Temperatures in southwestern North America are projected to increase 3.5-4 degrees C over the next 60-90 years. This will precipitate ecological shifts as the ranges of species change in response to new climates. During this shift, rapid-colonizing species should increase, whereas slow-colonizing species will at first decrease, but eventually become reestablished in their new range. This successional process has been estimated to require from 100 to over 300 years in small areas, under a stable climate, with a nearby seed source. How much longer will it require on a continental scale, under a changing climate, without a nearby seed source? I considered this question through an examination of the response of fossil plant assemblages from the Grand Canyon, Arizona, to the most recent rapid warming of similar magnitude that occurred at the start of the Holocene, 11,700 years ago. At that time, temperatures in southwestern North America increased about 4 degrees C over less than a century. Grand Canyon plant species responded at different rates to this warming climate. Early-successional species rapidly increased, whereas late-successional species decreased. This shift persisted throughout the next 2700 years. I found two earlier, less-extreme species shifts following rapid warming events around 14,700 and 16,800 years ago. Late-successional species predominated only after 4000 years or more of relatively stable temperature. These results suggest the potential magnitude, duration, and nature of future ecological changes and have implications for conservation plans, especially those incorporating equilibrium assumptions or reconstituting past conditions. When these concepts are extended to include the most rapid early-successional colonizers, they imply that the recent increases in invasive exotics may be only the most noticeable part of a new resurgence of early-successional vegetation. Additionally, my results challenge the reliability of models of future vegetation and carbon balance that project conditions on the basis of assumptions of equilibrium within only a century.