Transmission and perception of acoustic signals in ...

4 downloads 14502 Views 751KB Size Report
low, and any variation in the male's call will center about a species' mean to ... we here measure the maximum hearing distance of the call of L. coquilletti in.
Journal of lnsect Behavior, VoL 6, No. 2, 1993

Transmission and Perception of Acoustic Signals in the Desert Clicker, Ligurotettix coquilletti (Orthoptera: Acrididae) W. J. Bailey, 1 M. D. Greenfield, z and T. E. Shelly 3 Accepted March 24, 1992; revised April 28, 1992

Males of the desert clicker, Ligurotettix coquilletti (Acrididae: Orthoptera) defend a female-required resource, the creosote bush Larrea tridentata, in desert habitats of the southwestern United States. Males signal acoustically to each other as well as to searching females. The call is produced by tegminal/femoral stridulation where one or both legs are used for sound production. Sound pressure levels, measured laterally, are influenced by the intervening tegmen between the stridulating leg and the microphone. Differences in measured sound pressure levels between sides can vary up to 7 dB. When clicks are produced multiply, these multiple clicks may be 4 dB louder than single clicks. We examine the structure of the call and the effective broadcast area of single males by monitoring acoustic ascending neurons of the ventral nerve cord in the neck. By taking the neurophysiological preparation into the fieM, we were able to map the broadcast area of isolated males and also of males calling within aggregations. The distance over which the signal of isolated males could be detected was 8-14 m, whereas neural representation of the calls of males within aggregation were detectable within 4-6 m. The sound spectrum of the song, although having a major lower-frequency component around 10 kHz, has extensive power in the ultrasonic range. The tuning characteristics of the ascending auditory Department of Zoology, University of Western Australia, Nedlands, Western Australia 6009. 2Department of Entomology, Division of Biological Sciences, The University of Kansas, Lawrence, Kansas 66045-2106. Address at time of study: Department of Biology, University of California, Los Angeles, California 90024. 3Hawaiian Evolutionary Research Center, University of Hawaii, Gilmore 310, Honolulu, Hawaii 96822.

141

0892-7553/93/0300-0141507.00/09 1993PlenumPublishingCorporation

142

Bailey, Greenfield, and Shelly

neuron matched the overall structure o f the male call. The importance o f the acoustic cue, as compared to visual cues, is discussed in relation to female attraction. KEY WORDS: Acrididae,Ligurotettix coquilletti, soundproduction,broadcastarea, resourcede-

fence.

INTRODUCTION The songs of male acridid grasshoppers are used to attract conspecific females (Otte, 1977; Riede, 1987), and in these circumstances the song provides cues as to species' identity (Skovmand and Pederson, 1978; Helversen and Helversen, 1983; Butlin, 1989) and the location of the male (Ronacher et al., 1986). Song also has the potential to reveal information on the male's competitive status, where components of the call are correlated with male fitness (Searcy and Anderssen, 1986). Studies on nocturnal ensiferan Orthoptera, such as those for gryllids and tettigoniids, have shown that certain aspects of the call have the potential to be under selection through female choice. But cues used in this way should provide on average consistent evidence to the female of male fitness under the complex conditions of the animal's natural habitat (cf. Gerhardt, 1991). Such cues include the call's duration [Grylloidea (Hedrick, 1986)], its carrier frequency [Tettigoniidae (Latimer and Sipple, 1987; Bailey and Yeoh, 1989)], and its intensity [Tettigoniidae (Bailey et al., 1990)]. In comparison, for diurnal orthopterans the auditory cue takes on a less significant role in intersexual selection, for in daylight females can locate the approximate position of their mates by sound, and if female choice occurs, they can make a final assessment of a male through visual signals (Otte, 1977). Predictably then, selection on the call through intersexual selection should be low, and any variation in the male's call will center about a species' mean to which females will be attracted (Butlin et al., 1985). Ligurotettix coquilletti is a diurnal grasshopper, common in the southwestern part of north America, where it calls from the creosote bush Larrea tridentata (Otte and Joern, 1975; Greenfield et al., 1987). From here males call either singly or in groups of up to five within one bush. Males in such aggregations have been referred to as "dominion" owners (Brown, 1975; Shelly and Greenfield, 1991), since they consist of mutually aggressive individuals that remain on and defend overlapping portions of the same resource. Females appear in higher numbers on bushes harboring groups rather than at those containing isolated males (Greenfield and Shelly, 1985). There are two possible reasons for this. One is that groups are acoustically more obvious than the calls of isolated males (Shelly and Greenfield, 1991), and the second is that females are

Acoustic Signaling by the Desert Clicker

143

attracted to certain creosote bushes because of the plant's high nutritional value. The second of these hypotheses was considered unlikely since the number of females arriving at low- and high-quality bushes from which loudspeakers broadcast the male's song was the same. Further, females showed an attraction to these playback calls that was contingent on intensity (Shelly and Greenfield, 1991); there was no evidence that females could detect nutrient rich bushes from a distance through smell alone. Where female attraction is dependent on the call of either a single male or the collective sound of a dominion, it is instructive to know the range over which these calls are effective and also whether females can distinguish individual males within these groups. The effective distance of the call will be dependent on five factors: (i) the intensity of the outgoing signal, (ii) any inherent directional characteristics of the emitter, (iii) the effects of attenuation through spherical spreading, (iv) the degree of signal degradation due to the scattering of sound from structures within the habitat (Wiley and Richards, 1982), and finally, (v) both the hearing sensitivity of the female and its ability to distinguish and localize a conspecific's call. If the female detects several males, either calling singly from a number of closely distributed bushes or from within a group on one bush, she may selectively isolate the call of one individual (vide Pollack, 1988; R t m e r and Bailey, 1986, 1990). Her ability to distinguish any one male from the calls of others will depend, at least in part, on the proportion of time the male spends calling as well as the call's intensity. For where intensity changes are random, due to the male either changing its position or altering the way in which the signal is produced, localization by the female using intensity cues will be far more difficult (cf. Gerhardt et al., 1989). By using the animal's hearing system as a living microphone (Rheinlaender and Rtmer, 1986), measures of the distance over which the call may be heard, the maximum hearing distance, have been made in tettigoniids (Rtmer and Bailey, 1986) and frogs (Brenowitz et al., 1984). In both cases the area over which the call was projected was influenced by the position from which the male called within the habitat: those males calling from a higher perch had a measurable increase in effective range. Using a similar neurophysiological assay we here measure the maximum hearing distance of the call of L. coquilletti in the feld. The advantage of such a simple technique is that measurements of threshold can be made within the insect's natural environment and are therefore subject to the stochastic effects of noise and signal degradation. We equate hearing distance of individual insects with the output from a male. We measure both the directional characteristics of a male's call and any attenuating properties of the habitat. Where males were calling close to their conspecific neighbors, we assess whether an approaching L. coquilletti can distinguish the call of a single male from the calls of other males within an aggregation. By measuring the insect's hearing threshold we were able to calculate the maximum theoretical

144

Bailey, Greenfield, and Shelly

hearing distance based on this threshold and this value could then be compared to the maximum hearing distance measured in the field. MATERIALS AND METHODS

Study Site Field work was carded out in the alluvial plain at the Boyd Deep Canyon Desert Research Center, 10 km south of Palm Desert, California. The climate and flora of this site are representative of the Californian portion of the Sonoran Desert (Zabriskie, 1979). Larrea bushes at the study site average 2.3 m in width and 1.7 m in height. There are approximately 190 bushes ha -1, and the mean nearest-neighbor distance (edge to edge) is 2.4 m. Air temperatures taken at midday during the study periods ranged from 37 to 43~ in the shade.

Signal Intensity, Attenuation, and Frequency Spectrum Sound pressure levels (SPL; 0 dB re 20 /~Pa) of calls produced by L.

coquilletti males were measured in the field with a General Radio 1982 sound level meter. The detector response of the meter, operated with an octave-bandwidth frequency filter centered on 8 kI-Iz, was set on " p e a k " to provide a response time of less than 50/~s. Although this limited our measurements to the sonic components of the call, our observations concerned long-distance communication, and in this regard we may presume that high frequencies, because of the severe attenuation they suffer compared to lower frequencies (Wiley and Richards, 1982), would be of much less significance. Additionally, during earlier field playback trials females were attracted to speakers delivering only this sonic part of the call (Shelly and Greenfield, 1991). L. coquilletti produces sound by the tegminal-femoral stridulation typical of most gomphocerine grasshoppers. Single clicks are produced by either the movement of one hindleg or the simultaneous movements of both. Occasionally, multiple clicks, successions of two to four stridulations spaced at approximately 50-ms intervals, are produced. The successive stridulations in multiple clicks are generally made by alternate hindlegs, but sometimes the same hindleg is used. We measured the intensity of single clicks produced by one or both hindlegs and noted the difference between measurements made ipsi- and contralateral to the stridulating leg. Additionally, we compared the intensity of single clicks measured ipsilaterally to that of multiple clicks. All SPL values were measured 0.5 m from the insect, and the mean of five values was obtained for each type of measurement. Only solitary insects, those not courting females or aggressively interacting with neighboring males, were used. Immediately

Acoustic Signaling by the Desert Clicker

145

following SPL determination, we measured the air temperature at the insect's perch with a Bailey BAT-12 thermocouple thermometer. The insect was captured and the hind femur was measured to the nearest 0.05 mm with Helios callipers. We measured the attenuation of L. coquilletti clicks due to vegetation by broadcasting prerecorded stridulations through a loudspeaker (Nagra DSM monitor) alternately positioned in front of and adjacent to a Larrea bush. In both situations SPL of the broadcast was measured 3 m from the loudspeaker at a height of 1 m. Each measurement was repeated five times on four bushes when winds were very light. A single click of a L. coquilletti male singing in a laboratory was recorded via a Racal Store 4S tape recorder operated at 152.4 cm s -1 and fitted with a Bmel and Kjaer 3135 (6.25-mm) microphone (flat response to 100 kHz). The temporal structure and frequency spectrum of the recording was analyzed via a custom built A-D board. Although the temperature at which these calls were recorded (27~ is lower than that experienced in the field, we are confident that the carrier frequency and the detail of the pulse structure within each click are representative of the sounds measured in the field.

Acoustic R e c e p t i o n in the Field

During July 1988 we employed a simple neurophysiological apparatus mounted on a battery-driven Iselworth preamplifier to monitor the response of a L coquilletti "preparation" to acoustic stimuli in the field. The preparation consisted of an insect oriented ventral side up, with its ventral nerve cord (VNC) revealed by dissection in the cervical region. An unshielded steel hook electrode lifted the VNC at a point near the suboesophageal ganglion, and an indifferent silver electrode was inserted in the abdomen. The VNC was severed anterior to the suboesophageal ganglion, and the wound covered with petroleum jelly to prevent desiccation. Such preparations usually lasted 1-3 h in the field, and one remained viable for 9 h. If the preparation survived the initial hour, the signalto-noise ratio of the summed action potentials to background neural activity improved. The response patterns we observed were similar to those obtained from the serially homologous G and B1 neurons of the locust (Boyan, 1984). The amplified action potentials were recorded on one track of a Sony Walkman Professional cassette recorder and monitored aurally by headphones. Periodically, the signal-to-noise ratio of the recorded signal was checked by a portable Tektronix oscilloscope. The second channel of the cassette recorder (and headphones) received the signal of the acoustic stimulus via a microphone (QMC ultrasound detector) placed within 2 m of the focal male. We found no difference in the auditory response of either sex and so used both sexes for field analysis.

Bailey, Greenfield, and Shelly

146

Maximum Hearing Distance We estimated the greatest distance over which L. coquilletti could hear the sonic components of the conspecific signals in either of two ways. First, the neurophysiological preparation, with the effective tympanum ipsilateral to a focal male, was hand-held approximately 1 m from an isolated calling male. We then walked away from this male to a point where the signal was no longer discernible through the headphones. This procedure was repeated over a number of radial axes to chart the perimeter of the "broadcast area" surrounding the focal male. This distance indicated the maximum distance over which the preparation could discern the call and presumably the distance over which it could be heard by a conspecific. An observer, stationed close to the calling male, indicated the timing of the clicks and also noted whether the animal moved. In the latter case, measurements were discontinued. We produced maps of broadcast areas for eight focal males, using a different preparation for each. When males were calling from bushes that were close to the focal male it was often difficult, by aural monitoring, tO identify the neural response of the preparation to each click. Under these conditions we employed a second method where the call (stimulus) was recorded through the microphone onto one track of a Sony Walkman tape recorder and the neural activity of the preparation on the other. The preparation was then moved away from the focal male along a radial axis. A poststimulus time histogram (PSTH) was used to analyze the relationship between call and nerve response. This technique employs a custombuilt AD board with appropriate software. It takes the recorded call as a trigger and measures the interval between this event and the neural responses of the hearing system. Where the response of the preparation is constant with regard to the stimulus, the counted neural events will form a histogram whose mode will represent the temporal response to the call. Where there is no maximum, then it may be assumed that the call is lost within background noise: any neural response will become random with respect to the stimulus. The maximum hearing distances of 16 focal males, using 10 different preparations, were determined by this method. We abandoned the test if the focal male moved during the recording.

Acoustic Sensitivity We tested the nerve responses of preparations from five insects to a range of pure frequencies. Repeated pulses of sound (200-ms duration with an 800ms interpulse interval with rise and decay times of 20 ms) were broadcast via a Radio Shack 40-1310A loudspeaker (flat response from 5 to 40 kHz _+ 3 dB). Each preparation was oriented with the effective tympanum ipsilateral to the loudspeaker located 2 m distant. Threshold responses were measured aurally via

Acoustic Signaling by the Desert Clicker

147

headphones. The responses to stimulus frequencies between 5 and 45 kHz were measured in steps of 5 kHz. Absolute SPLs of the 20-kHz signals were measured with the sound level meter.

RESULTS The Song of Ligurotettix coquilletti The acoustic signaling of L. coquilletti is best described as an incessant production of clicks that vary in amplitude and interval. There are two extreme forms of sound production, single clicks with long interclick intervals and sequences of more rapid double or multiple clicks. Single clicks are produced at a rate of 30 clicks min-~: the mean interval between single clicks is 1.9 _+ 1.0 (SD) s -~ (Fig. 1A). Either leg may be used to produce these clicks and each click is formed by a series of four or five pulses lasting 20-30 ms (Fig. 1B). Males may intersperse these long sequences of single clicks with double or multiple clicks where both legs are used. The interval between multiple clicks, measured for isolated males, again shows considerable variation (mean interval, 0.08 _+ 0.06 s-1). We observed that males tend to produce more multiple clicks when other males were on the same bush than when calling alone. The frequency spectrum of the song measured from a single click shows a high-energy component between 8 and 20 kHz and then a broad spectrum of high frequencies with a second maximum around 30 kHz (Fig. 2A). Side had a consistent and significant effect on call intensity (Figs. 1A and C). Single clicks measured ipsilateral averaged 7.4 dB (+2.13-1.63: n = 4 males) louder than those measured contralaterally. The absolute intensity of these calls, measured ipsilaterally, ranged from 66.4 to 68.3 dB over 0.5 m. Comparisons between the intensity of multiple and that of single clicks showed that multiple clicks were 2.8 to 4.7 dB louder than single clicks, and further clicks produced by synchronous movements of both hindlegs were 1.5 to 7.2 dB louder (n -- 3) than single clicks produced by one hindleg. In addition to side and synchrony of leg movements, several other factors contributed to interindividual variation in call intensity. Size was found to influence call intensity, with larger males calling more loudly [SPL = 1.77 (hind femur) + 42.18; r 2 = 0.241, t = 3.435, d f 3 3 , P < 0.01). Moreover, it was shown elsewhere (Shelly and Greenfield, 1991) that L. coquilletti can vary the intensity of their calls in response to male density: males call more loudly when sharing occupancy of a Larrea bush with other calling males than when isolated. Finally, the foliage of a Larrea bush attenuates song intensity by 1.6-2.2 dB over a 3-m distance compared with sound broadcast over open ground (MannWhitney one-tailed U = 15, P = 0.05; n = 4, 4).

Bailey, Greenfield,and Shelly

148

A B

20 ms

C

D

Tr,lrr,:Tr 1 s

Fig. 1. The call of Ligurotettix coquilletti. (A) Three single clicks recorded lateral to the insect. The first and third are produced by the leg closer to the microphone (ipsilateral) and the second by the contralateral leg. (B) The temporal structure of one click consisting of five sound pulses (recorded in the laboratory). (C) Field recording of multiple clicks of a single male recorded lateral, showing variation in intensity dependent on the leg used by the male. (D) The complex acoustical environment within a dominion; the calls of three males at the edge of a Larrea bush.

Tuning of the Receptors U s i n g the h o o k e d n e c k c o n n e c t i v e s with the preparation ipsilateral to a previously calibrated sound source, w e w e r e able to establish that the tuning o f the receptor system had m i n i m u m values at 10 and 25 k H z (Fig. 2B). The threshold a p p r o x i m a t e d 45 dB for frequencies close to 25 k H z and 48 dB at 10 kHz.

Acoustic Signaling by the Desert Clicker

149

(A) 0 Cl3 "13

-lo t-cD "~ - 2 0 113 "- -30 0 o 0:2 -40 0

50

25

Frequency

kHz

(B) lO

0 d8 :

60 dB SPL

or7 "o

o c

-5

0) .->

-10.

o "n," ~

-15. -20

0

,

,

I0

t

2'0

~

~

3~0

430

5~0

Frequency kHz F i g . 2. (A) The sound spectrum of a single click o f a m a l e L. coquilletti. (B) Tuning characteristics of

the ascending auditory interueurons of L. coquilletti ( _ S E ; n = 5).

Maximum Hearing Distance The effective distance of the call of an isolated singing male, as determined by the maximum hearing distance, ranged between 4.5 and 14.3 m (Fig. 3A). The polar plots we obtained from these measurements showed considerable differences in effective distances between individual males as well as the influence of side on the plot's symmetry (Fig. 3B). Any asymmetry is predicted from our data on the relationship between side and intensity as well as effects due to attenuation through foliage. Maximum hearing distances are approxi-

150

Bailey, Greenfield, and Shelly

(A) 12 t~

L

8.

m

E 6L

4-

E z

2 4 Maximum

6 fl 10 12 14 Hearing Distance m

16

03)

Male # 3

Fig. 3. Maximum heating distance. (A) Histogram of measured hearing distances of eight isolated males as estimated by auditory threshold. (B) Directional plots of three males showing the effects of orientation on the effective distance of the call. Males were calling from stems in a near-vertical or -horizontal position, with the tight of the figure as ventral for male 1 and the top of the figure anterior for males 2 and 3.

mately two to six times the m e a n nearest-neighbor distance b e t w e e n calling males situated on neighboring bushes (Greenfield and Shelly, 1985). M a x i m u m hearing distances o f isolated males that were measured by P S T H analysis support data obtained by aural monitoring. Using this technique we found that the calls of males within an aggregation were masked by those of neighbors at a m e a n distance o f 7.3 _ 1.3 m (n = 6). However, for isolated

Acoustic Signaling by the D e s e r t C l i c k e r

151

1

m

3 m

L ~ rm~,

6 m

,llll

9 m . . . . . . . . . . . ILL.__1..... ~L . . . . .............. gr ~T . . . . . . J'. . . . . . . .

12

m

I

Fig. 4. Oscillographic analysis of the neural response to a focal male at increasing distances. The m a l e ' s call recorded by the microphone is on the lower trace and the nerve response on the upper trace. The increasing delay between stimulus and response corresponds to the sound transmission time between the preparation and the focal male.

u

50

ms

males this distance was much further (mean = 12.3 + 2 m; n = 4). Figure 4 provides representative data from one nerve recording to demonstrate how the signal of an isolated male is still detectable by the nerve preparation at 12 m.

DISCUSSION To the orienting female, the clicks of males provide information for locating both mating opportunities and food resources. It is also conceivable that females may use these distant sound cues in a comparative sense to choose a better male and/or a preferred plant. For many acoustically orienting insects the intensity of a conspecific's call is an overriding determinant for phonotaxis (Latimer and Sipple, 1987; Bailey and Yeoh, 1989), but for L. coquilletti intensity is highly

152

Bailey, Greenfield, and Shelly

variable, changing by as much as 8 dB for one animal at one location (cf. Helversen and Helversen, 1983). Thus the effective range of the song will vary with the male's position on the perch, the side presented to the female, whether clicks are produced singly or in multiples and if calls are produced by synchronous movements of the legs. Further, where males are aggregated, they may produce on one bush more multiple clicks and so call more loudly than solitary males (Shelly and Greenfield, 1991). It is arguable that once females are within an aggregation of males, the call may be of significance to its choice between competing males. But extreme intraindividual variation makes this unlikely, and we propose that once the female is on the same bush as the male, vision will take over as the primary cue for attraction, and if sound were to play any part in female choice there would be a high risk of cheating. For example, an intruding smaller male could use the contralateral (with respect to the resident male) leg for advertising its presence to a distant female. This call would represent a quiet, distant male to the resident but a loud close male to an approaching female. Based on the distribution of males throughout the habitat and the regular spacing of Larrea bushes, we suggest that a female could hear not only the calls of its immediate neighbors, but also those of males in bushes much further away. The mean edge-to-edge distance between bushes of is 2.4 m, and if the sound loss through spherical spreading accounts for 6 dB for each doubling of distance, a male's call of 65-dB SPL measured at 1 m would be close to 56 dB at a neighboring bush. Fifty-six decibels is still 10 dB above the measured physiological threshold, and if there were intervening bushes between caller and receiver, this value could be diminished to 52-54 dB. (We have already noted that sound loss through a single bush over a distance of 3 m was 1.6 to 2.2 dB.) The maximum hearing distance of an isolated male was 14 m, and at this distance sound levels from a calling male producing 65-dB SPL at 1 m would be close to 43 dB, which approximates the measured threshold obtained in the laboratory for pure tones. Given the insect's ability to detect more than one isolated male, an approaching female will be presented with a complex of clicks at different intensities and from different directions. Where the distance between males falls beneath 5-6 m, each male's unique temporal pattern will be lost to the searching female within a clatter of conspecific clicks. The summed action potentials of the ascending neuron follow the focal male's clicks to a distance of 4-6 m, after which the receptors responded as much to the clicks of neighbors as to those of focal males: the response was random with respect to the call of the focal male. But these limits will be dependent on the behavior of the male, its posture, the leg used in stridulation, and the occurrence of multiple clicks: intensity will go through extremes of variation.

Acoustic Signaling by the Desert Clicker

153

Based on these results and on those o f other studies, we propose the following scenario for mate attraction in L. coquilletti. W e expect that initial settlement decisions by males will be influenced by the presence or absence o f conspecific stridulation, and these decisions may be reinforced by the chemistry o f the plant (Greenfield et al., 1987). Once within a bush, a territorial male defends only that bush, and does so mainly through visual detection of intruding males. The unreliable association between intensity and distance suggests that visual rather than acoustic cues provides a better indicator o f distance, defining an economically defensible unit. The acoustic cues of very close intruders have been found to stimulate aggression from resident males (Wang, G . - y . , and Greenfield, M. D., 1991), and this may lead to an increase in the ratio o f multiple to single clicks. While possibly yielding information on male quality in certain contexts, stridulation may actually present relatively useless cues for regulating resource defense, an unexpected finding in this territorial insect. It is also questionable whether the calls o f individuals are robust enough to act as a cue for female choice. As with other acoustical animals attraction may be to the most intense signal (Parker, 1983; Arak, 1988; Bailey e t a l . , 1991). This attraction will draw her to bushes o f high nutritional value occupied by groups o f reproductively active males.

ACKNOWLEDGMENTS W e thank H. R6mer for advice and discussion on the use o f our field recording equipment, and to Hayward Spangler for a high fidelity recording of L. coquilletti. W e acknowledge the support o f Australian Research Committee (W.J.B.) Grant A 18315120 and the United States National Science Foundation grants BSR 83-05824 and BSR 86-00606 ( M . D . G . ) and BSR 86 12325 (T.E.S.).

REFERENCES Arak, A. (1988). Female mate selection in the natterjack toad: Active choice or passive attraction? Behav. Ecol. Sociobiol. 22: 317-327. Bailey, W. J., and Yeoh, P. B. (1988). Female pbonotaxis and frequency discrimination in the bushcricket Requena verticalis. Physiol. Entomol. 13: 363-372. Bailey, W. J., Cunningham, R. J., and Lebel, L. (1990). Song power, spectral distribution and female phonotaxis in the bushcricket Requena verticalis (Tettigoniidae: Orthoptera): Active female choice or passive attraction. Anim. Behav. 40: 33-42. Brenowitz, E. A., Wilczynski, W., and Zakon, H. H. (1984). Acoustic communication by spring peepers. Environmental and behavioral aspects. J. Comp. Physiol. 155: 585-592. Brown, J. L. (1975). The Evolution of Behavior, Norton, New York. Butlin, R. K. (1989). Reinforcement of premating isolation. In Otte, D., and Endler, J. A. (eds.), Speciation and Its Consequences, Sinauer Press, Sunderland, Mass., pp. 158-179. Butlin, R. K., Hewitt, G. M., and Webb, S. F. (1985). Sexual selection for intermediate optimum in Chorthippus brunneus (Orthoptera: Acrididae). Anim. Behav. 33: 1281-1292.

154

Bailey, Greenfield, and Shelly

Boyan, G. S. (1984). Neural mechanisms of auditory information processing by identified interneurons in Orthoptera. J. Insect Physiol. 30: 27-41. Gerhardt, H. C. (1991). Female mate choice in treefrogs: Static and dynamic acoustic criteria. Anim. Behav. 42: 615-635. Gerhardt, H. C., Diekamp, B., and Ptacek, M. (1989). Inter-male spacing in choruses of the spring peeper, Pseudacris (Hyla) crucifer. Anim. Behav. 38: 1012-1024. Greenfield, M. D., and Shelly, T. E. (1985). Alternative mating strategies in a desert grasshopper: Evidence for density dependence. Anim. Behav. 33: 1192-1210. Greenfield, M. D., Shelly, T. E., and Downum, K. R. (1987). Variation in host plant quality: Implications for territoriality in a desert grasshopper. Ecology 68: 828-838. Hedrick, A. V. (1986). Female preferences for calling bout duration in a field cricket. Behav. Ecol. Sociobiol. 19: 73-77. Helversen, D. von (1984). Parallel processing in auditory pattern recognition and directional analysis by the grasshopper Chorthippus biguttulus L. (Acrididae). J. Comp. Physiol. 154: 837-846. Helversen, O. von, and Helversen, D. (1983). Species recognition and acoustic localisation in acridid grasshoppers: A behavioural approach. In Huber, F., and Markl, H. (eds.), Neuroethology and Behavioural Physiology, Springer Verlag, Berlin, pp. 95-107. Latimer, W., and Sippel, M. (1987). Acoustic cues for female choice and male competition in Tettigonia cantans. Anita. Behav. 35: 887-910. Otte, D. (1977). Communication in Orthoptera. In Sebeok, T. (ed.), How Animals Communicate, Indiana University Press, Bloomington, pp. 334-361. Otte, D., and Joern, A. (1975). Insect territoriality and its evolution: Population studies of desert grasshoppers on creosote bushes. J. Anim. Ecol. 44; 29-54. Parker, G. A. (1983). Mate quality and mating decisions. In Bateson, P. (ed.), Mate Choice, Cambridge University Press, Cambridge, pp. 141-164. Pollack, G. S. (1988). Selective attention in an insect auditory neuron. J. Neurosci. 8: 2635-2639. Riede, K. (1987). A comparative study of mating behaviour in some Neotropical grasshoppers (Acridoidea). Ethology 76: 265-296. Rheinlaender, J., and Rrmer, H. (1986). Insect heating in the field. I. The use of identified nerve cells as "biological microphones." J. Comp. Physiol. 158: 647-651. Rrmer, H., and Bailey, W. J. (1986). Insect hearing in the field. II. Male spacing behavior and correlated acoustic cues in the bushcricket Mygalopsis marki. J. Comp. Physiol. 159: 627638. Rrmer, H., and Bailey, W. J. (1990). Insect heating in the field. Comp. Biochem. Physiol. 97: 443-447. Rrmer, H., Bailey, W., and Dadour, I. (1989). Insect heating in the field. III. Masking by noise. J. Comp. Physiol. 164: 609-620. Rrmer, H., Rheinlaender, J., and Dronse, R. (1981). Intracellular studies on auditory processing in the metathoracic ganglion of the locust. J. Comp. Physiol. 144: 305-312. Ronacher, B., Helversen, D. von, and Helversen, O. von (1986). Routes and stations in the processing of auditory directional information in the CNS of a grasshopper, as revealed by surgical experiments. J. Comp. Physiol. 158: 363-374. Searcy, W. A., and Anderssen, M. (1986). Sexual selection and the evolution of song. Annu. Rev. Ecol. Syst. 17: 507-533. Skovmand, P., and Pederson, S. B. (1978). Tooth impact rate in the song of shorthorned grasshoppers: A factor carrying specific behavioural information. J. Comp. Physiol. A 124: 27-36. Shelly, T. E., and Greenfield, M. D. (1991). Dominions and desert clickers (Orthoptera: Acrididae): Influences of resources and male signaling on female settlement patterns. Behav. Ecol. Sociobiol. 28: 133-140. Shelly, T. E., Greenfield, M. D., and Downum, K. R. (1987). Variation in host plant quality: Influences on the mating tactics in male grasshoppers. Behaviour 109: 1200-1209. Wang, G.-y., and Greenfield, M. D. (1991). Effects of territory ownership on dominance in the desert clicker (Orthoptera: Acrididae). Anim. Behav. 42: 579-587. Wiley, R. H., and Richards, D. G. (1982). Adaptations for acoustic communication in birds: Sound transmission and signal detection. In Kroodsma, D. E., Miller, E. H., and Ouellet, H. (eds.), Acoustic Communication in Birds, Vol. 1, Academic Press, New York, pp. 131-181. Zabriskie, J. G. (1979). Plants of Deep Canyon and the Central Coachella Valley, California, University of California Press, Riverside.