A new study of bed load transport—the movement of the gravel or other grains on a stream bed—has turned up a previously undetected effect. Marquis and Roy used several different methods to monitor bed load activity in a gravel bed river, Beard Creek in Quebec, Canada. They examined streamfow, bed load, and bed morphology before, during, and after 20 food events. The researchers found that two of the methods—measuring changes in bed topography between successive foods and surveying bed activity—gave inconsistent results. Changes in elevation of the bed did not always correspond to movement of bed load.