Data on the past intensity of Earth's magnetic field (paleointensity) are essential for understanding Earth's deep interior, climatic modeling, and geochronology applications, among other items. Here we demonstrate the possibility that much of available paleointensity data could be biased by instability of thermoremanent magnetization (TRM) associated with non-single-domain (SD) particles. Paleointensity data are derived from experiments in which an ancient TRM, acquired in an unknown field, is replaced by a laboratorycontrolled TRM. This procedure is built on the assumption that the process of ancient TRM acquisition is entirely reproducible in the laboratory. Here we show experimental results violating this assumption in a manner not expected from standard theory. We show that the demagnetization−remagnetization relationship of non-SD specimens that were kept in a controlled field for only 2 y show a small but systematic bias relative to sister specimens that were given a fresh TRM. This effect, likely caused by irreversible changes in micromagnetic structures, leads to a bias in paleointensity estimates.paleomagnetism | paleointensity | thermoremanent magnetization | multidomain T he aim of paleointensity research is to reconstruct variations in the absolute intensity of the ancient geomagnetic field throughout Earth history. Paleointensity data are essential for constraining the conditions in the core (1-4), for studying the role that the geomagnetic field plays in controlling Earth's atmosphere (5-7), and as a geochronological tool (8, 9). Despite the necessity for a large amount of reliable paleointensity data, there are still significant ambiguities in the available paleointensity information (10-12); these call for a reevaluation of the existing database (13).One fundamental problem in paleointensity research arises from the difficulty in locating dateable ancient materials that fulfill the most basic requirement of the absolute paleointensity method. This requirement states that the natural remanent magnetization (NRM) should be a pure thermoremanent magnetization (TRM) carried exclusively by noninteracting singledomain (SD) particles (14). As purely SD materials are rare in nature, much of the published data are based on materials that do not entirely fulfill the strict assumption of pure SD, but still demonstrate a reasonably satisfying relationship between the blocking and unblocking temperatures. That is to say, pseudo SD (PSD) or even small multidomains (MD) are frequently assumed to carry stable and reproducible magnetizations.We first outline the principles of the absolute paleointensity method (14, 15), as most of the data in the paleointensity database rely on some variant of this method (10, 16). The basic underlying assumptions of any absolute paleointensity method are that TRM is quasi-linearly proportional to the intensity of the field (B) in which it was acquired (TRM = α · Β), and that in the absence of chemical and physical alteration, the proportion between the TRM and B is an intrinsic...