In this research, the map-use ability of primary school students of grade 1-4 has been discussed, that based on the theories of children's cognitive development. The studying items have included four dimensions: coordinate location, spatial orientation, measurement, and symbolization. That has formed the basis of study-ability criteria on map-use in primary school, so as to design the sampling test of map-use ability. The number of 1,162 sampling subjects were chosen from over Taiwan area, that were composed of three parts: urban, rural, and aboriginal. The relationships between the four-dimensions and the topics of regional, sexual, study-grade were discussed. The survey analysis has shown that: 1) in the lower-grade (1~2) student's map-use ability, the average score is 75(%), the coordinate location (87%) and measurement (94%) dimensions have higher score, and the spatial orientation (68%) and symbolization (63%) score worse. In the regional scope, there were no obvious difference between the urban and the rural student, but are much better that the aboriginal. In the sexual-part, there were still no obvious differences between the body and girls students, except in the 1st grate that the boy's measurement dimension is better. In the grade-part, the 2nd grade students are obvious better than 1st grade in all dimensions of the urban and aboriginal parts, but the superior arts are different in each other. In the rural part, there is no difference between 1st and 2nd grade student. 2) in the median-grade (3~4) part, the average score of the student's map-use ability is 82(%). The coordinate location (88%) and symbolization (85%) have higher score, the spatial orientation score secondary (68%), and measurement dimensions score worst (67%). In the regional scope, the urban and rural parts are obviously better than the aboriginal in all dimensions, but they have no difference between each other. In the sexual-part, the median-grade girl students have some better map-use ability in coordinate location, spatial orientation and measurement dimensions than the boys, but still have no obviously difference in these four-dimensions. In the grade-part, 4th grade students had the better score than 3rd grade in the four dimensions of the urban, rural, and the aboriginal. This has accommodates the theories of children's cognitive development.