Marks (2001) is critical of our study of nest-site selection in Great Horned Owls (Bubo virginianus) because we compared our data from owl nests with data from random points rather than from unused stick nests. We argue that Great Horned Owls have so many options for nesting in eastern forests that there is little constraint on nest-site selection. Therefore, their choice of nest sites is determined largely by vegetation characteristics in the surrounding landscape, and comparison of owl nests with random points is the best way to assess nest-site selection. Furthermore, we believe that use of unused nests as controls, as advocated by Marks, is biased because control nests will have many of the same characteristics as nests used by owls.