Dallmann R. & Geissmann, T. (2001). Different levels of variability in the female song of wild silvery gibbons (Hylobates moloch). Behaviour 138: 629-648.

Different levels of variability in the female song of wild silvery gibbons (Hylobates moloch)

Robert Dallmann & Thomas Geissmann

Institute of Zoology, Tierärztliche Hochschule Hannover, Germany

Key Words: Hylobates moloch; silvery gibbon; song; individuality; vocalisation.

Abstract: Gibbon songs are known to include species- and sex-specific characteristics. It has been suggested frequently that these songs also exhibit a high degree of individuality, but quantifying individuality has rarely been attempted. Because the statistical methods used in earlier studies were highly dependent on sample size (Kruskal-Wallis one-way analysis of variance by ranks), it was not possible to compare results among studies directly. We introduce a mean pairwise difference (MPD) of scaled variables in order to quantify great-call variability and individuality. Because of its construction as simple normalised difference, the MPD is largely independent of sample size. This makes it possible to compare results directly with those of other studies on other populations or species. Even various levels of variability (intra- vs. inter-individual, intra- vs. inter-population variability) can be determined and compared with this method. In addition, the MPD can be calculated independently for any acoustic variable. This opens up a broad variety of research options in the area of comparative analysis of acoustic communication. For instance, variability in various parts of a phrase, in various parts of a song bout, in various contexts or in various seasons can be compared, and this is possible even if the variables under comparison are not the same. As an example we analysed female great-call phrases of wild silvery gibbons in Java (Indonesia). We found that inter-individual variability is significantly higher than intra-individual variability. This implies that females can be distinguished by their great-calls, although this was not examined during the present study. Additionally, variability of female songs was found to be significantly lower within one population than among any two populations. The various sections of the great-call differ in their variability. The first half of the great-call (excluding the introductory note) shows the highest potential for individual recognition.