Communication sounds or ‘social calls’ of 16 European bat species (Chiroptera, Vespertilionidae) were recorded at a range of roost and foraging sites. A comparative analysis of more than 5400 individual calls for general structures and for inter- as well as intraspecific variability resulted in 50 types of calls, which differed by their specific structure and by the calling species. These types could be grouped into four different general types of calls, according to the kind and complexity of their structure, independent of the calling species. Distinct types of calls seem to have similar functions in different bat species. One general type may be used predominantly in female–infant interactions as an isolation or direction call, which serves as mutual recognition. This type of social call was also used in ‘tandem flights’ of pairs of bats, which might increase individual knowledge of roost sites and foraging success. A second type was used in mate attraction, and a further one in an aggressive context. The fourth one was used by hindered or distressed bats. The group of ‘aggressive’ calls is least variable, but the complex mating calls and isolation calls are very diverse. Species-specific sound structures were identified, which allowed a computational species distinction. The measured inter-individual variability of social calls should be significant for their functions in individual recognition. So, beyond common features concerning the frequency structure of bat social calls, interspecific differences, as well as the intraspecific variability of details of sonagraphic parameters, should elucidate the specific functions of the calls.

The aim of this study was to describe the structure of social calls produced by Kuhl's pipistrelles Pipistrellus kuhlii. Bats foraging around street lamps in the Campania region, Southern Italy, were recorded. Calls were produced by bats during chases and, similarly to those of Pipistrellus pipistrellus, were probably used to repel conspecifics from hunting sites. Calls often had three components, lasted on average 34 ms and contained most energy at about 17 kHz. A positive correlation was found between the frequencies of maximum amplitude of echolocation and social calls. Social calls from two adjacent populations differed in their peak frequencies, and possible hypotheses for this variation are given. The structure of P. kuhlii social calls was compared with those of the 45 and 55 kHz P. pipistrellus phonic types. Kuhl's pipistrelle calls lasted longer and showed lower values of minimum and peak frequencies. These differences can help to discriminate between field recordings of P. kuhlii and P. pipistrellus where the two species occur sympatrically.