Holarctic Marmots as a factor of Biodiversity.
Rumiantsev V.Yu;, Nikol'skii A.A. & Brandler O.V. eds.,
Abstracts, 3d Conference on Marmots (Cheboksary, Russia, 25-30 August 1997),
Moscow ABF 1997, 216p., 100 (Russian), 141-142 (English)


SOME PHONETIC PARAMETERS OF THE ALARM CALLS IN MARMOTA MARMOTA L.: DIFFERENCES BETWEEN THE FIRST AND THE SECONDARY WHISTLES RELATED TO THE SILHOUETTES


C. Facello*, E. Macchi*, M. Panzera**, R. Piazza*, E. Tesio*

* Dip. Morfofisiologia Veterinaria, Universite Torino
** Istituto Fisiologia Veterinaria, Universite Messina, Italy


A preliminary evaluation of the experiences with artificial aerial stimuli shows that different models elicit a different number of acoustic answers by Marmots. Now it is useful to verify if and how phonetic parameters change with the changing of artificial danger situation. Firstly data were considered into two big groups: those concerning the first whistle (FW), at least the first vocalization could be recorded in every run of the shape, and those concerning the following calls. Starting from this, data were compared using Fisher test showing interesting results. FW elicited by predators models (Golden Eagle, Goshawk and Buzzard) has a first harmonic of lower mean frequency (Fme) than following ones (P<0.05); moreover, considering the only occasional predators (Goshawk and Buzzard), FW= shows lower Fme (P<0,01) and lower difference of amplitude (a.d.) between second and first harmonic (P<0.01) (Fig. 1). Testing FW elicited by Eagle's model versus FW of occasional predator=92s ones, they are different considering= both Fme (higher for Eagle with P<0.05) and a.d. (higher for Eagle with P<0.01).

The same differences don't subsist considering the following whistles. May be that these differences depend on the dimension of models, in fact it's immediatly evident that the Marmot's predator birds are bigger than non predator ones (Kestrel and Chough). To test this hypothesis three new models were built; they were of simple rectangular shape, reflecting respectively the dimensions of Golden Eagle (2x0.53 m); occasional predators (0.9x0.35 m) and non predators (0.6x0.3 m). Surprising their use elicited a low number of calls (the average not being dissimilar to that of non predators models).

Spectral analysis shows that these whistles are quite different from those recorded with other models; Fme is highest (P<0.01) and also a.d. (P<0.01) in fact second harmonic is frequently absent (fig. 1). Geometrical models seem to represent a kind of stimulus quite different from others: there are no differences changing rectangular dimension nor it is possible to reproduce the effects due to other models of similar dimension. Marmots show more concern for predators' shapes than for non predators' ones (difference as number of whistles). In the case of predator raptors models, first call seems correlated with danger kind or degree, while following vocalizations are used to manteining high attention and alert in the colony until danger is ended. The model's silhouette greatly influences acoustic antipredatory alert behaviour and surely more than model's dimension. Although models and real birds represent two different perceptive realities, it's most probable that differences seen testing the animals' reactions to artificial shapes are just due to the simple aspects (as dimension or silhouette) that these stimuli show.

Fig. 1. Left sonogram (band width 43 Hz) shows a whistle with a high amplitude of second harmonic (most typical of the first call). On the right a call whose second harmonic is almost absent (most typical of secondly whistles and of calls related to geometrical shapes).



Back to Abstracts contents