Several studies have shown that an alteration chemical remanent magnetization (CRM) may have a direction intermediate between that of the remanence of the parent phase and the field during remagnetization. In a fold a surviving natural remanent magnetization (NRM) is exposed at varying angles to the remagnetizing field, giving rise to varying degree of remanence deflection. A laboratory experiment was carried out to investigate how the CRM deflection varies with angle between parent phase NRM and remagnetizing field, HCRM, and we discuss how this variation would affect the results of a fold test. The experiment was performed on a Silurian lava, with a well-defined single component NRM carried by deuterically oxidized titanomagnetite (TM). Specimens of the lava were arranged in a fold configuration with NRM at various angles to a 50 ¿T laboratory field. Heating to 525¿C resulted in oxidation of part of the primary TM phase to hematite, leading to the formation of stable remanence components of intermediate direction, as revealed by thermal demagnetization. Stepwise unfolding of the CRM gave a syntectonic fold test result. The variation of remanence deflection with angle between HCRM and initial NRM can be understood in terms of vector addition of fields or remanence components. A model for CRM acquisition derived from the experimental results was employed to explore CRM in other fold configurations. It appears from analysis of synthetic data that it is possible to falsely conclude both synfold and prefold magnetization from fold tests on postfold or synfold CRM.