native child Olympus is as remote as heaven and as unreal. To a second as described in "Snowbound,' it is the local Huckleberry Hill magnified and peopled with strange beings. To a third, whose imagination has not been stirred, it is merely a word. Manifestly, the second state is the only one that leads to geography.

As to astronomical geography, the child's concepts are even more vague. He has almost nothing to go on. He sees the heavenly bodies, it is true, but he cannot imagine the earth one of them. Such grotesque pictures as appear from time to time in the comic journals are more likely to convey to him some clear ideas of the earth as a heavenly body than do all the geography books. However, in childhood it is of little consequence whether the earth is a globe or a platter, so long as the sun and the moon and the stars rise in the east and set in the west, and season succeeds season, though of course it is well to tell children that the earth is round and let them grow into the knowledge.

To recapitulate, the average child beginning to use a primary geography, if previously untaught, has a meager knowledge of the people he has seen, their characteristics and employments. He also knows the topography of his home town more or less completely and of such places as he may have visited. That is his equipment for the study of geography, except an unlimited imagination which tends to make a supernatural product of everything partially known, of everything described in words outside his experience. This statement covers fairly well the range of knowledge, hence of apperceiving centers, of the average child not specially taught.

Acquired Concepts. But children in the average modern school when taking a textbook in geography are not in "the natural state." They have been given a great number of elementary geographic concepts in school, through oral instruction and through supplementary readers. It is important to consider this possible enlargement of the apperceiving possibilities. What does it commonly give? What should it give? How does it affect the first use of the textbook?

We must assume that instruction in any subject, at any time, is for the training of the minds and the enrichment of the lives of the children at the time, and not merely to prepare them for the use of a textbook later.

The Imagination. The first three years of school, before a textbook is used in geography, cover the period of the most vivid imagination of the whole school life. They are the days when fairy tales and nature myths appeal most strongly, when the mind can see in the unknown likeness to the known, but when differences readily run into grotesque images. Little boys and little girls and their fathers and mothers and brothers and sisters and pet animals are the same the world over, but unfamiliar creatures, unknown powers, readily become hideous monsters or kindly, absurd beasts.

It is the childhood of the race over again. Gods and goddesses, fairies, nymphs, ogres, dragons, dwarfs, giants, valkyrs, angels, represent flights of the imagination attempting to compass that which has no prototype or parallel in experience.

So the early geographies made the unknown parts of the world totally unlike the known parts. The imagination, unchecked by knowledge, ran riot in two-headed

giants, one-footed men, serpents coiled about the earth, turtles and fish as its supports, the heavens above, the earth beneath, and the waters under the earth. Scylla and Charybdis, Niflheim and Valhalla, Olympus and Hades, were mere attempts at geography, unguided by a sane apperception. The voyages of Ulysses, the travels of Sir John Mandeville, the wanderings of Sindbad, all belong to the same class; and children, given geographic terms beyond their apperceptive powers, but appealing to their imaginations, are likely to make just such journeyings. Truly, however, this is better than the deadly instruction whose words are drugs stupefying even the imagination.

Important Concepts. -What concepts of the world should teachers seek to implant during the first three years of school? Naturally, such as grow out of the most vivid knowledge that the children have. That is, people, the family, the playmates, the neighbors; the home environment, food, bed, fireside, play, work; animals and plants, all the concerns of child life. Which of these can be extended most readily into the world beyond the immediate ken? That may vary with the circumstances. But I should say, first, child life in its most material aspects, food, clothing, games, parental care, cold, heat, school, also the employments of others within the daily observation. How can this be done? I must repeat, all geographic knowledge except that obtained by direct observation comes through the imagination. Hence, facts taught to children, to be of value, must be extensive, that is, they must prepare the minds for the production of correct images of the remote and unseen when described in words. It is the

failure to consider this that has often reduced to absurdity so-called "home geography," making it merely a statement of the obvious. The bare fact, like a smooth round pebble, has little educational worth. A fact must be covered with hooks to give it value.

Hence, the geographic ideas to be given during these primary school years are to be estimated by their possible relations, the strength and tenacity of their tentacles, rather than by their manifest substance. They must produce the sort of intelligence that prepares the way for further intelligence. This is rather vague, I admit. It does not tell what to teach. And, indeed, it is impossible to dogmatize about that.

Influence of Environment. So much depends upon what the children of any school are likely to have seen and heard before coming to school. The first lessons for a class in Louisiana should be different in many respects from those for a class in Maine, because the 66 known" differs. To a child in southern California grass is an exotic raised on lawns with great difficulty. To a child in Minnesota bananas and oranges occur only on fruit stands and at the grocer's.

The first steps from the known cannot be very long, especially because of the psychological fact above noted, that children readily see likenesses but can make little educational use of differences. Almost the only differences to be noted with profit at first are those of degree, the adding to the known, or substracting from it some other known. A child never having seen a cow cannot imagine one except as like a dog, or a mouse, or some other known animal. (See "Contents of a Child's Mind," Stanley Hall.) But the same child has no difficulty whatever

with a two-headed giant or an animal with the power of speech, or a bodiless head. Children in a temperate zone can understand Agoonack, because they can imagine a very long winter; but to children in southern Florida, or in Cuba, Agoonack's home belongs with the land at the top of Jack's beanstalk. To Anna Josefsky, of Baxter Street, New York, a rolling prairie is a wheel on a cart or perhaps a barrel rolling along the sidewalk, or possibly a roller skate. And there is no way of correcting these images at the time. All attempts to give knowledge of conditions widely different from those familiar should be abandoned with young children. Contrasts may be used, but only contrasts both of whose elements are known. Differences in climate as to heat and cold may be explained because children know both. Contrasts used must be of degree or quality of known elements.

The first steps should be to likenesses with slightly differing aspects, and these should be made easy by graphic and constructive illustrations. For instance, the dress and food of children in other lands can easily be made clear by illustration. Games, if known, are good beginnings. Homes and the modes of sleeping and eating are not difficult to teach and are very suggestive. When the children can read quite readily, such books as "Seven Little Sisters" are very helpful. As Dr. Taylor clearly shows in his valuable articles in "Educational Foundations," stories with geographic settings are among the best means of imparting concepts.

The Study of Physical Features. The question of studying physical features is sure to arise. The answer must depend very largely upon the locality.

Little can