ИСТИНА

Although killer whale (Orcinus orca) dialects have been studied in detail in several populations, little attempt has been made to compare dialect characteristics between populations. In this study we investigated diversity of monophonic and biphonic calls among four resident populations from the North Pacific Ocean: Southern residents (SR, about 86 animals), Northern residents (NR, about 216 animals), southern Alaska residents (AR, >751 animals), and eastern Kamchatka residents (KR, >650 animals). Call contours were extracted using an algorithm in MATLAB and compared using dynamic time warping. We calculated the similarity between each pair of calls from each major category (monophonic/biphonic) within each population. SR had the highest median similarity (that is, the lowest call diversity) for monophonic calls followed by NR, AR and KR. KR had the highest median similarity for biphonic calls followed by AR, NR and SR. So, the diversity of monophonic and biphonic calls appears to be negatively correlated and related to the population size: monophonic calls are more diverse and biphonic calls are less diverse in larger populations. In human languages, speaker population size was shown to be a significant predictor of phonemic diversity, with a smaller population size predicting smaller overall phoneme inventories. The diversity of monophonic calls follow the same pattern, which suggests that their evolution is driven by the same stochastic processes that also affect human phonemes. By contrast, diversity in biphonic calls shows the opposite pattern, which suggests that the relative effect of directional selection is greater for biphonic calls. Therefore, we suppose that dialect evolution is a complex process influenced by an interaction between evolutionary forces.