This is the first report of the study carried out during the last 4 years. The objective of this research is to clarify the acquiring process of the number concept in children by the “method of experimental education.” Here, its problems, principles and methods are discussed. In the orthodox study of the developments of the children's number concept, the central problem seems to be the discovery of the hierarchy of the developmental stages of this, concept, as are typical in Piaget's or Werner's researches. In their theories, we think, it is assumed implicitly that the acquisition of number concept is initiated by internal factors such as maturation or readiness. There is no doubt that Piaget's study is one of the most brilliant works in this field. Internal factors, however, should de the most important determinant and the role of learning should have little significance, if the establishment of the order of developmental stages is the most important point concerning the number concept or mathematical education. But we think that in acquiring the number concept the symbolic mediation process is sine qua non, which has cultural origin. So, the external factors, learning, experience or education should have equal importance. Does the term “maturation” mean after all the sudden appearance of new behavior pattern? It is extremely difficult to tell whether such behavior pattern is caused by maturation or by experience. Using the traditional method of simple observation of the developmental phenomena, it might not be possible to determine the significance or the role of maturation in causal relations. This methodological ref lexion leads us to the point that the method of exprimental control of the developmental processes has to be applied in order to get basic solution of the role of maturation or experience. Such method has already been used by Gesell and Luria successfully. We try to extend the same type of method into a more general methodology and give a name of “the method of experimental education.” It has analogical significance with the method of experimental embryology in biology. That is to say, the method of experimental education applies some operations on the apparently natural process of the number development, and tries to find out, by theconsequent responses, what causal factors are working on it. In the area of the developmental study of number concept up to date, the main theories have been to discover the natural sequence of developmental stages, and paid little attention to the role of learning or experience. However, the number has highly systematic characters, and can only be acquired according to cerfain order. Therefore, it is necessary to attempt to give the number concept to children according to certain order. We consider that the essential character of number concept is not the abstraction of the numerical between numerical operations such as like plus and minus, and the operations of relationships between concrete objects. The numerical operations, however, have highly abstract characteristics, while, the operation of objects is possible by the concrete way of thinking. There is so much discrepancy between these two dimensions. Mediating them, we use here the dice-patterns as the semi-concrete and semi-abstract dimension. By the above principle, the model of learning process is discussed and the actually performed schedule of the research is presented.