This study aimed at investigating the developmental study of topographical localization in school children. The purpose of Experiment I was to examine the effect of perceptual and spatial cues on topographical localization in landscape task. For this purpose, three task were varied as to the degree of cues. Each task required 216 subjects (1st, 2nd, 3rd, 4th, grades) to observe both position and gaze direction of one doll on unrotated landscape, and then to place the other doll on rotated landscape in the same position as the doll on unrotated landscape, facing the same direction. The results were: (1), the cues had effect on the acceleration of localization.(2) the degree of acceleration was different in the degrees of cues, age grades, the points of landscape and the kind of localization-(a) position in the correspondent quadrant formed by the intersection of a road and a stream (b) gaze direction of a doll (c) precise position (d) simultaneous consideration on gaze direction and precise position. These results suggested: that by using remarkable objects as cues, one reference system of one's own body as a reference point to define the position of objects was easily constructed, and the change of reference directions between rotated and unrotated landscape was easily grasped, and the reference system of one's body on unrotated landscape was reconstructed on rotated landscape. The purpose of Experiment II was to investigate the difference of transformation of relation of topographical position and gaze direction in the landscape task among cognitive formation levels of grasping the change of relation of objects. The Ss, who could not transform the relation of positions, were 60 school children (6:9-8:7). The Ss were divided into three groups, and trained in accordance with the method of experimental education. Group C: trained for one day. The Ss defined the position of objects by front-back, right-left relation to one's own body and other person's body at any points. Group E₂: trained for two days. After the same training as group C, the Ss could transform the relation of positions by concrete acts, in relation to one's own body as measure. Group E₁: trained for five days. After the same training as Group E₂, the Ss could discover the transformational rule, and used it to transfer the relation of positions. The results were; group E₁ was better than group E₂, group E₂ was better than group C in landscape task. Conclusion,(1) it became clear that important variable was as follows: to grasp the change of reference directions by concrete acts, and to reconstruct the reference system of one's body as a reference point to define the position of objects. But these concrete acts had little effect on the acceleration of acquisition of transformation of topographical position and gaze direction.(2) After the concrete acts, using the transfofmational rule had much effect not only on the acceleration of acquisition of transformation of relation of topographical position and gaze direction but also on its retention.