Land managers and scientists often ask: “How do I identify the threshold?” or “Where is the threshold?” Such questions presume a great deal about the nature of thresholds that are, as yet, not generally understood and not necessarily general at all. In this talk, I review how the threshold concept has been considered in land management and argue five points. 1) Attempting to predict threshold responses is likely to be a waste of time in many cases. 2) We have not yet produced an adequate number of case histories about the consequences, scales, rates, and mechanisms of threshold behavior within different kinds of ecosystems, so it is difficult to structure monitoring programs to detect threshold responses. 3) Managing for resilient conditions, with attention to slow social and ecological variables, to minimize the risk of crossing a threshold is potentially useful but is controversial for managers. 4) We have a dim understanding of how inherent (i.e., not management-related) variations in landscapes influence threshold responses, but such understanding can be used to guide planning. 5) Recognizing the consequences of post-threshold states for restoration actions is the most straightforward use of the threshold concept at present. Examples are drawn from restoration approaches in the southwestern United States and elsewhere to support these assertions.