Certain species group themselves into natural associations, that is to say, into communities which we meet with more or less frequently and which exhibit the same combination of growth-forms and the same facies. As examples in northern Europe may be cited a meadow with its grasses and perennial herbs, or a beech forest with its beech trees and all the species usually accompanying these. Species that form a community must either practice the same economy, making approximately the same demands on its environment (as regards nourishment, light, moisture, and so forth), or one species present must be dependent for its existence upon another species, sometimes to such an extent that the latter provides it with what is necessary or even best suited to it (Oxalis Acetosella and saprophytes which profit from the shade of the beech and from its humus soil); a kind of symbiosis seems to prevail between such species. In fact, one often finds, as in beech forests, that the plants growing under the shade and protection of other species, and belonging to the most diverse families, assume growth-forms that are very similar to one another, but essentially different from those of the forest trees, which, in their turn, often agree with one another. The ecological analysis of a plant-community leads to the recognition of the growth-forms composing it as its ultimate units. From what has just been said in regard to growth-forms it follows that species of very diverse physiognomy can very easily occur together in the same natural community. But beyond this, as already indicated, species differing widely, not only in physiognomy but also in their whole economy, may be associated. We may therefore expect to find both great variety of form and complexity of interrelations among the species composing a natural community; as an example we may cite the richest of all types of communities--the tropical rain-forest. It may also be noted that the physiognomy of a community is not necessarily the same at all times of the year, the distinction sometimes being caused by a rotation of species. The different communities, it need hardly be stated, are scarcely ever sharply marked off from one another. Just as soil, moisture, and other external conditions are connected by the most gradual transitions, so likewise are the plant-communities, especially in cultivated lands. In addition, the same species often occur in several widely different communities; for example, Linnaea borealis grows not only in coniferous forests, but also in birch woods, and even high above the tree limit on the mountains of Norway and on the fell-fields of Greenland. It appears that different combinations of external factors can replace one another and bring into existence approximately the same community, or at least can satisfy equally well one and the same species, and that, for instance, a moist climate often completely replaces the forest shade of dry climates.