Aims Because rhinolophids have been hypothesized to use echolocation call frequency to recognize conspecifics, sympatric species calling at similar frequencies should be subject to acoustic character displacement, i.e. a drift in frequency values to minimize the risk of misidentification of conspecifics. However, it has been proposed that insufficient geographical separation between populations in sympatry and allopatry may counter the establishment of frequency differences by character displacement. Here we tested the hypothesis that insular populations should exhibit acoustic divergence, and this should be revealed by comparing call frequencies with those observed in mainland, allopatric populations of conspecifics. We also tested whether the evolutionary pressure towards acoustic divergence should be especially strong at sites where rhinolophid species emitting similar call frequencies roost together in order to minimize interspecific frequency overlap.
Location Sardinia and southern Italy (Campania, Lazio, Abruzzo).
Methods Time-expanded echolocation calls and body size were recorded from Sardinian populations of Rhinolophus mehelyi Matschie, Rhinolophus hipposideros (Bechstein) and Rhinolophus euryale Blasius. Both call frequencies and forearm length of insular R. hipposideros and R. euryale were compared with those of populations from mainland areas of Italy where R. mehelyi is absent, to explore the hypothesis that the presence of the latter species (which calls at frequency values intermediate between the other two) may determine acoustic divergence in the other species. For Sardinian R. mehelyi and R. euryale, we also carried out intraspecific comparisons of call frequencies between bats from monospecific colonies and those from mixed colonies.
Results As hypothesized, Sardinian R. hipposideros and R. euryale called at frequencies higher and lower, respectively, than in the peninsula. In this way, overlap with R. mehelyi is avoided. Body size showed no difference between insular and peninsular populations, i.e. frequency differences are not a by-product of difference in body size determined by insularity. Frequency values in Sardinian R. euryale from monospecific colonies did not differ from those of bats roosting together with R. mehelyi. However, R. mehelyi showed frequency values significantly higher when associated with R. euryale, possibly to minimize the risk of species misrecognition.
Main conclusions At least under geographical isolation, character displacement may be a causal mechanism for shifts in call frequency of sympatric rhinolophids. Species recognition and facilitation of intraspecific communication (with possible implications for mate recognition) constitute the best candidate factors for the phenomenon we observed.