Objective: The aim of this research was to study in primary teeth the relationship between the accuracy of the electronic apex locator Root ZX and each of the following factors: root canal curvature, canal size, the position of the apical foramen and the degree of resorption. Materials and methods: The pulp tissue was removed from 71 root canals of 21 primary molars. After irrigation with NaOCl (1%), the working length of 71 root canals was measured by Root ZX (Morita, Tokyo, Japan) clinically prior to extraction. The actual length of each canal was recorded after extraction, and compared with the electronically measured length. Results: Paired samples t test was performed at a significance level of 95% (P ≤ 0.05); electronically measured values were lower than those of the actual length, with a tendency to estimate the root canal length just short of the apex. Both Cronbach's alpha and intraclass correlation coefficient values were near to 1. Chi-square analysis showed that P > 0.05 for canal curvature, file size and resorption degree, but P 0.05 for the position of the apical foramen. Mann-Whitney U test P values were 0.05 when comparing between the "≥ 2.0 mm before the apex" subgroup and other subgroups. Conclusions: The accuracy of Root ZX is acceptable in primary teeth, and is not affected by root canal curvature, canal size or the degree of resorption. However, when the apical foramen is ≥ 2.0 mm short of the anatomical apex, the possibility of incorrect measurement increases. Palabras clave: apex locator, canal curvature, primary teeth, resorption, root canal instruments, Root ZX, working length