The migratory tree-roosting hoary bat (Lasiurus cinereus) and silver-haired bat (Lasionycteris noctivagans) are among the bat species with the highest reported prevalence of rabies in North America. However, bats submitted for rabies testing typically have been those that have come in contact with humans or pets. Given the roosting ecology of L. cinereus and L. noctivagans, contact with healthy individuals of these species is expected to be rare, with a bias in contact and submission of infected individuals and thus an overestimation of rabies prevalence. We tested 121 L. cinereus and 96 L. noctivagans specimens, collected during mortality surveys at wind energy facilities in Southern Alberta, Canada in 2007 and 2008, for rabies. None of the L. cinereus (0%) and one L. noctivagans (1%) tested positive for rabies. Prevalence of rabies was significantly lower than previously reported estimates, passive and active, for L. cinereus and L. noctivagans. In a review of the literature including multiple bat species, we found a significant difference in estimates of rabies prevalence based on passive versus active surveillance testing. Furthermore, roosting ecology influenced estimates of rabies prevalence, with significantly higher prevalence among passive surveillance submissions of nonsynanthropic species compared to synanthropic species, a trend not evident in active surveillance reports. We conclude that rabies prevalence in randomly collected L. cinereus and L. noctivagans is low and comparable to active surveillance estimates from other species (≤ 1%), and that roosting ecology influences estimates of rabies prevalence among bats submitted to public health laboratories in North America.