Fringed myotis (Myotis thysanodes) is a bat species of conservation concern in western North America that may be impacted by increased recreational activity near roost sites, changes in water resource availability caused by increased urban and agricultural water use, and anthropogenic climate change. Our purpose was to describe and model maternity roost use by fringed myotis in Colorado. We compared differences between roosts occupied by maternal fringed myotis and randomly selected potential roosting locations that were not known to be occupied by this species during the maternity period. We evaluated the strength of evidence for competing hypotheses on 2 scales: one that included landscape variables and a second that included roost-site variables. We used logistic regression, Akaike's information criterion, and multimodel inference to investigate maternity roost use by fringed myotis. The model explaining the most variability in our landscape data included grade and aspect, and the model explaining the most variability in our roost-site data was estimated volume of the roost. Understanding maternity roost use by fringed myotis can guide conservation and management decisions related to roost protection in the Rocky Mountain West. When feasible, we believe that developing knowledge about maternity roost use, as well as autumn and winter roost use, will help improve management decision making related to forest bats of conservation concern in western North America.