J Comp Physiol A (2008) 194:597­609
DOI 10.1007/s00359-008-0344-0
123
REVIEW
Otoacoustic emissions from insect ears: evidence of active
hearing?
Manfred Kössl  Doreen Möckel  Melanie Weber 
Ernst-August Seyfarth
Received: 3 February 2008 / Revised: 23 April 2008 / Accepted: 3 May 2008 / Published online: 31 May 2008
 Springer-Verlag 2008
Abstract Sensitive hearing organs often employ nonlinear
mechanical sound processing which generates distortion-product
otoacoustic emissions (DPOAE). Such
emissions are also recordable from tympanal organs of
insects. In vertebrates (including humans), otoacoustic
emissions are considered by-products of active sound
ampliWcation through specialized sensory receptor cells in
the inner ear. Force generated by these cells primarily augments
the displacement amplitude of the basilar membrane
and thus increases auditory sensitivity. As in vertebrates,
the emissions from insect ears are based on nonlinear
mechanical properties of the sense organ. Apparently, to
achieve maximum sensitivity, convergent evolutionary
principles have been realized in the micromechanics of
these hearing organs--although vertebrates and insects
possess quite diVerent types of receptor cells in their ears.
Just as in vertebrates, otoacoustic emissions from insects
ears are vulnerable and depend on an intact metabolism, but
so far in tympanal organs, it is not clear if auditory nonlinearity
is achieved by active motility of the sensory neurons
or if passive cellular characteristics cause the nonlinear
behavior. In the antennal ears of Xies and mosquitoes, however,
active vibrations of the Xagellum have been demonstrated.
Our review concentrates on experiments studying
the tympanal organs of grasshoppers and moths; we show
that their otoacoustic emissions are produced in a frequency-speciWc
way and can be modiWed by electrical stimulation
of the sensory cells. Even the simple ears of
notodontid moths produce distinct emissions, although they
have just one auditory neuron. At present it is still uncertain,
both in vertebrates and in insects, if the nonlinear
ampliWcation so essential for sensitive sound processing is
primarily due to motility of the somata of specialized sensory
cells or to active movement of their (stereo-)cilia. We
anticipate that further experiments with the relatively simple
ears of insects will help answer these questions.
Keywords Hearing organs  Tympanal organ 
Nonlinear sound processing  Electromotility
Abbreviations
OAE Otoacoustic emission
DPOAE Distortion-product otoacoustic emission
SOAE Spontaneous otoacoustic emission
OHC Outer hair cell
IHC Inner hair cell
SPL Sound pressure level
TRP Transient receptor potential
Ears produce mechanical energy
Sense organs are comprised of sensory neurons and accessory
cells that convert speciWc stimulus energy into changes
in membrane potential. This conversion, termed sensory
transduction, is based either on intracellular reception of the
stimulus (as in photoreceptors) or on direct activation of
membrane proteins, as is the case in mechanically sensitive
systems. As a rule, such energy transduction operates only
in one direction.
Hence it came as a great surprise, when in 1978 David
Kemp recorded sound events in the human (outer) ear canal
that were apparently produced by the inner ear (Kemp
M. Kössl  D. Möckel  M. Weber  E.-A. Seyfarth (&)
Institut für Zellbiologie und Neurowissenschaft,
J.W. Goethe-Universität, Siesmayerstrasse 70,
60323 Frankfurt am Main, Germany
e-mail: seyfarth@bio.uni-frankfurt.de
598 J Comp Physiol A (2008) 194:597­609
123
1978). He stimulated the ear with brief acoustic clicks and
demonstrated that sound waves of a particular frequency
appeared in the external ear meatus after a distinct delay of
several milliseconds. Initially, these events were called
"Kemp-echos". However, they do not originate from echolike
reXections at the ear-drum or the middle-ear but
depend on an anatomically and physiologically intact
cochlea. Subsequent studies in numerous mammalian species
have shown that the outer hair cells (OHCs) in the
cochlea are the elements responsible for the retrograde generation
of sound (Kemp 2002). While the inner hair cells
(IHCs) are extensively innervated by the dendrites of aVerent
auditory neurons, the OHCs possess only a few auditory
nerve synapses but are directly contacted by eVerent axons
that provide centrifugal information from the brain stem to
the inner ear. Zheng et al. (2000) discovered a protein,
called prestin, in the cell membrane of OHCs which undergoes
voltage-dependent changes in conformation, resulting
in shortening or elongation of the hair cell soma upon depolarization
or hyperpolarization of its membrane (Liberman
et al. 2002; Dallos and Fakler 2002). Following an acoustic
stimulus, OHCs are thereby capable of producing forces
which can signiWcantly augment the displacement amplitude
of the basilar membrane in the inner ear. This voltagedependent
motility of the OHCs has been considered the
basis of the so-called "cochlear ampliWer", which provides
an improvement of sensitivity by 40­60 dB in the mammalian
ear (Geleoc and Holt 2003). It is up for debate, however,
whether other ampliWcation mechanisms might
contribute to the cochlear ampliWer in OHCs that are based
on force production by the stereocilia. At present, the issue
remains unsolved (see the discussion below).
Nonlinear mechanical ampliWcation of sound energy is
not restricted to the ears of mammals or of vertebrates as
such, but seems to be a general principle in sensitive hearing
organs. In the ears of insects, both nonlinear ampliWcation
and otoacoustic emissions have recently been
demonstrated. The potential underlying mechanisms are
currently being investigated; they are the subject of our
present review.
Two types of otoacoustic emissions
Otoacoustic emissions (OAEs) are a by-product of the
cochlear ampliWer and have been used to study its function in
detail. OAEs have also been recorded from the ears of nonmammalian
vertebrates whose hair cells do not express electromotile
prestin. Hence, other ampliWcation processes must
be responsible in these cases (Manley 2001, 2006; Manley
et al. 2001). The most likely candidates are interactions of
transduction proteins with actin and myosin in the stereociliary
bundle of the hair cells (see our discussion below).
We distinguish two types of otoacoustic emissions:
place-dependent OAEs and wave-dependent OAEs (see
also Shera and Guinan 1999).
Place-dependent OAEs are comprised of distinct sound
frequencies and hence can be attributed to a particular frequency
position in the inner ear. Place-dependent OAEs
include emissions induced by external sound stimulation of
the inner ear (such as the click-evoked OAEs mentioned
above) as well as spontaneous OAEs (SOAE) that are
recorded in the absence of any external sound. SOAEs provide
the most compelling indication of active sound generation
in the inner ear (see Hudspeth 2000). In many subjects,
SOAEs are measurable in the form of a narrow maximum
of sound pressure at a distinct frequency (Fig. 1, left). Often
several SOAEs appear in a given ear. In humans, they occur
at frequencies that lie in the range of best hearing, i.e.,
between 0.5 and 11 kHz. Spontaneous motion of the OHCs
has been considered as the cause of SOAEs (review Kemp
2002), but to explain the frequency-speciWcity of these
emissions, additional mechanisms are required such as the
emergence of standing waves at cochlear sites that are discontinuous
(Russell and Kössl 1999).
Amphibians, reptiles, and birds display distinct SOAEs
as well (Fig. 1, left). The main diVerence to mammals cases
is their limitation to frequencies below ca. 10 kHz. SOAEs
at this frequency have been measured in the barn owl (Taschenberger
and Manley 1997). In the case of mammals,
SOAE-frequencies up to 62 kHz have been recorded in bats
(Kössl 1994; Fig. 1, left). This Wnding demonstrates that the
mammalian ear can actively generate (i.e., without external
sound stimulation) extremely rapid movements that could
be employed for acoustic ampliWcation, corresponding to
the hearing range of many mammalian species that extends
into extreme ultrasound. Electromotility of OHCs has in
fact been found up to at least 79 kHz (Frank et al. 1999).
Possibly, cochlear ampliWers that are based on prestin even
operate up to the auditory frequency limits of mammals,
which are beyond 200 kHz in certain species relying on
echolocation.
Wave-dependent OAEs are evoked emissions that arise
following auditory stimulation by several pure tones of
diVerent frequencies. Due to nonlinear sound processing in
the inner ear, the amplitude of faint acoustic signals
increases in comparison to louder signals and the overall
signal wave form becomes distorted. Such distortions are
most pronounced when several pure tone signals are
applied simultaneously. The standard procedure to evoke
such distortions is stimulation with two pure tone stimuli of
diVerent frequency (f1, f2). The wave form distortions are
then measured in the frequency spectrum as additional frequency
components, so-called distortion-products, at frequencies
of nf1  (n  1)f2 and nf2  (n  1)f1. These
components propagate back across the middle ear to the
J Comp Physiol A (2008) 194:597­609 599
123
tympanic membrane and are recorded as distortion-product
OAE (DPOAE). In all hearing organs examined so far, the
DPOAE at 2f1  f2 is the loudest one (Fig. 1, right).
Distortion is the cost that has to be paid for highly sensitive
sound ampliWcation in biological systems. While many
technical systems, such as hiW-setups, produce distortions
when they become overloaded, auditory systems primarily
generate distortions at low sound pressure when the
cochlear ampliWer is operating at maximal gain. Unlike
place-dependent OAEs that are generally limited to one
particular frequency, DPOAEs can be elicited across the
entire hearing range of the subject and hence provide more
detailed information on cochlear ampliWcation. Increasingly,
DPOAEs are used as diagnostic tools, for instance to
examine newborn children for auditory impairment (see
e.g., the review by Kemp 2002). It should be emphasized
here that any nonlinear system can produce distortions, no
matter whether the nonlinearity is due to active cellular processes
or to passive properties of the system. Moreover, if a
physical or biological system would produce linear ampliWcation
that does not change its gain with sound level, no
two-tone distortions would be generated. In the case of the
mammalian cochlea, however, numerous studies have
shown that sensitive DPOAEs are due to active and nonlinear
ampliWcation that is dependent on metabolic processes
in the OHCs (see e.g., Johnson and Canlon 1994; Frolenkov
et al. 1998; Kössl and Vater 2000).
The auditory organs of insects and measuring OAEs
Many insects have excellent hearing and use it to detect
food or predators and to communicate with conspeciWcs
(see e.g., the reviews by Stumpner and von Helversen
2001; Gerhardt and Huber 2002). Unlike non-mammalian
vertebrates, their hearing range is not limited to relatively
low frequencies below 10 kHz but can extend far into the
ultrasound range. This applies to many moth species and to
crickets and katydids. For instance, certain oak bushcrickets
(Meconematinae) produce sound at 129 kHz
(Montealegre-Z et al. 2006); so far nothing is known of
their auditory capabilities, but it seems likely that they can
hear their own songs.
The anatomy of hearing organs in insects greatly diVers
from that of vertebrate ears. Tympanal organs, in which the
primary auditory neurons lie next to a tympanic membrane,
Fig. 1 Origin and characteristics of otoacoustic emissions (OAE) in
various mammalian and non-mammalian vertebrates. Lower left: In the
absence of acoustic stimuli, spontaneous OAEs are recordable at the
tympanic membrane via a microphone; they appear as amplitude peaks
at speciWc sound frequencies. Lower right: Distortion-product OAEs
are evoked by simultaneous stimulation with two pure tones (f1, f2)
and appear as additional peaks (red) at deWned frequencies in the
amplitude spectrum. They emerge at the basilar membrane in the region
of overlap of the two travelling waves f1, f2 (top: area in red), due
to nonlinear ampliWcation by outer hair cells (OHC). Data sources:
Long et al. 1996 (frog); Kössl, unpubl. (human); Manley 2006 (iguana
and Anolis); Taschenberger and Manley 1997 (barn owl); Kössl 1994
(moustache bat, DPOAE on lower right)
humanhuman
reptiles
frog
barn owl
moustached
bat
2 4 6 8 10 12 62 63
soundlevel(dBSPL)
Iguana
Anolis
-20
-10
0
10
20
30
travelling waves
f2f1
spontaneous OAE
frequency (kHz)
microphone
soundlevel(dBSPL)
loudspeaker
distortion
energy
f1
f2
frequency
stimulus
f1 f2
soundlevel
distorsion product OAE
57 59 61 63
0
20
40
60
f1 f2
2f1-f2
3f1-2f2
2f2-f1
3f2-2f1
inner ear
frequency (kHz)
OHC
600 J Comp Physiol A (2008) 194:597­609
123
are distinguished from antennal hearing organs, in which
the sensory neurons are arranged ringlike at the base of
each Xagellar antenna (called "Johnston's organ"). Tympanal
organs react to frequencies ranging far into ultrasound.
Possessing a relatively stiV tympanic membrane, they can,
in principle, generate sound in the form of otoacoustic
emissions. Characteristically, tympanal organs are situated
in the thorax, abdomen, or the legs of grasshoppers and
moths (Hoy and Robert 1996). Antennal hearing organs, on
the other hand, perceive air-particle movements in the
acoustic near-Weld and are situated on the heads of insects
such as Xies and mosquitoes (Fig. 2c).
In the case of locusts, ca. 80 sensory neurons are gathered
into a peripheral sensory ganglion called "Müller's
organ" that lies beneath a relatively large tympanic membrane
in the Wrst abdominal segment (Fig. 2a). As in all
insect auditory organs, both tympanal and antennal ones,
the sensory neurons are scolopidial mechanoreceptors, each
consisting of a bipolar sensory neuron with a ciliated dendrite
and two types of accessory cells, the scolopale cell
and the attachment cell (see Yack 2004 for a recent detailed
review). The sensory cilium is anchored to the root-apparatus
of the dendric process (Fig. 2d). Responsible for sound
transduction is the distal region at the ciliary tip that is
enclosed by an extracellular cap. The sensory cell somata
are gathered in the ganglion of Müller's organ, and their
dendrites reach out to various sclerotized attachment points
on the tympanic membrane (Gray 1960). The tympanum is
divided into relatively thin and thick regions. Upon acoustic
stimulation, these respective regions are maximally deXected
at high versus low frequencies. Thus, the sound
frequency to which the sensory cells react best depends on
the exact site of attachment on the tympanum. For instance,
the sensory neurons whose dendrites attach via the "pyriform
vesicle" in a region of thin membrane (Fig. 2a), react
best to relatively high sound frequencies above 10 kHz.
Fig. 2 Important anatomical features of auditory organs in insects. a
Tympanal organ in the Wrst abdominal segment of the locust (Locusta
migratoria), inside view. The relatively large tympanum (tym) spans
across a cuticular ring. The peripheral sensory ganglion (Müller's
organ, MO) lies near the anterior edge, from where sensory dendrites
reach out to sclerotized attachment points on the tympanum. The dendrites
in the pyriform vesicle (pv) insert in a relatively thin membrane
region, those of the elevated process (ep) in a thicker region; tyn tympanal
nerve. b Tympanal organ of tropical bushcricket (Mecopoda
elongata) in the opened foreleg tibia and after impregnation with methylene
blue. The characteristic crista acustica (ca) extends along a
branch of the acoustic trachea (tr). It comprises ca. 40 scolopidial
sensory neurons and is part of a larger mechanosensory organ complex
in the insect leg; sn sensory neurons; ac attachment cells; post tym posterior
tympanum. c Scanning electronmicrograph of Drosophila head
showing the two antennal hearing organs next to the compound eyes
(ce). The branched arista (ar) is Wrmly attached to the third antennal
segment (3); both oscillate in the acoustic near-Weld (see also Fig. 7);
1, 2, 3 segments of left antenna. d Schematic representation of a single
scolopidium with bipolar sensory neuron and accessory cells. The dendrite
(dt) of the sensory neuron (sn) runs out into a ciliated terminal
(ci); cd ciliary dilatation; ac attachment cell; sc scolopale cell; cr ciliary
root; bb basal body; ro rod; cp scolopidial cap
J Comp Physiol A (2008) 194:597­609 601
123
Other dendrites terminate via sclerites in thicker membrane
regions (such as the "folded body") and react to lower frequencies.
Hence the tympanal organ of locusts is characterized
by mechanical frequency Wltering and reaction to
distinct sound frequencies just like the vertebrate ear
(Michelsen 1971; Römer 1976; Breckow and Sippel 1985;
Jacobs et al. 1999; Windmill et al. 2005).
Distinct tonotopic representation of sound frequencies is
found in the tympanal organ of bushcrickets (katydids).
Here the auditory receptors form a linear array of cells
inside the tibia of each foreleg called "crista acustica"
(Fig. 2b). The anatomical arrangement resembles the situation
in the cochlea of the vertebrate ear. The ears of bushcrickets
are primarily stimulated by sound entering the
foreleg via a spiracle in the prothorax, from where it is
transmitted to the crista acustica by an acoustic trachea
(Lakes and Schikorski 1990). The spiracular tracheal system
acts primarily as a high-frequency gain Wlter, amplifying
sound in the high audio-frequency range or ultrasonics.
Tympanic membranes exist near the sense organ in the
tibia, but these are not directly contacted by the sensory
dendrites; rather they seem to aid in impedance matching
(Bangert et al. 1998). The scolopidial sensory neurons
(with their dendritic caps) are oriented perpendicular to the
long axis of the crista acustica. The linear anatomical organization
is the basis for a linear tonotopic coding of the
receptor neurons, that is, sensory cells in the proximal
region of the crista code for low sound frequencies and
those in the distal part for high frequencies, with a continuous
gradient of frequency coding along the sense organ
(OldWeld 1982, 1988; Stumpner 1996). Microscopic
inspection of the organ shows that the size of the dendritic
caps gradually decreases from proximal to distal (Fig. 2b),
suggesting that the mechanical features of these structures
might contribute to the tonotopic coding characteristics. Of
course this does not rule out that other, tonotopically
changing properties of the sensory cells might exist as well
(see also the discussion by Pollack and Imaizumi 1999).
Acoustic distortions in tympanal organs
Upon sound stimulation with pure tones, tympanal organs
generate distinct DPOAEs (Fig. 4, 5) that resemble those
measured in vertebrate ears in many respects (Kössl and
Boyan 1998a, b; Coro and Kössl 1998, 2001; Kössl and
Coro 2006; Möckel et al. 2007; Kössl et al. 2007). The
emissions are recordable via a probe that is adapted to the
diameter of the tympanum (in the case of locusts and
moths) or the spiracle (of bushcrickets) and that comprises
a stimulation channel with two loudspeakers and a recording
channel with a microphone (for technical details see
Fig. 3). In locusts and moths, DPOAEs can be induced even
with sound stimuli that lie near the auditory threshold; the
2f1  f2 emission is always the loudest (see the example in
the power spectrum of Fig. 4a). This indicates that the
motion of the tympanum/sense organ is nonlinear close to
the auditory threshold. By determining the stimulus level
(at various sound frequencies) that suYces to evoke a
DPOAE of a certain amplitude (e.g., 10 dB SPL), it is
possible to calculate DPOAE-threshold curves. Such calculations
predict the frequency-speciWc sensitivity of the nonlinear
mechanics of the tympanal organ. In fact, they
resemble the neuronal threshold of the locust ear, exempliWed
by two neuronal tuning curves (Fig. 4b). Hence, noninvasive
DPOAE-measurements can be used to determine
objective auditory thresholds in insects, comparable to the
procedure applied in mammals, particularly in humans.
Insect DPOAEs are vulnerable to manipulations that aVect
the physiological state of the animal, such as the application of
certain anesthetics or hypoxic substances (Fig. 4c). Similar
treatment clearly reduces DPOAE-levels also in mammals,
due to apparent blockade of the cochlear ampliWer. In the case
of insects, it is still unclear if it is actually the sensory cells
and potential ampliWcation mechanisms that are aVected by
Fig. 3 Recording DPOAEs in the bushcricket Mecopoda elongata,
experimental setup. The insect is secured to a metal holder and the
forelegs Wxed to two supports with beeswax. For acoustic stimulation,
two pure-tone signals are generated via D/A-conversion and fed into
two loudspeakers (ls1, ls2) following calibration and power ampliWcation.
Separation of the two stimulus channels is essential to preclude
distortions in the technical sound-producing system. The resulting
sound signal, which consists of the two stimuli as well as the DPOAE,
is recorded via a microphone, then ampliWed and fed into a A/D converter.
Sound-producing and sound-recording channels are gathered
into a coupling device, whose tip is adapted to the diameter of the
opening of the acoustic trachea (spiracle) in the prothorax of Mecopoda.
One advantage of this arrangement is that the stimulating/recording
device is applied to the prothoracic spiracle while the tympanal
organ with its crista acustica remains freely accessible for experimental
manipulations
tympanal
organspiracle
microphone
ls1
ls2
amplifier
sound
control
computer
D/A A/D
converter
602 J Comp Physiol A (2008) 194:597­609
123
such manipulations. Scolopidial sensory neurons have (aVerent)
axons that connect directly to the central nervous system
via the auditory nerve. This fortuitous anatomical situation
oVers an opportunity to stimulate auditory aVerents electrically
at some distance away from the sense organ so that its
mechanical structures remain unharmed by surgical procedures.
This way we were able to test if induced changes of the
membrane potential aVect the generation of otoacoustic emissions.
In fact, electrical stimulation causes reversible changes
in DPOAE-amplitude (Fig. 4d; Möckel et al. 2007). The
result suggests that events intrinsic to the sensory neurons
themselves (and possibly the cap cells) contribute to the production
of emissions.
The number of sensory cells in tympanal organs varies
from species to species. Notodontid moths like the mapleprominent,
Ptilodon cucullina, or the buV-tip, Phalera sp.,
seem to have the most simple organs. Here the tympanum
is stretched tightly between two elastic exoskeletal elements,
and only a single scolopidium is positioned between
a cuticular anchor and the tympanum (Fig. 5a,b; Eggers
1919; Surlykke 1984). In this strategic position, the auditory
neuron will react very sensitively to deXections of the
tympanum. Should the scolopidial sensory cell possess
active components of ampliWcation, it should also be capable
to inXuence the mechanics of the tympanic membrane
directly. It turns out that even the simple metathoracic ear
of notodontids emits prominent DPOAEs, whose growth
function is comparable to that of more complex tympanal
organs and that of vertebrate ears (Fig. 5c). Following treatment
with ether (as an anesthetic), the DPOAE-amplitudes
decrease considerably, which suggests that metabolic processes
are actively involved in the generation of emissions
(Kössl et al. 2007). In notodontids, DPOAEs can be elicited
at frequencies near 100 kHz which is in accordance with a
sensitive high-frequency hearing range in these moths (Surlykke
1984).
More detailed information on the location of DPOAE-generation
can be obtained from the ear of locusts after setting
Wne surgical lesions at the attachment points of sensory dendrites
at the tympanum (Fig. 6c). Here the vibrations of the
tympanic membrane are frequency-speciWc in the form of
travelling waves (Fig. 6a,b; Windmill et al. 2005). At the
location of the "pyriform vesicle" (PV in Fig. 6c), maximal
deXections are achieved at frequencies > ca. 12 kHz (26 kHz
in the example of Fig. 6b). At the "folded body" (FB), that is,
another site of dendritic attachment, maximal deXection can
only be induced at much lower frequencies (3.3 kHz in the
example of Fig. 6b). After the connection between pyriform
vesicle and the peripheral sensory ganglion has been severed,
high-frequency-DPOAEs are drastically reduced (Fig. 6d,
top). Additional ablation of the folded body together with the
whole sensory ganglion results in a reduction of DPOAE-levels
across the entire frequency range (5­30 kHz; Fig. 6d, bottom).
This Wnding underpins the notion that the
mechanosensory cells themselves are contributing to
DPOAE-generation in a frequency-speciWc way.
Vibrating antennal hearing organs
Although the results discussed in the previous chapter provide
circumstantial evidence, Wnal proof is still lacking in
the case of tympanal organs demonstrating that an active
ampliWcation process does in fact exist within the sensory
cells and how such a putative mechanism is constituted. In
the case of the antennal ears of Drosophila and of male
mosquitoes, however, Göpfert and Robert (2001, 2003) and
Jackson and Robert (2006) have demonstrated with laserFig.
4 DPOAE-measurements in the tympanal organ of the locust. a
DPOAE-spectrum following two-tone stimulation with 10 and 12 kHz;
distortion products (peaks in red) reach sound intensities that are clearly
distinguishable from background noise. Asterisk denotes 2f1  f2
distortion product. b Auditory threshold curves determined electrophysiologically
for two groups of neurons (in red; adapted from Römer
1976) and expressed as sound intensity suYcient to elicit DPOAEs (in
black; adapted from Kössl and Boyan 1998b) within the normal hearing
range of locusts. c Reduction of DPOAE-levels following CO2treatment
demonstrates the importance of active metabolic processes
for generation of otoacoustic emissions (adapted from Kössl and Boyan
1998a, b). d Brief electrical stimulation of the auditory nerve (horizontal
bars) causes transient reduction of emission levels (acoustic
stimuli: 10.5/12 kHz with 60/50 dB SPL; electrical nerve stimulation:
bipolar pulses of 10 A, pulse frequency 200 Hz, pulse duration 2 ms;
modiWed from Möckel et al. 2007)
20 240 660 10
-15
-10
0
0
15
10
post-treatment
pre-treatment
CO2
time (min)
hypoxia electrical stimulation
of auditory nerve
stimulus level (dB SPL)
DPOAE-level(dBSPL)
1
DPOAE-
thresholds
neuronal
thresholds
60
80
20
40
1 10 100
60
40
20
0
-20
frequency (kHz)
soundlevel(dBSPL)
5 10 15
a
c
b
d
*
J Comp Physiol A (2008) 194:597­609 603
123
vibrometric measurements that mechanical energy is in fact
generated in these organs (see the examples in Fig. 7).
Compelling evidence of such active force generation are
the spontaneous vibrations of the antennae at frequencies
below ca. 1 kHz. They can be induced by pharmacological
and genetic manipulations (Fig. 7b,c). Such vibrations do
not generate sound pressures that are loud enough to be
recordable as OAEs (because the organs lack a tympanic
membrane), but the underlying cellular mechanisms may
resemble those in tympanal organs. The quest for the
molecular basis of force generation focuses on proteins
associated with the transduction apparatus at the cilium.
Although it is still inconclusive in the case of vertebrate
hair cells which protein constitutes the transduction channel
(see the review by Corey 2006), with some circumstantial
evidence that a member of the TRP-group (TRP Transient
Receptor Potential) may be the most promising candidate
(Cuajungco et al. 2007), in Drosophila at least, there is
good evidence that a TRP-protein is indeed responsible for
sensory transduction (Kernan 2007). Antennal vibrations
are signiWcantly altered after deleting the genes that either
code for a putative TRP transduction channel or for associated
proteins (Göpfert and Robert 2003; Göpfert et al.
2005, 2006). These results indicate that the transduction
apparatus with its associated proteins is critically involved
in force production, even though the detailed molecular
interactions are still unclear (Göpfert et al. 2006). Interestingly,
the scolopidial sensory neurons of Drosophila also
possess a prestin-homologous protein (Weber et al. 2003),
but so far it appears unlikely that it is relevant here for the
production of mechanical forces.
Conclusion and a comparison of active ear mechanics
in vertebrates and insects
In reptiles and birds (sauropsids) active and nonlinear
mechanical ampliWcation in the inner ear is not based on
Fig. 5 The metathoracic tympanal
organ of notodontid moths
has only one auditory neuron;
nonetheless, the ear produces
distinct DPOAEs in the high frequency
range. a Detailed drawings
of Eggers (1919;
plate 23)demonstrate that the
tympanal scolopidium of the
buV-tip (Phalera bucephala)
comprises only one sensory neuron
(T tympanum; N tympanal
nerve; L ligament; remaining labels
denote putative cell nuclei).
b Schematized representation of
the tympanal organ of notodontids;
the scolopidial dendrite is attached
to the tympanum. c
DPOAE growth-functions in
tympanal organ of another notodontid,
the maple-prominent
(Ptilodon cucculina), before and
after treatment with ether acting
as an anesthetic; the two sound
stimuli were at 69.5 kHz (f1) and
75 kHz (f2) (modiWed from
Kössl et al. 2007)
tympanum
axon
scolop-
idium
Ptilodon
cucculina
b
DPOAE-level(dBSPL)
stimulus level (dB SPL)
35 55 75
-20
0
20
40
60
pre-treatment
ether
69.5/75 kHz
c
a
604 J Comp Physiol A (2008) 194:597­609
123
motility of the sensory cell soma, but on motility of the stereovilli
of the hair cells. Correspondingly, the stereovilli are
responsible for the generation of spontaneous OAEs (Manley
et al. 2001; Manley 2001). In mammals, there is ample
evidence that prestin-based somatic motility of outer hair
cells is an important component of the cochlear ampliWer
(reviewed by Ashmore 2008). Recent evidence, however,
has shown that there is also active force production in the
stereocilia of inner and outer hair cells. It is employed for
adaptation during auditory transduction and could amplify
sound waves on a cycle-by-cycle basis (reviewed by Fettiplace
and Hackney 2006). Hence at present, the question
remains unresolved which amplifying mechanism is the
relevant one in mammals. Techniques are required here that
can dissect the somatic versus the stereocilia mechanisms
in vivo. As to somatic motility of outer hair cells, it is
known that changes in the membrane potential cause prestin
to change its conformation which instantly shortens or
Fig. 6 Müller's organ of Locusta migratoria; travelling waves and
DPOAEs before and after lesions. a External view of tympanum; diagonal
line indicates measuring points (in mm) for laser-vibrometric
recordings of tympanic vibrations. Arrows mark position of pyriform
vesicle (PV) and of folded body (FB). b Stimulation with pure tones
generates vibration patterns that resemble travelling waves across measuring
points. Low frequency stimulation (at 3.3 kHz) causes displacements
of PV and FB, while high-frequency stimulation (at 26 kHz)
generates amplitude maxima at the PV without displacing FB. c Internal
view of locust tympanum showing Müller's organ and PV (scanning
electronmicrograph). The PV was severed from Müller's organ by
setting a lesion through the dendritic attachments (yellow spot).
d DPOAE-levels at sound stimuli of constant intensity and varying frequency
(f2), i.e., so-called DP-gram (f2/f1 ratio maintained at 1.08; L1/
L2 of 60/50 dB SPL). Upper: following a lesion at PV, DPOAE-levels
above ca. 15 kHz (red curve) drop nearly to noise level (dashed black
lines noise level expressed as the average (§1SD). Lower: subsequent
lesion of FB and of Müller's organ itself (in the same preparation)
cause a marked reduction of DPOAE-levels across the entire frequency
range. The results indicate that high-frequency DPOAEs are only
generated as long as the connection between sensory neurons and PV
remains intact (a,b modiWed from Windmill et al. 2005; c,d modiWed
from Möckel et al. 2007)
0.2
-5
0
0.6 1.0 1.4
5
PV FB
0.2
-30
0
0.6 1.0 1.4
distance across tympanum (mm)
-1gain(mmsPa)
DPOAE-level(dBSPL)
30
PV FB
3.3 kHz
26 kHz
lesion at PV
-40
-20
0
20
lesion at FB/ganglion
frequency f2 [kHz]
5 6 7 8 9 15 20 25 3010
-40
-20
0
20
DPOAE audiograms
tympanal
travelling wave
DPOAE
audiograms
PV
PV
lesion
intact animal
a
b
c
d
J Comp Physiol A (2008) 194:597­609 605
123
elongates the cell soma. This process is rapid and can follow
very high sound frequencies as indicated by fast motile
responses induced electrically in isolated outer hair cells up
to at least 79 kHz (Frank et al. 1999). However, since the
receptor potential of these cells in vivo decreases with frequency
due to membrane capacitance, it is not clear if it is
suYcient to drive somatic motility at very high frequencies.
Several possibilities have been discussed as to how the
required driving voltage could be provided at very high frequencies,
including extracellular potentials (Liao et al.
2007; Ashmore 2008).
AmpliWcation in stereovilli, which relies on interactions
between the transduction molecules and actin/myosin-like
proteins, seems to operate at a much slower rate than
somatic motility. So far, all available evidence gathered in
non-mammalian vertebrates demonstrates that active stereovilli
movements, adaptation mechanisms, and even electrically
induced OAEs are restricted to frequencies below
ca. 3 kHz (Manley et al. 2001; Fettiplace et al. 2001). Incidentally
in mammals, potential active stereovilli processes
are limited to a similar, low frequency range (Ricci et al.
2002; LeMasurier and Gillespie 2005). Correspondingly,
also the hearing range of non-mammalian vertebrates is
limited to relatively low frequencies (Fig. 8). Barn-owls
appear to be the front-runners among non-mammalian vertebrates
as far as high-frequency auditory processing is
concerned; their hearing range extends to ca. 11 kHz.
Accordingly in these owls, spontaneous OAEs are recordable
at ca. 9­11 kHz (Fig. 1). If an adaptation time constant
of 120 s in mammalian outer hair cells can serve as an
indicator of the frequency limits of hair bundle mechanisms
(Kennedy et al. 2003), then the calculated frequency limit
would be close to 8 kHz. To summarize, two potential
modes of cochlear ampliWcation exist in the case of vertebrates:
(1) a stereovilli-based process that covers low frequencies,
and (2) a somatic mechanism that ampliWes both
low and very high frequency signals.
By comparison, insects possess an auditory ampliWer in
the lower frequency range (up to ca. 1 kHz) that seems to
be based on a ciliary mechanism. This has clearly been
demonstrated in the antennal organs of Drosophila (Göpfert
and Robert 2001, 2003) and of male mosquitoes (Jackson
and Robert 2006); it is analogous to the situation in
non-mammalian vertebrates (see Albert et al. 2007). So far
in the case of insect tympanal organs, which can process
very high frequencies above 100 kHz like the ear of mammals,
a mechanical ampliWer based on transduction proteins
of the cilium has not been demonstrated (Fig. 8). If force is
generated here similar to the mechanism in antennal organs,
the putative ampliWcation process would be much more
Fig. 7 Vibrating antennal ear of
Drosophila. a Schematic representation
of the two distal antennal
segments with branched
arista and arrangement of laservibrometric
measurements
(adapted from Göpfert et al.
2005). Near-Weld sound vibrations
of the arista rotate the third
antennal segment relative to the
second. b Time course of spontaneous
("self-sustained") antennal
oscillations in nan-mutant
(having deWcient TRPV channels);
by comparison, in animals
with deWcient NompC TRPchannels
spontaneous activity is
reduced below that found in wild
type Xies. c Power spectra of
such oscillations in the same
Drosophila strains. Black traces,
spectrum obtained for one arista;
gray, spectra from additional
specimens of the same strain
(b, c adapted from Göpfert et al.
2006)
arista
sensory
neuron
rotational
vibrations
laser spot
antennal
socket
wild type
NompC-mutant
20 ms
800nm
spontaneous antennal
oscillations
nan-mutant
frequency (kHz)
2spectraldensity(nm/Hz)
wild type
(NompC control) NompC-mutant
nan-mutant
0.1
-4
10
-2
10
4
10
2
10
0
10
0.1 0.10.3 0.3 0.31 1 1
branched
a b
c
606 J Comp Physiol A (2008) 194:597­609
123
rapid than is currently known to exist in cilia or stereovilli.
Unlike the Wnding in antennal organs, so far no spontaneous
vibrations or SOAEs have been observed in tympanal
organs. Nonetheless, DPOAE-recordings from tympanal
organs show that the sensory cells are capable of processing
faint sound in a sensitive, nonlinear fashion and that this
process retroacts on the vibration properties of the tympanic
membrane.
Even if a potential ampliWer does not directly act on the
transduction apparatus but in the cell soma (like prestin in
mammals), it will still crucially depend on the transduction
apparatus and the resulting receptor (generator) potential.
Voltage-dependent mechanical deformation of the cell that
contributes to raising the receptor potential (which drives
the process), is a typical positive feedback system. If the
ampliWcation factor is close to 1, the system is prone to
spontaneous oscillations. The generation of DPOAEs in
mammals can easily be simulated, using the characteristic
behavior of the transduction currents in hair cells as the
nonlinear element. Such S-shaped transduction characteristics
are generally approximated by Boltzmann-functions;
they reXect the level-dependent properties of DPOAEs
quite well. A single Boltzmann-function suYces to simulate
the DPOAEs in mammals realistically (see e.g., Frank
and Kössl 1996; Lukashkin and Russell 1999; Lukashkin
et al. 2002). Using such simulations, and comparing how
the DPOAE-levels depend on the respective stimulus levels
in tympanal organs and in the mammalian cochlea, reveals
clear diVerences between the two types of auditory organ.
In the case of the tympanal organs of moths (for example in
Empyreuma aYnis, Kössl and Coro 2006), varying the two
stimulus levels results in a triangular DPOAE-activation
range (Fig. 9a), which can be simulated by a Boltzmanfunction
(Fig. 9b), but only as long as the zero-point or
operating point of the function lies asymmetrically. Using
symmetrical Boltzmann-functions, the DPOAE-activation
range has a diVerent shape (Fig. 9c) and resembles the one
found in the ear of mammals (Kummer et al. 2000) or frogs
(Meenderink and Van Dijk 2005). Interestingly, receptor
potentials measured directly in tympanal cells of locusts are
restricted to depolarizations and do not show hyperpolarizations
upon acoustical stimulation (Hill 1983a, b). This
Fig. 8 Comparing the ears of mammals, sauropsids, and insects: fundamental
anatomical and bio-acoustical parameters. Hearing ranges
are indicated without considering infra-sound perception. DPOAEsensitivities
are listed for minimal stimulus-levels that are just suYcient
to evoke DPOAEs of 10 dB SPL. IHC inner hair cell; OHC outer
hair cell; TM tectorial membrane; BM basilar membrane. Data
sources: Frank et al. 1999 (motility of OHC-somata); Ricci et al. 2002
(motile stereovilli in mammalian cochlea); Manley et al. 2001 (electrically
evoked OAEs in reptiles); Manley et al. 1993 (DPOAEs in lizards).
See text for detailed discussion
TTM
BM
OHCIHC
TM
outer hair cell
tympanal organ
tympanum
short hair cell
kinocilium
ciliary root
cilium
primary
mechano-
receptor
mammals sauropsids insects
BM
short
HC
TM
long HC
a b
c d
stereovilli
stereovilli
cuticular plate
hearing range 10 Hz - >200 kHz 80 Hz - 11 kHz 1 kHz - 100 kHz 100 - 1000 Hz
spontaneous some species some species spontaneous vibrations
OAE 0.5 - 62 kHz 0.8 - 11 kHz ? 100 - 1000 Hz
up to 40 dB SPL up to 25 dB SPL
DPOAEsensitivity
ca. -3 dB SPL ca. 10 dB SPL ca. 12 dB SPL ?
cell soma 70 kHz negative ? ?
motility
motile cilia/ < 3 kHz bis ~3 kHz ? ?
stereovilli (force generation) (electr. evoked OAE)
antennal
organ
antenna
J Comp Physiol A (2008) 194:597­609 607
123
Wnding can be explained by strongly asymmetrical transfer
functions. Hence, the single cilium of scolopidial sensory
neurons seems to create somewhat diVerent constraints than
the hair cells in the mammalian cochlea. Nevertheless, similar
proteins appear to be involved in both types of cells
(scolopidial and hair cells), and mechanical feedback
appears to be an integral part of sensitive hearing.
Acknowledgments We thank Frank Coro and Manuela Nowotny for
stimulating discussions and Kerstin Röhr for providing the photograph
of Fig. 2a. Support was provided by grants from the Deutsche Forschungsgemeinschaft
(DFG) and by stipends from the Evangelisches
Studienwerk (to D. M.) and the Jürgen Manchot Stiftung (to M. W.).
All our experiments reported here comply with the "Principles of Animal
Care", Publication No. 86-23, revised 1985, of the National Institute
of Health (USA), and also with current legislation in Germany.
References
Albert JT, Nadrowski B, Göpfert MC (2007) Mechanical signatures of
transducer gating in the Drosophila ear. Curr Biol 17:1000­1006
Ashmore J (2008) Cochlear outer hair cell motility. Physiol Rev
88:173­210
Bangert M, Kalmring K, Sickmann T, Stephen R, Jatho M, Lakes-Harlan
R (1998) Stimulus transmission in the auditory receptor organs
of the foreleg of bushcrickets (Tettigoniidae). I. The role of
the tympana. Hearing Res 115:27­38
Breckow J, Sippel M (1985) Mechanics of the transduction of sound in
the tympanal organ of adults and larvae of locusts. J Comp Physiol
A 157:619­629
Corey DP (2006) What is the hair cell transduction channel? J Physiol
576:23­28
Coro F, Kössl M (1998) Distortion-product otoacoustic emissions
from the tympanic organ in two noctuoid moths. J Comp Physiol
A 183:525­531
Coro F, Kössl M (2001) Components of the 2f1­f2 distortion-product
otoacoustic emission in a moth. Hearing Res 162:126­133
Cuajungco MP, Grimm C, Heller S (2007) TRP channels as candidates
for hearing and balance abnormalities in vertebrates. Biochim
Biophys Acta 1772:1022­1027
Dallos P, Fakler B (2002) Prestin, a new type of motor protein. Nat Rev
Mol Cellbiol 3:104­111
Eggers F (1919) Das thoracale bitympanale Organ einer Gruppe der
Lepidoptera Heterocera. Zool Jb (Anat) 41:273­376
Fettiplace R, Hackney CM (2006) The sensory and motor roles of auditory
hair cells. Nat Rev Neurosci 7:19­29
Fettiplace R, Ricci AJ, Hackney CM (2001) Clues to the cochlear
ampliWer from the turtle ear. Trends Neurosci 24:169­175
25 35 45 55 65 75
25
35
45
55
65
75stimulus2(dBSPL)
E.affinis
shift 0.25
30 40 50 60 70 80
30
40
50
60
70
80
DPOAE level (dB SPL)
-10 10 30 50
model
asymmetric
input
output
30 40 50 60 70 80
30
40
50
60
70
80
model
symmetric
input
output
stimulus 1 (dB SPL)
a
b
c
Fig. 9 DPOAE-generation in the (metathoracic) tympanal organ of
the moth Empyreuma aYnis (Arctiidae) (a), and in an approximation
model that is based on a nonlinear Boltzmann-function (b, c). Such
functions are used to describe the transduction behavior of hair cells in
the vertebrate cochlea. Here, the DPOAE-amplitude (color-coded) is
represented as a function of the amplitude of the two pure-tone stimuli.
The insect data are well represented by the approximation in (b) as
long as the operating point (or zero-point) of the function lies asymmetrically.
Using a symmetrical model (c), the DPOAE-activation
range resembles the one characteristically found in vertebrate ears (see
text for a detailed discussion)
608 J Comp Physiol A (2008) 194:597­609
123
Frank G, Kössl M (1996) The acoustic two-tone distortions 2f1­f2 and
f2­f1 and their possible relation to changes in the operating point
of the cochlear ampliWer. Hearing Res 98:104­115
Frank G, Hemmert W, Gummer AW (1999) Limiting dynamics of
high-frequency electromechanical transduction of outer hair cells.
Proc Natl Acad Sci USA 96:4420­4425
Frolenkov GI, Belyantseva IA, Kurc M, Mastroianni MA, Kachar B
(1998) Cochlear outer hair cell electromotility can provide force
for both low and high intensity distortion product otoacoustic
emissions. Hearing Res 126:67­74
Geleoc GSG, Holt JR (2003) Auditory ampliWcation: outer hair cells
pres the issue. Trends Neurosci 26:115­117
Gerhardt HC, Huber F (2002) Acoustic communication in insects and
anurans: common problems and diverse solutions. Univ of Chicago
Press, Chicago
Göpfert MC, Robert D (2001) Active auditory mechanics in mosquitoes.
Proc R Soc Lond B 268:333­339
Göpfert MC, Robert D (2003) Motion generation by Drosophila
mechanosensory neurons. Proc Natl Acad Sci USA 100:5514­
5519
Göpfert MC, Humphris ADL, Albert JT, Robert D, Hendrich O (2005)
Power gain exhibited by motile mechanosensory neurons in Drosophila
ears. Proc Natl Acad Sci USA 102:325­330
Göpfert MC, Albert JT, Nadrowski B, Kamikouchi A (2006) SpeciWcation
of auditory sensitivity by Drosophila TRP channels. Nat
Neurosci 9:999­1000
Gray EG (1960) The Wne structure of the insect ear. Phil Trans R Soc
Lond B 243:75­94
Hill KG (1983a) The physiology of locust auditory receptors. I. Discrete
depolarizations of receptor cells. J Comp Physiol 152:475­482
Hill KG (1983b) The physiology of locust auditory receptors. II. Membrane
potentials associated with the response of the receptor cell.
J Comp Physiol 152:483­493
Hoy RR, Robert D (1996) Tympanal hearing in insects. Ann Rev Entomol
41:433­450
Hudspeth AJ (2000) Hearing. In: Kandel ER, Schwartz JH, Jessel TM
(eds) Principles of neural science (4/e). McGraw-Hill, New York,
pp 590­613
Jackson JC, Robert D (2006) Nonlinear auditory mechanism enhances
female sounds for male mosquitoes. Proc Natl Acad Sci USA
103:16734­16739
Jacobs K, Otte B, Lakes-Harlan R (1999) Tympanal receptor cells of
Schistocerca gregaria: correlation of soma positions and dendrite
attachment sites, central projections and physiologies. J Exp Zool
283:270­285
Johnson AC, Canlon B (1994) Progressive hair cell loss induced by toluene
exposure. Hearing Res 75:201­208
Kemp DT (1978) Stimulated acoustic emissions from within the human
auditory system. J Acoust Soc Am 64:1386­1391
Kemp DT (2002) Otoacoustic emissions, their origin in cochlear function,
and use. Br Med Bull 63:223­241
Kennedy HJ, Evans MG, Crawford AC, Fettiplace R (2003) Fast adaptation
of mechanoelectrical transducer channels in mammalian
cochlear hair cells. Nat Neurosci 6:832­836
Kernan MJ (2007) Mechanotransduction and auditory transduction in
Drosophila. PXügers Arch Eur J Physiol 454:703­720
Kössl M (1994) Otoacoustic emissions from the cochlea of the `constant
frequency' bats Pteronotus parnellii and Rhinolophus rouxi.
Hearing Res 72:59­72
Kössl M, Boyan GS (1998a) Otoacoustic emissions from a non-vertebrate
ear. Naturwissenschaften 85:124­127
Kössl M, Boyan GS (1998b) Acoustic distortion products from the ear
of a grasshopper. J Acoust Soc Am 104:326­335
Kössl M, Coro F (2006) L1, L2 maps of distortion-product otoacoustic
emissions from a moth ear with only two auditory receptor neurons.
J Acoust Soc Am 120:3822­3831
Kössl M, Vater M (2000) Consequences of outer hair cell damage for
otoacoustic emissions and audio-vocal feedback in the mustached
bat. J Assoc Res Otolaryn 1:300­314
Kössl M, Coro F, Seyfarth E-A, Nässig WA (2007) Otoacoustic emissions
from insect ears having just one auditory neuron. J Comp
Physiol A 193:909­915
Kummer P, Janssen T, Hulin P, Arnold W (2000) Optimal L1­L2 primary
tone level separation remains independent of test frequency
in humans. Hearing Res 146:47­56
Lakes R, Schikorski T (1990) Neuroanatomy of tettigoniids. In: Bailey
WJ, Rentz DCF (eds) The Tettigoniidae: biology. systematics and
evolution. Crawford House Press, Bathurst, pp 166­190
LeMasurier M, Gillespie PG (2005) Hair-cell mechanotransduction
and cochlear ampliWcation. Neuron 48:403­415
Liao Z, Feng S, Popel AS, Brownell WE, Spector AA (2007) Outer
hair cell active force generation in the cochlear environment.
J Acoust Soc Am 122:2215­2225
Liberman MC, Gao J, He DZZ, Wu X, Jia S, Zuo J (2002) Prestin is
required for electromotility of the outer hair cell and for the cochlear
ampliWer. Nature 419:300­304
Long GR, Van Dijk P, Wit HP (1996) Temperature dependence of
spontaneous otoacoustic emissions in the edible frog (Rana esculenta).
Hearing Res 98:22­28
Lukashkin AN, Russell IJ (1999) Analysis of the f2­f1 and 2f1­f2 distortion
components generated by the hair cell mechanoelectrical
transducer: dependence on the amplitudes of the primaries and
feedback gain. J Acoust Soc Am 106:2661­2668
Lukashkin AN, Lukashkina VA, Russell IJ (2002) One source for distortion
product otoacoustic emissions generated by low- and highlevel
primaries. J Acoust Soc Am 111:2740­2748
Manley GA (2001) Evidence for an active process and a cochlear
ampliWer in nonmammals. J Neurophysiol 86:541­549
Manley GA (2006) Spontaneous otoacoustic emissions from freestanding
stereovillar bundles of ten species of lizard with small
papillae. Hearing Res 212:33­47
Manley GA, Köppl C, Johnstone BM (1993) Distortion-product otoacoustic
emissions in the bobtail lizard. I. General characteristics.
J Acoust Soc Am 93:2820­2833
Manley GA, Kirk DL, Köppl C, Yates GK (2001) In vivo evidence for
a cochlear ampliWer in the hair-cell bundle of lizards. Proc Natl
Acad Sci USA 98:2826­2831
Manley GA, Fay RR, Popper AN (eds) (2008) Active processes and
otoacoustic emissions in hearing. Springer handbook of auditory
research, vol 30. Springer, New York
Meenderink SWF, Van Dijk P (2005) Characteristics of distortion
product otoacoustic emissions in the frog from L1, L2 maps. J Acoust
Soc Am 118:279­286
Michelsen A (1971) The physiology of the locust ear. I. Frequency sensitivity
of single cells in the isolated ear, II. Frequency discrimination
based upon resonances in the tympanum, III. Acoustical
properties of the intact ear. Z Vergl Physiol 71:49­128
Möckel D, Seyfarth E-A, Kössl M (2007) The generation of DPOAEs
in the locust ear is contingent upon the sensory neurons. J Comp
Physiol A 193:871­879
Montealegre-Z F, Morris GK, Mason AC (2006) Generation of extreme
ultrasonics in rainforest katydids. J Exp Biol 209:4923­
4937
OldWeld BP (1982) Tonotopic organisation of auditory receptors in
Tettigoniidae (Orthoptera: Ensifera). J Comp Physiol 147:461­
469
OldWeld BP (1988) Tonotopic organization of the insect auditory pathway.
Trends Neurosci 11:267­270
Pollack GS, Imaizumi K (1999) Neural analysis of sound frequency in
insects. BioEssays 21:295­303
Ricci AJ, Crawford AC, Fettiplace R (2002) Mechanisms of active hair
bundle motion in auditory hair cells. J Neurosci 22:44­52
J Comp Physiol A (2008) 194:597­609 609
123
Römer H (1976) Die Informationsverarbeitung tympanaler Rezeptorelemente
von Locusta migratoria (Acrididae, Orthoptera). J
Comp Physiol 109:101­122
Russell IJ, Kössl M (1999) Micromechanical responses to tones in the
auditory fovea of the greater mustached bat's cochlea. J Neurophysiol
82:676­686
Shera CA, Guinan JJ Jr (1999) Evoked otoacoustic emissions arise by
two fundamentally diVerent mechanisms: a taxonomy for mammalian
OAEs. J Acoust Soc Am 105:782­798
Stumpner A (1996) Tonotopic organization of the hearing organ in a
bushcricket. Naturwissenschaften 83:81­84
Stumpner A, von Helversen D (2001) Evolution and function of auditory
systems in insects. Naturwissenschaften 88:159­170
Surlykke A (1984) Hearing in notodontid moths: a tympanic organ
with a single auditory neurone. J Exp Biol 113:323­335
Taschenberger G, Manley GA (1997) Spontaneous otoacoustic emissions
in the barn owl. Hearing Res 110:61­76
Weber T, Göpfert MC, Winter H, Zimmermann U, Kohler H, Meier A,
Hendrich O, Rohbock K, Robert D, Knipper M (2003) Expression
of prestin-homologous solute carrier (SLC26) in auditory organs
of nonmammalian vertebrates and insects. Proc Natl Acad Sci
USA 100:7690­7695
Windmill JFC, Göpfert MC, Robert D (2005) Tympanal travelling
waves in migratory locusts. J Exp Biol 208:157­168
Yack JE (2004) The structure and function of auditory chordotonal
organs in insects. Microsc Res Tech 63:315­337
Zheng J, Shen W, He DZZ, Long KB, Madison LD, Dallos P (2000)
Prestin is the motor protein of cochlear outer hair cells. Nature
405:149­155