Open Access

Brian Bertram

• The Indian Hill Mynah (Gracula religiosa) was studied in the field in Assam in north-east India. The aims of the study were two-fold: (i) to understand better this bird's exceptional ability in captivity to imitate human speech; and (ii) to provide background understanding to studies of the importance of early auditory experience and of vocal imitation in the development of normal song patterns in birds. First is given a brief description of the distribution, general behaviour, and breeding biology of this arboreal, sexually isomorphic, semi-gregarious species. The remainder of the monograph deals with vocalizations; these were either tape-recorded in the field, or transcribed directly using a written notation developed for the purpose. Any wild adult Mynah of either sex possesses four categories of vocalizations: (i) 'Chip-call'; a loud piercing squeak made in contexts which include alarm. (ii) 'Um-sound'; a soft grunt, acting in close range social contexts, and (like chip-calls) common to all individuals. (iii) 'Whisper-whistles': several soft sounds of types unique to the individual. (iv) 'Calls': several loud noises, of extremely varied patterns. The bulk of the monograph deals with 'calls', as defined thus. Calls were compared quantitatively with one another by a method developed which measured the degree of overlap of one sonogram with a tracing of a second sonogram. Both by this method and by ear, calls were divided into discrete types, without intermediates. Birds of either sex have a repertoire of usually between five and twelve such call types, some of which are produced much more commonly than others. Repertoires tend to be larger in birds which call more frequently, or which have mates with large repertoires. The repertoire of a given bird stays largely constant from year to year in size. composition, and proportions. No bird shares any of its call types with its mate, but it shares several of them with near neighbours of the same sex. There is a progressive change of dialect with distance, such that birds nesting more than about 14 km apart have no call types in common. No general characteristics of call structure could be found which were indicative of the sex of the caller, but in a known locality the call type made immediately reveals the sex of the bird producing it. Call types are learnt by selective imitation of neighbouring individuals during a young bird's first several months. A call type common in the repertoire of one bird tends also to be common in the repertoire of a neighbour, except at the edge of the limited range of that call type. Which particular call type a calling bird selects from among those in its repertoire is discussed. Few call types could be related to non-auditory contexts. A bird is likely to repeat the call type last made, and also tends to standardize the order in which it produces its different call types; this standard order tends to be the same as that of its neighbours. A birdtends also to reply at once and to standardize the call type it makes in immediate reply to a particular call type of its mate; again, neighbouring pairs of birds tend to use the same standardized call and reply types. The length of the interval between a particular call and its reply tends to be constant in a given pair of birds, and approximately the same in neighbouring pairs. These are all further aspects of extensive but selective vocal imitation by Mynahs of adult birds; other species are not imitated. Information on calling when in contact with other pairs came mainly from playback experiments, when single calls were presented to nesting pairs of Mynahs. Response strength was measured by the incidence of flight, number of subsequent vocalizations, latency of response, and proportion of playbacks ignored. When presented with playbacks of calls of familiar types (of neighbours) and of unknown types (of strangers), birds responded more strongly to the familiar than to the unknown call types. They did, however, respond somewhat to the unknown call types, which were of patterns never previously heard by them, presumably recognizing these as being Mynah calls by their sound quality. Mynahs responded as strongly to playbacks of neighbours' calls which were not in their own repertoire as to playbacks of neighbours' calls which were. A bird tends to match at once the call last heard (either from a tape recorder or from a wild neighbour), itself producing the same call type at once, if it possesses it in its own repertoire. That call type, and others associated with it, also occurs more frequently thereafter. Thus calls heard affect calls made, and vice-versa since other individuals nearby behave similarly. A change of nearest neighbours in successive years was shown to affect one pair's repertoire proportions. Further playback experiments showed that Mynahs were able to distinguish between a single call made by their neighbours and a single call of the same call type made by their mates. Small but consistent differences were found in the sonograms of such calls of the same type made by different birds. The structure of a single call type may change gradually with distance. The development of vocalizations with age is briefly described. In the final discussion sections, the ways in which, and the extent to which, Mynahs are able to determine the species, home locality, sex and individual identity of other Mynahs are outlined. There follow consideration, and comparison with other species: (i) of various aspects of repertoires; (ii) of the distribution of call types among different individuals; (iii) of the dynamic aspects of calling, and a scheme is proposed which accounts for the selection for utterance of a particular call type from the repertoire; and (iv) of the organization and coordination of calling. The lack of imitation of other species in the wild is discussed, and contrasted with the several ways in which wild Mynahs imitate one another in various aspects of their calling.