lepidoptera: Crambidae). Revista Cenic 4:
olfactory and auditory mediated sex attraction in Achroia grisella (Fabr.).
Naturw. 58: 265-266.
FENTON, M. B., AND J. H. FULLARD. 1981. Moth hearing and the feeding strategies
of bats. Am. Sci. 69: 266-275.
FINN, W. E., AND T. L. PAYNE. 1977. Attraction of greater wax moth females to
male-produced pheromones. Southwest. Entomol. 2: 62-65.
FLINT, H. M., AND J. R. MERKLE. 1983. Mating behavior, sex pheromone responses
and radiation sterilization of the greater wax moth (Lepidptera: Pyralidae). J.
Econ. Entomol. 76: 467-472.
FUCHS, S., AND N. KOENIGER. 1974. Sound production as a means of defense for
the honey-bee colony (Apis cerana Fabr.). Apidologie 5: 271-278.
GREENFIELD, M. D., AND J. A. COFFELT. 1983. Reproductive behaviour of the
lesser wax moth, Achroia grisella (Pyralidae: Galleriinae): signalling, pair forma-
tion, male interactions, and mate guarding. Behaviour 84: 287-315.
GWYNNE, D. T., AND E. D. EDWARDS. 1986. Ultrasound production by genital
stridulation in Syntonarcha iriastis (Lepidoptera: Pyralidae): long distance sig-
nalling by male moths? Zool. J. Linn. Soc. 88: 363-76.
HAMPSON, G. F. 1917. A classification of the Pyralidae, subfamily Gallerianae. Novit.
Zool. 24: 17-58.
HODGES, R. J. 1979. A review of the biology and control of the rice moth Corcyra
cephalonica Stainton (Lepidoptera: Galleriinae). Tropical Products Inst. Pub.
G125. 20 p.
LEYRER, R. L., AND R. E. MONROE. 1973. Isolation and identification of the scent
of the moth Galleria mellonella, and a reevaluation of its sex pheromone. J.
Insect Physiol. 19: 2267-2271.
MICHELSEN, A. 1979. Insect ears as mechanical systems. Am. Sci. 67: 696-706.
MILLER, L. A. 1983. How insects detect and avoid bats. Pages 251-266, in Huber, F.
and H. Markl (eds.), Neuroethology and Behavioral Physiology, Springer-Ver-
lag, Berlin.
MOISEFF, A., G. S. POLLACK, AND R. R. HOY. 1978. Steering responses of flying
crickets to sound and ultrasound: mate attraction and predator avoidance. Proc.
Nat'l. Acad. Sci., USA, 75: 4052-4056.
MULLEN, M. A., AND C. H. TSAO. 1971a. Morphology of the tympanic organ of the
greater wax moth, Galleria mellonella. J. Georgia Entomol. Soc. 6: 124-132.
MULLEN, M. A., AND C. H. TSAO. 1971b. Tympanic organ of Indian meal moth,
Plodia interpunctella (Hfibner), almond moth, Cadra cautella (Walker) and to-


December, 1988

Spangler: Symposium on Agroacoustics

bacco moth, Ephestia elutella (Hfibner) (Lepidoptera: Pyralidae). Int. J. Insect
Morphol. and Embryol. 1: 3-10.
PAJNI, H. R., AND H. S. ROSE. 1977. Descriptions of a new genus Aswania
(Lepidoptera: Galleriidae) from north-west India with comments on this family.
J. Entomol. Res. 1: 13-17.
ROEDER, K. D. 1967. Nerve cells and insect behavior, Revised edition. Harvard
Univ. Press, Cambridge, MA 238 p.
SAMPSON, M. A., AND R. KUMAR. 1983. Population dynamics of the stem-borer
complex on sugar-cane in southern Ghana. Insect Sci. Applic. 4: 25-32.
SMITH, K. G. 1957. The occurrence and distribution of Aphomia gularis (Zell.) (Lep.,
Galleriidae), a pest of stored products. Bull. Entomol. Res. 47: 655-667.
SPANGLER, H. G. 1984a. Responses of the greater wax moth, Galleria mellonella L.
(Lepidoptera: Pyralidae) to continuous high-frequency sound. J. Kansas En-
tomol. Soc. 57: 44-49.
SPANGLER, H. G. 1984b. Attraction of female lesser wax moths (Lepidoptera:
Pyralidae) to male-produced and artificial sounds. J. Econ. Entomol. 77: 346-349.
SPANGLER, H. G. 1984c. How lesser wax moths find their mates. Glean. Bee Cult.
112: 467-470.
SPANGLER, H. G. 1985a. Sound production and communication by the greater wax
moth (Lepidoptera: Pyralidae). Ann. Entomol. Soc. Amer. 78: 54-61.
SPANGLER, H. G. 1985b. Detecting lesser wax moths acoustically. Glean. Bee Cult.
113: 207-209, 218.
SPANGLER, H. G. 1986a. Functional and temporal analysis of sound production in
Galleria mellonella (Lepidoptera: Pyralidae). J. Comp. Physiol. A 159: 751-756.
SPANGLER, H. G. 1986b. High-frequency sound production by honey bees. J. Apic.
Res. 25: 213-219.
SPANGLER, H. G. 1987a. Acoustically mediated pheromone release in Galleria mel-
lonella (Lepidoptera: Pyralidae). J. Insect Physiol. 33: 465-468.
SPANGLER, H. G. 1987b. Ultrasonic communication in Corcyra cepalonica (Stainton)
(Lepidoptera: Pyralidae). J. Stored Prod. Res. 23: 203-211.
SPANGLER, H. G. 1988. Moth hearing, defense and communication. Annu. Rev. En-
tomol. 33: 59-81.
SPANGLER, H. G., M. D. GREENFIELD, AND A. TAKESSIAN. 1984. Ultrasonic mate
calling in the lesser wax moth. Physiol. Entomol. 9: 87-95.
SPANGLER, H. G., AND C. L. HIPPENMEYER. 1988. Binaural phonotaxis in the
lesser wax moth, Achroia grissella (F.) (Lepidoptera: Pyralidae). J. Insect
Behav. 1: 117-122.
SPANGLER, H. G., AND A. TAKESSIAN. 1983. Sound perception by two species of
wax moths. Ann. Entomol. Soc. Amer. 76: 94-97.
SPANGLER, H. G., AND A. TAKESSIAN. 1986. Further observations on sound produc-
tion by the lesser wax moth, Achroia grisella (F.) (Lepidoptera: Pyralidae). J.
Kans. Entomol. Soc. 59: 555-557.
SURLYKKE, A., AND M. GOGALA. 1986. Stridulation and hearing in the noctuid moth
Thecophora fovea (Tr.). J. Comp. Physiol. A. 159: 267-273.
WHALLEY, P. E. S. 1964. Catalog of Galleriinae (Lepidoptera, Pyralidae) with de-
scriptions of new genera and species. Acta. Zool. Cracoviensia 10: 561-618.
ZAGATTI, P. 1985. Approche evolutine du comportement precopulatoire chez les
L6pidopteres Ditrysiens. State Thesis. Univ. Pierre et Marie Curie, Paris, 191


4m -


.,".. 'm

Florida Entomologist 71(4)


'Insect Attractants, Behavior, and Basic Biology
Research Laboratory, Agricultural Research Service,
U.S. Department of Agriculture, Gainesville, Florida 32604
2Grain Marketing Research Laboratory
Agricultural Research Service, U.S. Department of Agriculture
1515 College, Av., Manhattan, Kansas 66502


A small, easily constructed sound-insulated room suitable for use in acoustic detec-
tion of insect larvae in stored commodities is described. Sound measurements are pre-
sented to quantitate the reduction of exogenous noise. The directionality of the acousti-
cal detector used for detection of Sitophilus oryzae larvae feeding inside kernels of grain
was determined. "Dead spots" were located and suggestions for the design of a sample
holder are given which will minimize this problem.


Se describe un cuarto pequeno, insulado a prueba de ruido, y facil de construir, para
uso en la detecci6n acustica de larvas de insects en cultivos almacenados. Se presentan
medidas del sonido para cuantificar la reducci6n de ruidos externos. Se determine la
direccionalidad del detector acustico usado para detectar larvas de Sitophilus oryzae
comiendo dentro de los granos. Se localizaron "lugares muertos" y se hacen sugerencias
para el disefo de un porta-muestra que reducird el problema.

Recent interest in detecting internally feeding insects in kernels of grain has cen-
tered on the use of acoustical methods. Although the electronic filtering used by Webb
et al. (1988), Vick et al. (1988) and Hagstrum et al. (1988) reduces the ambient and
system noise to some extent, the high amplification required to detect the low-power
sounds that feeding larvae produce requires that the grain sample be isolated from
ambient noise. Although an anechoic chamber would be the ideal environment to use
with these detector systems, this often is impractical given constraints of budget and
space common at most laboratories. We describe a small sound-insulated room which is
easily constructed, inexpensive and suitable for use for insect detection using acoustical
The ideal acoustic detector should be one which detects sounds equally well from all
directions. Our preliminary data indicated that our detectors had some degree of direc-
tionality. This problem of directionality and consequent dead space in the grain sample
holder can be minimized with certain grain sample holder designs provided the direction-
ality of the transducer is known. We present data on this directionality for one type of
detector used in this laboratory and suggest a sample holder design to minimize direc-
tionality effects for this detector.

December, 1988

Vick et al.: Symposium on Agroacoustics


Fig. 1. A. Outline of sound-insulated room. B. Schematic representation of a wall
section to show construction details.


Sitophilus oryzae (L.), the rice weevil, (laboratory stock in culture at this laboratory
for 19 years) was reared on wheat incubated at 25 + 1C and 65% RH 5% with a 14:10


480 Florida Entomologist 71(4) December, 1988

(L:D) photoperiod. Adults were allowed to oviposit on uninfested wheat for 7 days and
then were removed. Individual kernels were checked for sounds using the acoustical
detector system described by Webb et al. (1988) to identify kernels containing larvae
17 days after the adults were removed.
The sound-insulated room was constructed in a corner of our laboratory (Fig. 1).
Sound control board (1/2 inch thick, Cellotex Co.) was cut to the size and shape of the
floor of the proposed room and laid directly on the concrete floor of the laboratory. A
frame for the floor was constructed of pine 2-inch X 4-inch wood (ca. 4 cm X 9 cm) and
laid on the sound control board. The floor of the room was constructed of 3/4-inch (ca.
1.8 cm) thick plywood attached to the upper side of the 2 inch X 4 inch frame with nails
and glue. The sound-insulated room used 2 existing laboratory walls for its back and
right side wall. To these, sound control board was attached with glue. The left side wall
(LSW) as well as the front wall (FW) were framed with 2 inch X 4 inch lumber on 16
inch centers (40.5 cm) using standard construction techniques. Three-inch thick
fiberglass bats were placed between the studs. The LSW and FW walls were faced
inside and out with sound control board. The door was a 28-inch (71 cm), hollow-core
veneered door typically used for interior doors in houses. The ceiling was constructed
of a 2-inch X 4-inch lumber frame covered inside and out with sound control board. A
medium length shag carpet was installed on the floor. The walls and ceiling and door
were then covered inside and out with 3-inch thick (ca. 7.5 cm) Sonex acoustical foam
(Illbruck Co., 3800 Washington Avenue No., Minneapolis, MN 55412) except that the
2 walls shared with the laboratory were covered only on the inside.
A work bench was constructed in the sound insulated room by stacking 6 standard
concrete building blocks into two, three-block high columns (61 cm high) upon which
was placed a concrete slab (66 X 40.5 X 5 cm thick) which served as the work surface.
The slab was covered with medium length shag carpet. A sound insulated box, 56 cm
deep X 54 cm wide X 64 cm tall, constructed of 3/4 inch plywood and lined with 3/4-inch
sound control board was placed on the work surface. Access to the box was gained
through a hinged door measuring 38 cm wide X 51 cm tall located on the side of the
box. The entire box including the door was lined with sound insulation foam. The sample
holder was suspended in this box by rubber bands to reduce transmission of building
vibrations to the sample holder and microphone.
The sound detection and amplification system used here was described in Vick et al.
(1988). The filter, signal processing, and computer system were located outside the
sound-insulated room. The sound insulation qualities of the room were tested by feeding
a 1 kHz signal from an oscillator into a speaker placed 2 meters in front of the room
door at a height of 1.5 meters. The sound in the room was measured with a 1-inch
condenser microphone, Bruel & Kjaer (B&K) model 4145, and a microphone amplifier,
B&K model 2610. The grain sample holder was made from a copper pail (ca. 0.5 mm
wall thickness) with the top slightly larger than the bottom and having the following
dimensions: 11 X 9.5 X 13.5 cm, respectively for the diameter of the top and bottom
and the height of the sides with the sides tapering towards the bottom. The angle
formed by the meeting of the sides and bottom was 93 degrees. A 33-mm diam hole was
bored at the center of the bottom of the grain sample holder into which was fitted the
plastic diaphragm of the acoustic coupler.
Five wheat kernels, each infested with a 17-26 day-old rice weevil larvae, were
placed in a small mesh pouch (ca. 1 X 1 cm with a thickness of one wheat kernel) which
was used as the sound source for measuring the effect of distance and angle on sound
detection. The same 5 kernels were used throughout the 3 days of testing. The bottom
of the sample container was covered with a 1 cm layer of uninfested wheat. The sound
source was placed in the middle of the container on top of the layer of wheat and the
detectable sounds emanating from the sound source were counted for 2 minutes. The

Vick et al.: Symposium on Agroacoustics

sound source was moved to a position 2.4 cm from the center towards the edge of the
container and the sounds counted as before. A third count was taken at a position 4.6
cm from the center. The counts thus were taken at the center of the container (im-
mediately above the diaphragm) (position A), at a point ca. midway between the middle
of the container (position B) and the edge and at a point near the edge (position C).
These 3 counts were taken in random order. The voltage spike counts at the 3 horizontal
locations were taken at 1, 2, 3, 4, and 5 cm depths, also, in random order. Measurements
were replicated 8 times per position.


The sound-insulated room, the sound-insulated box inside the room and even the
grain in the grain holder are components of an exogenous noise reduction system. To
determine the contributions of the various parts of this system to overall noise reduc-
tion, sound levels were measured at various locations inside and outside of the sound-in-
sulated room (Fig. 2). The sound level at the front of the door leading into the sound-in-
sulated room (2 meters from the sound generating oscillator) was 67 dB. The sound
level at the microphone in a full grain holder inside the sound-insulated box positioned
inside the sound-insulated room was 13 dB for a total reduction of 54 dB. Since each 6
dB reduction represents a 2X reduction in sound power (volume), the entire noise
suppression system reduces exogenous noise to a level ca. 0.15% of its original power.
The exogenous noise suppression afforded by this room when used with a sound-in-
sulated box and the present grain sample holder would be sufficient for most larval


Outside Chamber _Inside Chamber
67.0 dB (door open)
47.0 dB

Plywood Box
(door open)
37.5 dB


(door closed)
40.0 dB

(door closed) Grain Holder
25.0 dB (Empty)
18.5 dB

Groin Holder
CFu I ) <
13.0 dB
Fig. 2. Flow chart showing effect of each component of the exogenous noise reduction
system on the reduction of background noise.

Florida Entomologist 71(4)


Horizontal distance from center (cm)
Depth (cm) 0 2.4 4.6

1 4640 (1982)' 312 (474) 4 (7)
2 1635 (648) 180 (181) 7(14)
3 1174(1307) 101(117) 7(11)
4 814 (1022) 88 (105) 23(28)
5 154 (174) 95(114) 22(27)

'Number of larval sounds as determined by number of voltage spikes

detection needs. Routine use of this room in our laboratory has shown that normal
laboratory work, including the normal operation of laboratory equipment (refrigerators,
freezers, etc.) and conversation, can be carried out in the laboratory without interfering
with sound counts by the sound detection system in the insulated room. Problems might
develop if the amplification had to be set at very high levels for some reason (i.e. to
detect very young larvae). Extensive testing of the system with uninfested grain in the
sample holder did yield rare spurious sound spikes which seemed to be attributable to
building vibrations possibly made worse by the fact that this room is located on the
second floor. In our situation these spurious sound spikes were sufficiently rare that
they were easily identified as spurious sound spikes by the normal replication of the
experiments. In cases where building vibrations are common enough to be troublesome,
the problem might be solved by suspending the sound-insulated box from the ceiling of
the sound-insulated room by elastic bands.
The extent that sounds could be detected was both a function of distance of the sound
source from the diaphragm and angle of the sound source to a vertical perpendicular
line in the center of the grain mass of the sample holder. Although the number of sounds
detected decreased as the sound source was moved away from the acoustic coupler
(Table 1), the distance of the sound source from the detector was not as important in
this regard as the angle of the detector in relation to the sound source. The greatest
sensitivity occurred when the sound source was at 0 degree incidence to the detector
angle and decreased as the angle increased. This effect is especially apparent for the
results at position C (4.6 cm, Table 1). As the depth increased from 1 to 5 cm, the
distance from the sound source to the acoustic coupler increased but the angle de-
creased. The increase in sound counts with decreasing angle (even though the distance
increased) indicates that in sample holder design, one of the top priorities should be the
elimination of "dead spots", even if the design change would place some grain at a
greater distance from the detector.
The log of the number of voltage spikes (Table 1) was plotted against the distance
of the infested kernels from a zero degree incidence perpendicular line running from
the detector up into the grain mass of the sample holder. Regression lines (not shown)
were eye-fitted for grain depths of 1, 2, 3, 4, and 5 m. Thus from the resulting 5
regression lines one could estimate for each of the grain depths the number of sounds
expected when the sound source was placed at positions A, B, or C (0, 2.4 and 4.6 cm,
respectively, from Table 1) or at any position in between. Conversely, one could esti-
mate at what point between position A and the sample holder wall that a given number
of sounds would be detected.
Figure 3 illustrates the effect of detector directionality with isobarss" drawn for
1500, 1000, 100 and 30 counts expected from our standard insect sound source. A con-

December, 1988


Vick et al.: Symposium on Agroacoustics







Fig. 3. Schematic diagram of the grain sample holder used in this study. A constant
sound source when placed anywhere on each of the lines would yield sound counts
equivalent to any other point on the same line.

stant sound source would give a constant number of voltage spike counts at any point
on each of the lines shown. This figure dramatically illustrates the need to eliminate the
"dead spot" formed at the angle where the bottom and sides of the container join. A
more satisfactory shape for the sample holder would be a cone shaped vessel that
eliminates the acute bottom to side angle, even if the total depth of the container had
to be increased to hold the same quantity of grain sample.

Florida Entomologist 71(4)


HAGSTRUM, D. W., J. C. WEBB, AND K. W. VICK. 1988. Acoustical detection and
estimation of Rhyzopertha dominica (F.) larval population in stored wheat.
Florida Entomol. 71: 441-447.
VICK, K. W., J. C. WEBB, B. A. WEAVER, AND K. LITZKOW. 1988. Sound-detection
of stored-product insects that feed inside kernels of grain. J. Econ. Entomol. 81:
WEBB, J. C., C. A. LITZKOW, AND D. C. SLAUGHTER. 1988. A computerized acous-
tical larval detection system. Applied Engineering in Agriculture. 4: 268-274.

a t a-- ---a -- -a- -


University of Florida
Gainesville, Florida 32611-0143


Development of sound-baited traps for insects has lagged behind that of light- and
chemical-baited traps. The principal successes for acoustic traps have been with mole
crickets (Gryllotalpidae), field crickets (Gryllidae), and ormiine flies (Tachinidae). The
crickets are attracted to the conspecific calling song and the flies to the calling songs of
their hosts. Electronic sound synthesizers facilitate routine operation of acoustic traps,
and increasing the intensity of the sound far above the levels of the natural call greatly
increases the numbers trapped. Acoustic traps are most likely to be useful for species
that exhibit long-range phonotaxis under natural conditions. Acoustic traps are unlikely
to be cost-effective for control but have proved valuable in studying behavior and ecol-
ogy, collecting specimens, and monitoring populations.


El desarrollo de trampas cebadas con sonido esta mas atrasado que trampas de luz
o cebadas con products quimicos. El 6xito principal de trampas acisticas ha sido con
los topogrillos (Grillotalpida), grills de campo (Grillida) y con moscas Taquinidas. Los
grills son atraidos a los cantos coespecificos y las moscas al canto de su hospedero. Los
sintetizadores electr6nicos de sonido facilitan la rutina de la operaci6n de trampas acuis-
ticas, y aumentan la intensidad del sonido much mas que los niveles del llamado natural,
aumentando el nfimero de atrapados. Es mas probable que trampas acusticas sean mas
tiles para species que demuestren una fonotaxis de largo alcance bajo condiciones
naturales. Es improbable que el costo de trampas acusticas valga la pena, pero han
demostrado ser valiosas en studios ecol6gicos y de comportamiento, en la colecci6n de
muestras, y en el chequeo de poblaciones.

Traps baited with lights or chemicals are widely used to collect insects and to monitor
their populations. Light traps catch a wide variety of insects that generally are attracted
in small numbers; why light attracts insects and from how far are poorly understood
(but see Baker 1985). On the other hand, chemical traps generally trap one or a few


December, 1988

Walker: Symposium on Agroacoustics

kinds of insects, frequently in large numbers; this is because chemical baits simulate
specific sex pheromones or chemicals released by particular foods. Individuals are often
attracted from 100 m or more. Acoustic traps are similar to chemical traps in that they
generally catch one or a few species, often in large numbers. Acoustic baits simulate
the mating call of the captured species or of the prey or host species.
In this paper I review the development and use of sound-baited insect traps, compare
the principal components of successful traps, and consider potential uses and limitations
of acoustic traps for agricultural insects.


Kahn & Offenhauser (1949; also Offenhauser & Kahn 1949) were apparently first to
field test a sound-baited trap. In a swamp in Cuba, they used a recording of the flight
sound of an Anopheles albimanus female to attract mosquitoes to a high-voltage elec-
trified screen. Their equipment was crude by today's standards-they played a 78 rpm
acetate disk on a record changer. Nonetheless, they killed more mosquitoes in the peak
10-minute interval of trapping than a nearby cattle-baited trap caught in a week. Be-
cause phonotaxis in mosquitoes is mainly a matter of males seeking a mate by homing
on the female's flight sound, the mosquitoes Offenhauser & Kahn killed were principally
males-rather than blood-seeking, disease-carrying, egg-laying females. In spite of this
limitation, Belton (1967) and, more recently, Ikeshoji and co-workers (Ikeshoji et al.
1985, Ikeshoji 1986, Ikeshoji & Yap 1987) further developed and field-tested sound
traps for mosquitoes. Such traps could reduce mosquito populations by reducing fertility
of females, either by removing or chemically sterilizing attracted males.
The first acoustic traps developed for agricultural insects were for the mole crickets,
Scapteriscus acletus and S. vicinus, which are important pests of pastures and crops
in the southeastern United States (Ulagaraj & Walker 1973, Walker 1982). These traps
(e.g., Fig. 1) broadcast the real or imitation calling song of the male and attract and
catch flying mole crickets of both sexes. A standard trapping station, consisting of one
S. acletus trap and one S. vicinus trap, generally yields thousands of mole crickets in
a year; catches of hundreds during one evening are not uncommon. The record catch of
S. acletus for one station in one night is 3,297; for S. vicinus, 2,174 (Walker 1982, and
unpublished). Forrest (1983a) and Chukanov & Zhantiev (1987) trapped mole crickets
of other species (Scapteriscus spp. and Gryllotalpa spp.) that flew or walked to repro-
ductions of their calling songs.
A major use of acoustic traps for mole crickets has been to acquire living material
for research. Adult mole crickets are exceedingly difficult to collect by other means,
and large-scale laboratory rearing has thus far proved impractical. Sound trapping has
made possible much of the research on the biology and on chemical and biological control
of Scapteriscus acletus and S. vicinus (Walker 1984).
Field crickets are a second group of agriculturally important insects that have been
caught with sound-baited traps. Campbell & Shipp (1974, 1979) and Campbell (ms. in
review) developed traps for Teleogryllus commodus, an important pest of pastures in
Australia and New Zealand. As in mole crickets, the bait was the natural or synthesized
male calling song, and both males and females flew or walked into the trap. In North
America, field crickets of the genus Gryllus have been acoustically trapped by Cade
(1979, 1981) and Walker (1986). Sound trapping field crickets has contributed to studies
of their migratory and mating behaviors (Campbell & Shipp 1979, Cade 1981, Walker
1987). It can also provide a ready source of live crickets for laboratory studies or for
feeding animals: for three years, a trap broadcasting G. rubens song at Gainesville,
Florida, caught hundreds of G. rubens most months and an annual average of 8,209
(Walker 1986).


University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs