Over the last decades, determination of ages in monazite using Electron Probe Microanalysis (EPMA) has progressed to the point that it is now considered as a routine procedure within several laboratories worldwide [1,2]. Nevertheless, the typical age errors of individual electron probe spots reported in the literature are an order of magnitude larger than attainable by Secondary Ion Mass Spectrometry (SIMS) and two orders of magnitude larger than attainable by Thermal Ionization Mass Spectrometry (TIMS). The age errors frequently range from 40 Ma to 120 Ma, although after statistical treatment of a set of homogeneous ages, a precision of approximately 20 Ma can be obtained. Despite the low accuracy, the numerous studies have shown that the Th-U-totalPb method using the EPMA is a reliable method, easy to use, for detecting age variations on a very small scale. The effective limitation of the EPMA monazite ages determination concerns its relatively poor accuracy regarding mainly Pb determination. Similarly the relative uncertainties for U and Th are generally larger than 2%, which can lead to large errors. The error on the age for each analysed point depends on the U, Th and Pb concentration. As the Pb concentration is directly related to the age of the monazite and to the U and Th concentration, older is the monazite, better is the relative error. Since the first monazite dating study by EPMA [1] there has been improvements in data age processing, but only few evaluations of uncertainties in the Pb, U, Th determinations. As the error calculated on the age for each analysed point is the whole inaccuracies on Pb, U, Th determinations, this method requires high accuracies determinations of trace elements [3] as well as of major elements.The aim of the present experimental and theoretical study is firstly for evaluating precisely the uncertainties of EPMA dating of monazite and secondarily to give suggestions for minimizing them. Precise evaluation is not straightforward because sources of uncertainties include not only counting statistic but also many parameters like microprobe operating conditions (e.g., accelerating voltage, beam current, beam diameter, pulse height analyser), spectral background subtraction, quantification procedure, overlap, line shape, standards, sample preparation (polish, carbon coat, contamination), irradiation damage etc.. To estimate experimentally these sources, measurements were done on a large homogeneous monazite from the Itambe pegmatite district (Brazil [4]). The crystal homogeneous at the micrometer scale is free of fluid and solid inclusions. The composition (6.5 wt. % Th, 0.12 wt. % U and 0.15 wt. % Pb) is typical for natural monazites. Multiple Isotope Dilution Thermal Ionisation Mass Spectrometer (IDTIMS) analyses yield concordant U-Pb ages of 474 ± 1 Ma [5] and document the isotopic homogeneity of the monazite. The precise knowledge of age and the low Pb and U content in this monazite allows us to estimate the main sources of uncertainties in age determination using EPMA. As th...