The flow of sap in Populus tremuloides was studied by the thermoelectric method immediately prior to and at several periods of time after radical girdling. Similarly, the flow of sap in Betula papyrifera was studied before and after girdling or injection with metabolic inhibitors or a filtrate of a culture of Ceratocystis ulmi.None of the treatments caused any significant change in the speed at which the sap flowed. However, the rate at which the temperature rose at the upper (downstream) thermocouple decreased significantly in girdled trees and in most trees treated with metabolic inhibitors. That decrease was first noticed 4 days after poisoning with 0.1 m mercuric chloride, and 1 day after poisoning with either 0.01 m picric acid or a culture filtrate of Ceratocystis ulmi. A 0.05 m trichloroacetic acid solution caused a decrease in the rate of temperature rise about 2 hours after the tree was injected. The rate of temperature rise returned to normal the next day after the injection with trichloroacetic but again showed a decrease after 42 days.Increases in the rate of heat rise were noted the day after the birch trees were girdled, and all girdled trees showed a significant decrease in the rate of heat rise sometime after they were girdled. Many of the girdled trees showed no sap flow or even an apparent downward flow of xylem sap sometime after girdling.Laboratory experiments indicated that the blocking of the vessels by air emboli would result in a significant change in the rate of temperature rise, not accompanied by a significant change in the speed at which the sap flowed. Since air emboli caused similar results to those actually experienced in the field, the data are interpreted as the result of blocked or nonfunctional vessels.All of the data agree with that of earlier workers. Cessation of upward flow occurs only in girdled trees after rather long periods of time. It is suggested that cessation could be caused by tyloses and air emboli. All of the data are also in agreement with the passive theories of sap ascent. However, the trichloroacetic acid experiments suggest the existence of a repair mechanism or a second means of sap movement.