A direct method for determining soil drought resistance to supplement field tests is described. Grass seedlings are grown for 6 to 8 weeks in the greenhouse and then given a soil drought treatment lasting from 6 to 9 days by placing them in a chamber designed to maintain constant environmental conditions of temperature (80 degrees F), light (175 foot-candles), relative humidity (30 to 35%), and air velocity (0.5 mile per hour). Seedlings surviving the drought treatment are allowed to recover under favorable growing conditions in the greenhouse. The percentage of plants renewing growth during the recovery period is used as the index of relative drought resistance of a species or strain. In preliminary trials the reaction of over 96,000 grass seedlings to the controlled soil drought treatment has been studied. The average survival values for species have been in agreement with their known behavior under natural conditions of critical drought. More important, however, was the wide variation observed between strains or seed sources within species. Although field data are available for only a small number of the strains tested, satisfactory correlation between field and greenhouse survival was obtained where comparisons were possible. The variability of strains within species suggests the possibility of breeding for soil drought resistance in grasses and the dangers of characterizing a species as to drought resistance from observations on a single strain or seed source.