Keith, S. A. (2005).Vocal diversity of female Kloss's gibbons (Hylobates klossii) in the Mentawai islands, Indonesia. M.Sc. thesis in Primate Conservation, Oxford Brookes University, U.K., x + 85 pp.
Sally A. Keith
The vocal diversity of female Kloss's gibbons was
assessed by spectrographically analysing the sex-specific, stereotypic great call
phrase of the female. Songs were recorded during a period from June - August 2005
in the Mentawai Islands 85-135 km off the coast of West Sumatra, Indonesia. Four
locations on three different islands in the chain were used: Simabuggai (Siberut),
Sikabei (Siberut), Saureinu (Sipora) and South Pagai (South Pagai). Songs were recorded
from three to six listening posts per location using a modified point census technique.
A total of 24 females and 137 great calls were of sufficient quality to be spectrographically
analysed from the four populations.
Spectrographic analysis was achieved using 'Raven 1.2' software and data was obtained for 12 variables: (1) Pre-trill + trill duration, (2) Frequency band exploited during pre-trill and trill, (3) Duration of 1st note, (4) Frequency modulation of 1st note, (5) Duration of 2nd note, (6) Frequency of 2nd note, (7) Number of pre-trill notes, (8) Pre-trill duration, (9) Notes/second in pre-trill, (10) Trill duration, (11) Number of trill notes, (12) Notes/second in trill. No variables were included from the post-trill element of the great call phrase because of insufficient recording quality. Statistical tests determined significance of diversity in the following categories: Intra-individual, inter-individual, inter-population, intra-island, inter-island, and Siberut group and Sipora & South Pagai group. Coefficient of variation, Mann Whitney U and Kruskal Wallis non-parametric analysis of variance tests demonstrated significant diversity of individual, population and island categories. Discriminant function analysis assigned great calls to individuals correctly in a range of 47.8-81.1% of cases in the 4 populations, assigned individuals to populations correctly in 58.3% of cases, and assigned individuals to their correct islands in 70.8% of cases. Discriminant function analysis for classification of individuals to either a Siberut group or Sipora & South Pagai group, assigned individuals correctly in 62.5% of cases when all variables were included, and in 79.2% of cases when the variable ë Number of notes in trill' was removed.
Results suggest individuals are distinguishable from each other and therefore there is potential for individual recognition among conspecifics. Three hypotheses are proposed to explain inter-individual diversity: (1) Trilling as a handicap, (2) Neighbour-stranger discrimination, (3) Relaxed selection. Inter-individual diversity differs between populations and this is suggested to be because of local differences in forest structure and characteristics. Inter-population differences are hypothesised to be due to rivers acting as facilitators to speciation. Intra-island (Siberut) diversity is lower than inter-island diversity as expected. Inter-island diversity is significant but implies an invalid clinal relationship throughout the island chain. The level of diversity between the Siberut group and the Sipora & South Pagai group indicates the Kloss's gibbon should remain with no Evolutionarily Significant Units (ESUs) despite a trend suggesting a process of divergence, and advocate a reassessment of the current classifications of the three endemic Mentawai monkey species.
Site by Thomas Geissmann.
For comments & suggestions, please email to