Alteration experiments on natural monazite crystals (Manangotry standard, Madagascar) under alkali conditions at 300, 400, 500 and 600 °C and 200 MPa were conducted to clarify mechanisms behind incomplete resetting of U-Th-Pb geochronological systems in monazite replaced by dissolution and precipitation processes. Above 400 °C, experimental products show typical replacement textures: a compositionally distinct monazite rim, referred as altered rim, surrounds the primary monazite (Mnz1). Isotopic and electron microprobe U-Th-Pb in situ dating of the altered rim yields intermediate ages between pristine monazite (555 Ma) and complete experimental resetting (0 Ma). Lead is systematically detected in altered rims, with concentration decreasing from 400 °C to 600 °C. The origin of incomplete resetting is elucidated with transmission electron microscope images that reveal an incomplete replacement of Mnz1 by a secondary monazite (Mnz2) within the altered rim. With increasing temperature, the size and volume of the Mnz2 within the altered rim become more important. Because no structural Pb or Pb nanoinclusions were observed, Pb in the altered rim is attributed to the Mnz1 component. Partial resetting of U-Th-Pb systems depends on the nanomixture of different Mnz1 proportions in the analyzed volume, and explains the higher rejuvenation at 600 °C than at lower temperatures. Although microanalytical techniques have the spatial resolution to date micrometer-sized rims, they are unable to resolve a nanoscale mixture of pristine and secondary monazite that could occur in altered rims formed by fluid-driven replacement, especially at low temperatures. Porosity and/or inclusions and complex age scattering in zoned monazite are significant markers that can indicate a possible nano-sized partial replacement.