Compression-creep data for zircaloy 2, Mo-Cu-Zr alloy and 18-8 stainless steel were obtained at room temperature, 250°C, 316°C(600°F) and 450°C for a period of 100 hours. Zircaloy 2 and Mo-Cu-Zr alloy were casted respectively as ingot by the consumable-electrode double-arc melting. The test specimens were machined from a bar obtained from the ingot by forging, and annealed at 700°C (Zircaloy 2) and 750°C (Mo-Cu-Zr alloy) for 1 hour in vacuum furnace.<br>The test equipment for compression creep is the conventional tension creep machine with a fixture consisted of two yokes which convert tensile loading into compressive loading. The fixture used is of the similar type to the one developed at the Westinghouse Research Laboratory by M.J. Manjoine. The compression-specimen which has a diameter of 12mm and an overall length of 36mm was compressed between two seats, the ends of the specimen and of the seats being ground and lapped. The relative displacement of the yokes was measured by dial gauge extensometer as a measure of the strain in the specimen.<br>For checking the magnitude of instantaneous strain in creep tests, short-time tension and compression tests were made for zircaloy 2 and Mo-Cu-Zr alloy at 316°C by using the test equipment above-mentioned. The continuous loading was given by moving a running weight sliding on the loading lever arm of the creep machine.<br>Although at room temperature zircaloy 2 and Mo-Cu-Zr alloy have smaller instantaneous and creep strain in comparison with 18-8 stainless steel which displays appreciable creep at room temperature, they tend to have poorer creep resistance at higher temperatures, and the steady-state creep component becomes conspicuous for Mo-Cu-Zr alloy at 316°C and for zircaloy 2 at 450°C. The creep strength of Mo-Cu-Zr alloy at 450°C is stronger than that of zircaloy 2 when the stress level is below 17kg/mm².<br>Comparison of tension-creep and compression-creep properties for Mo-Cu-Zr alloy (at 316°C and 450°C) show that the alloy has poorer resistance in compression than in tension within a certain limit of stress, above which an effect of decrease of stress resulted from the increase of cross-section of a compression specimen would appear. The similar phenomena for S 816 and nimonic 90 at 1600°F have been reported by L.A. Yerkovich.<br>This difference in creep-resistance may partly be explained by the anomalous variation of the stress-strain relationship in tension and compression. But it should be taken into account as well that the bedding-down of the ends of compression-specimen and the anisotropic effects in the resistance to deformation produced in the process of preparing the test specimen are related to the difference in creep-resistance, although in our experiment the bedding-down of the compression specimen was minimized by lapping the ends of compression specimen.