Figure 1.
Dominant male Astatotilapia burtoni produce pulsed sounds during courtship quiver behaviors towards females.
A) Photograph of a yellow dominant male in front of his pot territory performing a quiver display and courtship sound while presenting his anal fin egg-spots (arrow) towards two nearby, and attentive, gravid females. H, hydrophone. B) Representative waveform (top) and spectrogram (bottom) of a pulsed broadband courtship sound produced by a dominant male.
Figure 2.
Dominant male A. burtoni produce intentional sounds during courtship quivers.
A) Examples of the temporal sequence of courtship sounds and quiver behaviors produced by two individual males of different sizes. Top graph shows a small male (SL = 55 mm) that produced courtship sounds during ∼40% of behavioral quiver displays, while the bottom graph shows a larger male (SL = 82 mm) that produced sounds during ∼80% of quivers. Each vertical mark represents a single courtship sound or quiver behavior during the 30 min trial. B) Relationship between the percentage of quiver behaviors associated with courtship sounds and male standard length (SL) shows that larger males produce a greater proportion of behaviors with sounds than do smaller males.
Figure 3.
Characteristics of courtship sounds produced by male A. burtoni during quiver behaviors.
A) Relationship between mean peak frequency (Hz) of sounds and male body size (standard length) shows that larger males produce lower frequency sounds. Each point represents the mean±SE of several sounds produced by an individual fish. B) There is a positive linear relationship between the number of pulses per sound and total sound duration (ms).
Table 1.
Summary of characteristics of courtship sounds produced by dominant male Astatotilapia burtoni.
Figure 4.
Example of auditory evoked potential (AEP) traces recorded from A. burtoni.
Averaged AEP traces from a representative subordinate male in response to a 200 Hz stimulus at several different intensities. An averaged trace from a control dead fish at 120 dB in response to a 200 Hz stimulus shows no response. Bottom trace shows the actual stimulus waveform recorded by the hydrophone at the position of the fish head. Threshold at this frequency was 105 dBrms re: 1 µPa based on the repeatable waveform and the presence of an FFT peak at twice the stimulus frequency.
Figure 5.
Hearing thresholds in the cichlid fish A. burtoni.
A) Hearing thresholds for subordinate and dominant males show similar responses, but subordinate males had lower thresholds at 600–800 Hz. B) Hearing thresholds for females show that receptive gravid individuals have lower thresholds at low frequencies from 100–600 Hz compared to mouthbrooding females. Threshold data are plotted as mean±SE (left axis). Asterisks indicate statistical differences between reproductive states within a sex at each test frequency (p<0.05). Gray overlay lines represent the power spectra (128 point FFT, Hanning window) of a representative courtship sound and are plotted as relative amplitude in dB (right axis) for comparison of sound spectral energy to hearing thresholds. N = 8 fish for each reproductive state (for females) and social status (for males).
Table 2.
Correlations between auditory evoked potential hearing threshold, circulating sex steroid levels, and gonadosomatic index (GSI) in the African cichlid fish Astatotilapia burtoni.
Figure 6.
Gravid sexually receptive females prefer males associated with playbacks of courtship sounds over noise controls.
A) Experimental tank setup for female preference trials. A gravid female was placed in the center compartment with size-matched territorial dominant males on either side. An underwater speaker in one of the male compartments played either male courtship or noise control sounds to the female while she was visually isolated. Following playback, the opaque barriers were removed and the time spent by the female in neutral and preference zones, as well as male reproductive behaviors were recorded and analyzed. B) Gravid females showed a greater relative preference (preference index) for males associated with the playback of courtship sounds compared to playbacks of control noise sounds. Data are plotted as mean±SE. N = 10 gravid females per sound condition. C) Examples of the playback stimuli used for courtship sounds and brown noise control trials. Waveforms (top) and spectrograms (bottom) for a 1 min section of each sound type are shown.