Water superheated to the limit in an open tube develops aninternalnegative pressurePn(tending to tear the molecules apart) equal to the saturation vapor at that temperature, less one atmosphere. Kendrick, Gilbert, and Wisner heated water in an open thin‐walled capillary U‐tube to 270°C for 5 sec before it exploded (Pn=53 atmos). Using their method, the writer heated three tubes to 264°, 266°, and 267°C for 5 sec or more before explosion occurred (Pn=48 to 51 atmos). The cohesive strength of water is thus sufficient to withstand aninternalnegative pressure of over 51 atmos at 267°C. At this temperature an additional negative pressure (appliedexternallythrough centrifugal force) would rupture the water column. It has been predicted from van der Waals' equation that theexternalnegative pressure which the system could withstand would vanish at 273°C, in fair agreement with experiment.