This International Standard describes a method for the determination of the resistance of vulcanized rubbers to fatigue under repeated tensile deformations, the test piece size and frequency of cycling being such that there is little or no temperature rise. Under these conditions, failure results from the growth of a crack that ultimately severs the test piece. The method is restricted to repeated deformations in which the test piece is relaxed to zero strain for part of each cycle. Analogous fatigue processes can occur under repeated deformations which do not pass through zero strain and also, in certain rubbers, under static deformation, but this International Standard does not apply to these conditions. The method is believed to be suitable for rubbers that have reasonably stable stress-strain properties, at least after a period of cycling, and that do not show undue stress softening or set, or highly viscous behaviour. Materials that do not meet these criteria might present considerable difficulties from the points of view of both experiment and interpretation. For example, for a rubber that develops a large amount of set during the fatigue test, the test strain will be ill-defined and the fatigue life is likely to differ markedly under constant maximum load and constant maximum extension conditions; how the results for such a rubber should be interpreted, or compared with those for other rubbers, has not been established by basic work. As a general guide, a rubber for which the set determined in accordance with 9.5 and 10.2 exceeds 10 % is likely to fall into this category. For this reason, the method is not considered suitable for most thermoplastic elastomers. Similar considerations apply with regard to other changes in elasticity behaviour during testing. A distinction should be made between this fatigue test and the flexometer tests described in the various parts of ISO 4666, where fatigue breakdown occurs under the simultaneous action of stress and temperature. Advantages over the De Mattia flex cracking and cut growth test (see ISO 132) include the following. The test yields quantitative results which do not depend on operator interpretation and which can be recorded automatically. The initial deformation is clearly defined and can readily be varied to suit different applications. Great caution is necessary in attempting to relate standard test results to service performance since the comparative fatigue resistance of different vulcanizates can vary according to the test conditions used and to the basis by which the results are compared. Guidance on the selection of test conditions and on the interpretation of results is given in Annex A.