Finally, determination of SPL during playback was simultaneously video recorded, and positions of the camera and the observer were the same in both playback and recording trials. Sound recordings were analysed using Cool Edit 2000 (Syntrillium Software Corporation, Phoenix, AZ,

USA) and S_TOOLS-STx 3.7.8 (Acoustics Research Institute, Austrian Academy of Sciences, Vienna, Austria). We analysed 3–10 feeding clicks, Selleck Neratinib 1–12 courtship clicks and 30 sound pulses of growls per animal recorded (sampling rate 44.1 kHz). Pulse duration was analysed in all behavioural contexts. In distress growls, we measured pulse period (time between maximum peaks of consecutive pulses within a growl) and repetition rate (number of pulses s−1). The dominant frequency of the sounds was determined through cepstrum-smoothed power spectra (Noll, 1967). In order to attenuate the effect of tank resonances, all recordings were low pass filtered (3000 Hz) (Akamatsu et al., 2002). All data were verified for normal distribution and homogeneity of variances using Shapiro–Wilk’s and Levene’s tests, respectively. When these assumptions were not met, non-parametric tests were performed. Means of sound characteristics were calculated for each fish and Atezolizumab each behavioural context, and used for further analyses. Differences between sexes

in dominant frequency, pulse duration, pulse period and pulse repetition rate (when applicable), and SPL of sounds produced, were calculated using t-test. Relationships between seahorse height and sound characteristics were determined using Pearson’s (log-transformed growls’ pulse period values) and Spearman rank (for feeding clicks’ SPL values) correlation coefficients. In order to calculate differences in sound characteristics

recorded in different behavioural contexts (feeding, stress, courtship), a Kruskal–Wallis test was performed followed by a Dunn’s post hoc test. The difference in the number of sounds emitted during courtship was compared using the Mann–Whitney U-test (between males and females) and the Friedman test (among courtship days), followed by a Dunn’s post hoc test. Hippocampus reidi produced two distinct sounds (Table 1) in different 上海皓元 behavioural contexts: click sounds – single pulses, recorded during feeding and courtship, frequently audible to the observer during trials; and growling sounds – a series of sound pulses emitted only when handheld and never during intraspecific interactions. Feeding clicks were produced during prey capture and consisted of short broadband sounds that were typically uttered singly (mean duration: 16.1 ms), with the main energy ranging from 50 to 800 Hz. The mean SPL (LLFP re: 2 cm) of feeding clicks was 119.8 dB re 1 μPa (see Table 1). They were produced in all feeding events recorded (Fig. 1a). Click duration [r = −0.