We tested the hypothesis that clustering in the behaviour of emerging bats is a response to the risk of avian predation. We hypothesised that if avian predation was the cause of clustering, bats in the prolonged absence of avian predators, would not cluster or would cluster less during their emergences. We studied the Azorean bat (Nyctalus azoreum) in the Azores Archipelago. The Azores have a depauperate fauna with no raptorial birds likely to predate bats. The Azorean bat is an endemic mammal to the archipelago, which has an unusually extensive degree of diurnal activity that has been hypothesised to reflect release from the risk of diurnal predation by raptors. Contrary to our prediction Azorean bats clustered during emergence to the same extent as bat species which occur where there are raptors. Two interpretations of these data are possible. First, the hypothesis that the behaviour is anti-predatory may be incorrect. Most of the variation in clustering was explained by variation in ambient temperature possibly suggesting the bats emerged in groups to aid exploitation of sparsely distributed food. Alternatively, the behaviour may be anti-predatory, but the key factor precipitating clustering may not be the risk from aerial predators, but terrestrial predators, such as rats (Rattus norvegicus) and cats (Felis cattus), both of which were common around the roost sites.