In general, never show the representation of a thing unless it be impossible to show the thing itself; for the sign absorbs the child's attention, and makes him lose sight of the thing signified. The armillary sphere[2] seems to me poorly designed and in bad proportion. Its confused circles and odd figures, giving it the look of a conjurer's apparatus, are enough to frighten a child. The earth is too small; the circles are too many and too large. Some of them, the colures,[3] for instance, are entirely useless. Every circle is larger than the earth. The pasteboard gives them an appearance of solidity which creates the mistaken impression that they are circular masses which really exist. When you tell the child that these are imaginary circles, he understands neither what he sees nor what you mean. Shall we never learn to put ourselves in the child's place? We do not enter into his thoughts, but suppose them exactly like our own. Constantly following our own method of reasoning, we cram his mind not only with a concatenation of truths, but also with extravagant notions and errors. In the study of the sciences it is an open question whether we ought to use synthesis or analysis. It is not always necessary to choose either. In the same process of investigation we can sometimes both resolve and compound, and while the child thinks he is only analyzing, we can direct him by the methods teachers usually employ. By thus using both we make each prove the other. Starting at the same moment from two opposite points and never imagining that one road connects them, he will be agreeably surprised to find that what he supposed to be two paths finally meet as one. I would, for example, take geography at these two extremes, and add to the study of the earth's motions the measurement of its parts, beginning with our own dwelling-place. While the child, studying the sphere, is transported into the heavens, bring him back to the measurement of the earth, and first show him his own home. The two starting-points in his geography shall be the town in which he lives, and his father's house in the country. Afterward shall come the places lying between these two; then the neighboring rivers; lastly, the aspect of the sun, and the manner of finding out where the east is. This last is the point of union. Let him make himself a map of all these details; a very simple map, including at first only two objects, then by degrees the others, as he learns their distance and position. You see now what an advantage we have gained beforehand, by making his eyes serve him instead of a compass.