An experimental study was made of influence of strain magnitude and strain duration on dynamic fracture in uniaxial tension of low-strength gypsum plaster, high-strength gypsum plaster, high early strength portland cement mortar, and ordinary portland cement mortar. Dynamic test specimens were circular cylindrical bars with diameters ranging from 0.9 to 1.2 inches and lengths ranging from 18 to 58 inches. Static test specimens 2 inches long were cut from the long bars. A special loading device, designed and constructed by Melpar, generated a compressive pulse by longitudinal impact of two metal bars and applied the pulse to one end of the dynamic test specimens. The compressive pulse was reflected at the free end of specimens as a tensile pulse and caused fracture in tension at a section near the middle. Time from zero strain to maximum tensile strain varied from 10 to 35 microseconds, and total duration of tensile strain varied from 20 to 430 microseconds with few exceptions. All materials withstood tensile strains two to three times the static fracture strain for short periods. The straining time required to cause fracture varied with strain magnitude.