Two experiments were conducted to test Bryant(1982)'s hypothesis that what is critical to a concept development is not "conflict" as Piaget emphasized but "agreement" between strategies.In Experiment 1, spontaneous measurement was examined with 20 5-and 6-year-old children following Bryant's method in a simpler form. One group proved that the results of their "measurements" using rods agreed with the results of "direct side-by-side comparisons", while another group proved that the results of their "measurements" conflicted with the results of "direct comparisons from different levels". The result was that children who experienced agreements made more spontaneous measurements than those experiencing conflicts. In Experiment 2, number onservation was examined with 40 5-and 6-year-old children. One group proved that the results of their "countings" agreed with the results of "one-to-one correspondences" using sticks, while another group proved that the results of their "countings" conflicted with the results of "direct visual comparisons". The result was that children who experienced agreements conserved numbers of 4 to 10 better than those experiencing conflicts.