During the past decade, research on large in-stream wood has expanded beyond North America's Pacific Northwest to diverse environments and has shifted toward increasingly holistic perspectives that incorporate processes of wood recruitment, retention, and loss at scales from channel segments to entire watersheds. Syntheses of this rapidly expanding literature can be facilitated by agreement on primary variables and methods of measurement. In this paper we address these issues by listing the variables that we consider fundamental to studies of in-stream wood, discussing the sources of variability in their measurement, and suggesting more consistency in future studies. We recommend 23 variables for all studies of in-stream wood, as well as another 12 variables that we suggest for studies with more specific objectives. Each of these variables relates either to the size and characteristics of in-stream wood, to the geomorphic features of the channel and valley, or to the ecological characteristics of the riparian zone adjacent to the study reach. The variables were derived from an overview of those cited in the literature and from our collective field experiences.